MX 201010 en

User's Guide
IMEX
Advanced Oil/Gas Reservoir Simulator
Version 2010
By Computer Modelling Group Ltd.
This publication and the application described in it are furnished under license
exclusively to the licensee, for internal use only, and are subject to a confidentiality
agreement. They may be used only in accordance with the terms and conditions of
that agreement.
All rights reserved. No part of this publication may be reproduced or transmitted in any
form or by any means, electronic, mechanical, or otherwise, including photocopying,
recording, or by any information storage/retrieval system, to any party other than the
licensee, without the written permission of Computer Modelling Group.
The information in this publication is believed to be accurate in all respects.
However, Computer Modelling Group makes no warranty as to accuracy or
suitability, and does not assume responsibility for any consequences resulting from
the use thereof. The information contained herein is subject to change without notice.
Copyright 1987-2010 Computer Modelling Group Ltd.

All rights reserved.
The license management portion of this program is based on:

Reprise License Manager (RLM) Copyright (C) 2006-2010, Reprise Software, Inc
IMEX uses Intel(R) Compilers.
IMEX, CMG, and Computer Modelling Group are registered trademarks of
Computer Modelling Group Ltd. All other trademarks are the property of their
respective owners.
Computer Modelling Group Ltd.
Office #150, 3553 - 31 Street N.W.
Calgary, Alberta Canada T2L 2K7
Tel: (403) 531-1300
Fax: (403) 289-8502
E-mail: cmgl@cmgl.ca
Preface
IMEX is CMG's new generation adaptive implicit-explicit black-oil simulator which
includes features such as local grid refinement, comprehensive well management,
pseudo-miscible option, volatile oil option, polymer flooding, horizontal wells, dual
porosity/permeability, flexible grids, and many more. IMEX was developed to simulate
primary depletion, coning, water, gas, solvent, and polymer injection in single and double
porosity reservoirs.
This User's Guide provides a step-by-step procedure for preparation of an input data
set for this program. A tutorial section is provided as well as a set of appendices
describing the underlying theory. Use of this User's Guide requires a basic
knowledge of reservoir engineering and some exposure to reservoir simulation.
Every attempt has been made in the preparation of this User's Guide to provide the
user with all the necessary details. If questions arise, please contact:
Computer Modelling Group Ltd.
#150, 3553 31 Street N.W.
Calgary, Canada
T2L 2K7
Telephone: (403) 531-1300 Fax: (403) 289-8502 E-mail: cmgl@cmgl.ca
Confidentiality: All components of CMG technology including software and related

documentation are protected by copyright, trademark and secrecy. CMG technology
can be used only as permitted by your license from CMG. By the license, you have
agreed to keep all CMG technology confidential and not disclose it to any third party.
All rights reserved. No part of this publication may be reproduced or transmitted in any
form or by any means, electronic, mechanical, or otherwise, including photocopying,
recording, or by any information storage/retrieval system, to any party other than the
licensee, without the written permission of Computer Modelling Group.
Corrections/Errors: CMG ENDEAVORS TO PRODUCE TECHNOLOGY OF THE
HIGHEST QUALITY; NEVERTHELESS ERRORS OR DEFICIENCIES IN SUCH
TECHNOLOGY ARE INEVITABLE. IF YOU FIND AN ERROR OR DEFICIENCY, YOU
ARE REQUESTED TO PROVIDE DETAILS OF IT AND ILLUSTRATIVE DATA SET(S)
TO CMG SUFFICIENT TO PERMIT CMG TO REPRODUCE THE ERROR OR
DEFICIENCY. CMG SHALL ENDEAVOR TO REMEDY A DEFICIENCY IN A TIMELY
MANNER AND SHALL PERIODICALLY REPORT TO YOU AS TO THE STEPS BEING
TAKEN TO REMEDY THE DEFICIENCY. THE RESPONSE TIME FOR A DEFICIENCY
MUST BE PRIORITIZED FOR THEIR GENERAL APPLICATION TO CMG MEMBERS
AND WHETHER THEY FORM PART OF A CMG PROGRAM. CMG DOES NOT
WARRANT THAT DEFICIENCIES WILL BE REMEDIED.
Limited Liability: CMG does not warrant the accuracy or usefulness of the technology
and software - Refer to your license.
Contents
Introduction
Important Changes between IMEX 2010.10 and IMEX 2009.10.................................1

Important Changes between IMEX 2009.10 and IMEX 2008.10.................................4
Important Changes Between IMEX 2008.10 and IMEX 2007.10 ................................6
Important Changes Between IMEX 2007.10 and IMEX 2006.10 ................................8
Important Changes Between IMEX 2006.10 and IMEX 2005.10 ..............................12
Important Changes Between IMEX 1999.10 and IMEX 98.05 ..................................37
Important Changes Between IMEX 98.05 and IMEX 98.00 ......................................41
Important Changes Between IMEX 93.00 and 92.00 .................................................61
Introduction to IMEX .................................................................................................65
Tutorial
69
Introduction.................................................................................................................69
Data Groups in the Keyword Input System ................................................................70
How to Document Your Data Set ...............................................................................71
How to Do a Restart....................................................................................................72
Controlling Contents of the Output File .....................................................................74
Controlling Contents of the Graphics File (SR2)........................................................75
Describing Your Grid System.....................................................................................76
Describing Refined Grid .............................................................................................78
Using Dual Porosity/Dual Permeability......................................................................79
Input of Null Blocks ...................................................................................................83
Using the Aquifer Option (see Appendix G) ..............................................................85
Using the Pseudo-Miscible Option .............................................................................87
Using the API Tracking Option ..................................................................................88
User's Guide IMEX
Contents i
Using the Rock Compaction/Dilation Model Option ................................................. 89

Using the Polymer Option .......................................................................................... 98
Problems with Small Timesteps or Long Execution Times ....................................... 99
Problems with Solver Convergence ......................................................................... 102
Simulating a Gas-Water or Gas Reservoir ............................................................... 103
Simulating a Gas-Water with Condensate Reservoir ............................................... 104
Simulating a Tilted Water-Oil/Water-Gas Contact .................................................. 105
Horizontal Wells....................................................................................................... 109
Vertical Equilibrium Calculation ............................................................................. 110
Defining Multiple PVT Regions .............................................................................. 114
Defining Wells ......................................................................................................... 117
Defining the Well Type............................................................................................ 119
How to Shut In a Well and Reopen It....................................................................... 120
Voidage Replacement............................................................................................... 122
Using the Wellbore Model ....................................................................................... 123
Operating and Monitoring Constraints..................................................................... 125
Definition of Well Indices within IMEX.................................................................. 127
Input of Well Indices................................................................................................ 129
Stopping a Simulation Run....................................................................................... 131
Using Wildcards in Well Lists ................................................................................. 132
Guidelines for Setting Up Well Data........................................................................ 133
Recurrent Data from Other Sections ........................................................................ 135
Subsidence Output.................................................................................................... 137
Parallel IMEX........................................................................................................... 139
Using the Seawater Injection Option........................................................................ 141
Using the Scale Buildup Option ............................................................................... 142
Simulating a Volatile Oil Reservoir ......................................................................... 144
Using the Subdomain-Dual Permeability-Reinfiltration Model............................... 145
Modelling Non Darcy Flow in Hydraulic Fractures Accurately Using a Grid
Based Approach .................................................................................................. 147
Using the Fracture Clean-up Model ......................................................................... 153
Using the Oil Wet Option......................................................................................... 157
Special Keywords for Advanced Users (SHIFT) ..................................................... 165
Modelling Non Darcy Flow in Fractured Shale Gas Reservoirs .............................. 166
The WATER_FIX Keyword .................................................................................... 179
Introducing a Hydraulic Fracture in Recurrent Data ................................................ 180
The Trapped Oil Hysteresis Option for Pcow and Krow (Oil-Water System) ............. 181
Keyword Data Entry System
189
Introduction to Keyword System.............................................................................. 189

Comments (Optional) ............................................................................................... 195
Blank Lines (Optional)............................................................................................. 196
Scan Mode for Checking Errors ............................................................................... 197
Include Files (Optional)............................................................................................ 198
ii Contents
User's Guide IMEX
Controlling Data File Listing (Optional) ..................................................................199

Controlling Data File Listing Using NOLISTLIM (Optional) ................................200
Changing the Comment Indicator (Optional) ...........................................................201
Changing the Keywords by Using Translate Rules (Optional).................................202
Input of Grid Property Arrays...................................................................................203
Entering Matrix Grid Properties ...............................................................................205
Entering Fracture Grid Properties.............................................................................206
Entering Refined Grid Properties..............................................................................207
J and K Direction Data from I Direction ..................................................................208
Constant Value Arrays..............................................................................................209
Array Input in IJK Notation......................................................................................210
Array Input of Values that Vary in the I Direction ...................................................212
Array Input of Values that Vary in the J Direction...................................................213
Array Input of Values that Vary in the K Direction .................................................214
Values that Vary for Most or All Grid Blocks..........................................................215
Modifying Array Data (Conditional) ........................................................................216
Interpolating Table Data (Optional)..........................................................................219
Input/Output Control
221
Command Line Input/Output Control.......................................................................221

Values Stored in Binary Form ..................................................................................224
ASCII Characters ......................................................................................................226
Input/Output File Names (Optional).........................................................................227
Important Notes (Run Time Dimensioning) .............................................................231
Problem Type Where IMEX Fails to Allocate Adequate / Appropriate Storage ........232
Run Time Dimensioning (Optional) .........................................................................233
Background (Run Time Dimensioning)....................................................................238
Advantages of Run Time Dimensioned IMEX.........................................................239
Project Main Title (Optional)....................................................................................240
Project Second Title (Optional) ................................................................................241
Project Third Title (Optional) ...................................................................................242
Case Identification (Optional)...................................................................................243
Check Only (Optional)..............................................................................................244
Trap UNIX Signal 2 Interrupt (Optional) .................................................................245
ASCII Form of SR2 (Optional).................................................................................247
Precision of SR2 (Optional)......................................................................................248
Controlling Data File Listing Using NOLISTLIM (Optional) .................................249
Input Data Units (Optional) ......................................................................................250
Output Data Units (Optional)....................................................................................253
Data Range Checking (Optional)..............................................................................254
Maximum Number of Error Messages (Optional)....................................................255
Restart Timestep (Optional)......................................................................................256
Restart Record Writing Frequency (Optional)..........................................................257
Output Printing Frequency (Optional) ......................................................................259
User's Guide IMEX
Contents iii
Items in Output Print File (Optional) ....................................................................... 262

Simulation Results File Writing Frequency (Optional)............................................ 269
Items in Simulation Results File (Optional)............................................................. 271
Items in Diary File (Optional) .................................................................................. 277
Production Split Output (Optional) .......................................................................... 279
Debug Output (Optional).......................................................................................... 280
Reservoir Description
283
Fundamental Grid Definition (Required) ................................................................. 283

K Direction Index (Optional) ................................................................................... 290
Block Dimensions for the I Direction (Conditional) ................................................ 292
Block Dimensions for the J Direction (Conditional)................................................ 294
Block Dimensions for the K Direction (Conditional) .............................................. 296
Depth to the Centre of a Grid Block (Conditional) .................................................. 298
Depth to the Tops of Grid Blocks (Conditional) ...................................................... 300
Depths to Centre of Pay (Conditional) ..................................................................... 302
Depths to Top of Block (Conditional)...................................................................... 304
Grid Tilt Angles (Conditional) ................................................................................. 306
Corner Point Depths for Corner Point Grids (Conditional)...................................... 308
Lateral Corner Point Locations for Corner Point Grids (Conditional) ..................... 310
Line-Based Corner Point Locations for Corner Point Grids (Conditional).............. 312
Complete Corner Point Locations for Corner Point Grids (Conditional)................. 314
Local Refined Grid (Optional) ................................................................................. 317
Refined Grid Location (Conditional) ....................................................................... 327
Dual Porosity (Optional) .......................................................................................... 329
Dual Permeability (Optional) ................................................................................... 330
Dual Porosity Subdomain Method (Optional).......................................................... 331
Assigning Fractional Volumes to Subdomain blocks (Optional)............................. 333
Subdomain to Subdomain Flow Transmissibility Multipliers (Optional) ................ 335
Dual Porosity MINC Method (Optional) ................................................................. 336
Shape Factor Calculation (Conditional) ................................................................... 337
Matrix-Fracture Transfer Calculation (Conditional) ................................................ 339
Allowing Addition Matrix-Fracture Connections (Optional)................................... 341
Fracture Spacing (Conditional) ................................................................................ 342
Null Block Indicator (Optional) ............................................................................... 343
Porosity (Required) .................................................................................................. 345
Rock Compressibility (Required)............................................................................. 346
Pore Volume Modifiers (Optional) .......................................................................... 348
Permeabilities (Required)......................................................................................... 350
Netpay (Optional)..................................................................................................... 352
Netgross (Optional) .................................................................................................. 354
Transmissibility Multipliers (Optional).................................................................... 355
Transmissibility Multipliers for Lower Indexed Block Faces (Optional) ................ 358
Pinch Out Array (Optional)........................................................................................... 360
iv Contents
User's Guide IMEX
Pore Volume Cut-Off Threshold (Optional).............................................................362

Pinchout Tolerance (Optional)..................................................................................363
Corner Point Tolerance (Optional)................................................................................365
Faults (Optional) .......................................................................................................366
Aquifers (Optional) see Appendix G ........................................................................368
Pressure Influence Function (Conditional) see Appendix G ....................................375
Sectors (Optional) .....................................................................................................377
Sector Array (Optional) ............................................................................................379
Sector Assignment via Names and an Array (Optional)...........................................380
Flux Sector (Optional) ..............................................................................................381
Lease Planes (Optional) ............................................................................................383
Special Connections (Optional) ................................................................................385
Fault Array (Optional) ..............................................................................................387
Compaction/Dilation Rock Type (Optional) ............................................................389
Compaction/Dilation Rock Compressibility (Optional) ...........................................391
Compaction Rock Table (Optional)..........................................................................392
Compaction Hysteresis Rock Table (Optional) ........................................................394
Treat *CCPOR / *CROCKTAB / *CROCKTABH as Depletion Parameters/
Tables (Optional) .................................................................................................397
Compaction Irreversibility Flag (Optional) ..............................................................399
Dilation Rock Table in Elastic Zone (Optional) .......................................................400
Dilation Rock Table in the Plastic Zone (Optional) .................................................402
Dilation Rock Table in Unloading Zone (Optional) .................................................404
Dilation Rock Table in Recompacting Zone (Optional)...........................................407
Dilation Rock Table in Reloading Zone (Optional)..................................................409
Compaction/Dilation Rock Region (Optional) .........................................................417
Dispersion Coefficients (Conditional) ......................................................................423
Fault Transmissibilities (Optional) ...........................................................................425
Component Properties
427
Fluid Model (Required) ............................................................................................427

Reservoir Temperature (Optional)............................................................................431
Oil and Gas PVT Table (Optional) ...........................................................................432
Oil and Gas Differential Liberation Table (Optional) ..............................................437
Gas PVT Table for API Model (Optional) ...............................................................444
Oil PVT Tables for API Model (Optional) ...............................................................447
Gas PVT Table for GASWATER Model (Optional)................................................452
Condensate - Oil and Gas PVT Table (Optional) .....................................................455
Volatile - Oil and Gas PVT Table (Optional) ...........................................................461
Oil Compressibility as a Function of Pressure (Optional) ........................................467
Bo Above the Bubble Point as a Function of Pressure (Optional)............................470
Oil Viscosity Above the Bubble Point as a Function of Pressure (Optional) ...........473
Condensate Model - Undersaturated Eg, Bg, Zg Table ............................................476
Condensate Model - Undersaturated Gas Viscosity Table .......................................480
User's Guide IMEX
Contents v
Reference Pressure for Gas-Oil Capillary Pressure (Optional) ................................ 484

Solvent PVT Table (Conditional)............................................................................. 485
Densities (Required)................................................................................................. 489
Under-Saturated Oil Compressibility (Optional) ..................................................... 492
Oil Phase Viscosity Pressure Dependence (Optional).............................................. 493
Water Formation Volume Factor (Required) ........................................................... 494
Water Phase Viscosity (Required)............................................................................ 496
PVT Type (Optional)................................................................................................ 498
Maximum Rate of Increase of Solution Gas Ratio or Bubble Point Pressure
(Optional) ............................................................................................................ 499
Initial Gas-Oil Ratio (Conditional)........................................................................... 501
Gas and Solvent Mixing Parameter (Conditional) ................................................... 502
Minimum Solvent Saturation (Conditional)............................................................. 503
Polymer Adsorption Table (Conditional)................................................................. 505
Polymer Viscosity Mixing (Conditional) ................................................................. 507
Reference Polymer Viscosity (Conditional)............................................................. 510
Reference Polymer Concentration (Conditional) ..................................................... 511
Polymer Permeability Table (Conditional) .............................................................. 512
Reference Seawater Viscosity (Conditional)............................................................ 513
Rock-Fluid Data
515
Notes on Rock-Fluid Data........................................................................................ 515

Start of Rock-Fluid Property Input (Required) ........................................................ 517
Relative Permeability Curves by Table (Required).................................................. 518
Water-Oil Relative Permeability Table (Conditional).............................................. 520
Liquid-Gas Relative Permeability Table (Conditional)............................................ 531
Hysteresis Parameters (Optional) ............................................................................. 542
Reduced Vertical Extent of Pcow Cycle for Trapped Oil Hysteresis (Conditional)....... 547
Method for Evaluating 3-Phase Kro (Optional) ....................................................... 548
Rock Type (Optional)............................................................................................... 551
Saturation Endpoints for each Grid Block (Optional) .............................................. 552
Maximum Relative Permeability/Capillary Pressure Values for Each Grid Block
(Optional) ............................................................................................................ 559
Water Oil-Capillary Pressure (J Function) Shift for Each Grid Block (Optional) ...... 562
Non-Darcy Flow in Reservoir (Optional)................................................................. 563
Non-Darcy Coefficient Correction Factor (Conditional) ......................................... 566
Fracture Velocity Width Correction for PolymerWater Mixture Viscosity
(Conditional) ....................................................................................................... 567
Pressure Gradient Threshold Arrays ........................................................................ 568
Leverett J Function Option....................................................................................... 570
J Function Surface Tension Arrays .......................................................................... 573
Scale Deposition Table (Optional) ........................................................................... 574
Scale Damage Table (Optional) ............................................................................... 576
Subdomain Reinfiltration (Optional)........................................................................ 577
vi Contents
User's Guide IMEX
Initial Conditions
579
Notes on Initial Conditions .......................................................................................579

Initial Conditions Identifier (Required) ....................................................................582
User Input or Vertical Equilibrium Selection (Required) .........................................583
Initial Oil Phase Reservoir Pressure (Conditional) ...................................................587
Initial Bubble Point Pressure (Conditional) ..............................................................588
Initial Bubble Point Pressure vs. Depth (Conditional)..............................................589
Initial Dew Point Pressure (Conditional)..................................................................591
Initial Dew Point Pressure vs. Depth (Conditional)..................................................592
Initial Solvent Bubble Point Pressure (Conditional).................................................594
Initial Oil Saturation (Conditional)...........................................................................595
Initial Water Saturation (Conditional) ......................................................................596
Initial Polymer Concentration (Conditional) ............................................................597
Initial Seawater Volume Fraction (Conditional).......................................................598
Initial Light Oil Volume Fraction (Conditional).......................................................599
Initial Light Oil API Volume Fraction vs. Depth (Conditional)...............................600
Reference Depth and Reference Pressure (Conditional)...........................................602
Depth to Water-Oil Contact (Conditional) ...............................................................604
Datum Depth Specification (Optional) .....................................................................607
Specification of Capillary Pressure Values at Contacts (Optional) ..........................610
Specification of Method for Computing Oil Saturations in Gas Zone......................613
Specification of Water Saturation Above GOC or Below WOC (Optional) ............614
Water Saturation Used with Block_Center Initialization (Conditional) ...................616
Non Equilibrium Water and Oil Saturation Used with Gravity-Capillary
Equilibrium Initialization (Conditional) ..............................................................618
Numerical Methods Control
621
Numerical Methods Control Identifier (Optional) ....................................................621

Maximum Timesteps (Optional)...............................................................................622
Maximum CPU Seconds (Optional) .........................................................................623
Maximum Timestep Size (Optional) ........................................................................624
Minimum Timestep Size (Optional) .........................................................................625
Normal Variation in Variables per Timestep (Optional) ..........................................626
Relaxation Options (Optional)..................................................................................628
Adaptive Implicit Switching (Optional) ...................................................................630
Convergence of Newton's Method (Optional) ..........................................................632
Maximum Newtonian Cycles (Optional)..................................................................635
Minimum Number of Newtonian Cycles (Optional) ........................................ 636
Number of Cuts (Optional) .......................................................................................637
Convergence Tolerance for Linear Solver (Optional) ..............................................638
Convergence Tolerance for Linear Solver (Optional) ..............................................639
Orthogonalization (Optional)....................................................................................640
Solver Equation Ordering (Optional)........................................................................641
User's Guide IMEX
Contents vii
Solver Factorization Degree (Optional) ................................................................... 642

Pivot Stabilization (Optional)................................................................................... 643
Maximum Iterations (Optional)................................................................................ 644
AIMSOL/PARASOL Switch (Optional).................................................................. 645
Red-Black Ordering Check for Parasol (Optional) .................................................. 647
Factorization Degree within PARASOL Classes (Optional) ................................... 648
Factorization Degree between PARASOL Classes (Optional) ................................ 649
Parasol Class Partitioning Pattern (Optional)........................................................... 650
Target number of Planes per Jacobian Domain (Optional) ...................................... 655
Number of Threads to be Used (Optional) ............................................................... 656
Complete Storage Grid Array of Jacobian Domain Numbers (Optional) ................ 657
Target Number of Jacobian Domains (Optional) ..................................................... 658
Target Number of Jacobian Domains based on Parasol Classes (Optional) ............ 660
Well and Recurrent Data
663
Notes on Well and Recurrent Data........................................................................... 663

New Keywords and Options for the 2010.10 Release.............................................. 665
New Keywords and Options for the 98.05 Release.................................................. 683
Old Options No Longer Supported by IMEX 97.00 ................................................ 691
Well Management and Group Control ..................................................................... 692
Specifying the Well and Group Control Hierarchy .................................................. 694
Introducing Group Control after Start of Simulation ............................................... 696
Data Incompatibilities Between 93.00 and Previous Versions of IMEX ................. 697
Limitations of the Well Management and Group Control Module .......................... 698
Recurrent Data from Other Sections ........................................................................ 699
Well and Recurrent Data Identifier (Required) ........................................................ 701
Well Change Date (Conditional) .............................................................................. 702
Well Change Time (Conditional) ............................................................................. 703
Well Change First Time-Step Size (Optional) ......................................................... 704
viii Contents
User's Guide IMEX
Maximum Time-Step Size (Optional) ......................................................................705

Minimum Time-Step Size (Optional) .......................................................................706
Setting Grid Blocks to Implicit or Explicit (Optional) .............................................707
Setting Well Blocks and Neighbours to Implicit (Optional).....................................709
Set Frequency of Initialization of Bottom-Hole Pressure (Optional) .......................711
Group Identification (Optional) ................................................................................714
Well Identification (Required)..................................................................................719
Well Backflow Model (Optional) .............................................................................722
Well Head Method (Optional) ..................................................................................724
Well Head Iteration (Optional) .................................................................................727
Simultaneous BHP and Well Head Iteration (Optional)...........................................729
Perforations in Inactive Blocks (Optional) ...............................................................731
Well Type Definition (Required)..............................................................................733
Shut in Wells above Formation (Optional) ...............................................................735
Well Status Definition (Optional).............................................................................737
Shut and Reopen a List of Wells (Optional) .............................................................739
Tubing Data for Injectors (Conditional) ...................................................................741
Tubing Data for Producers (Conditional) .................................................................744
Composition of Injected Water, Gas, and Oil Phases (Conditional).........................749
Well Operating Constraints (Required) ....................................................................751
Maximum Number of Continue-Repeat (Optional)..................................................760
Monitored Well Constraints (Optional) ....................................................................761
Alter Primary Well Operating Constraint Value (Optional) .....................................768
Alter Well Constraint Value (Optional)....................................................................771
Sets/Alters a Wells Phase Productivity (Optional)..................................................775
User-Specified Reference Depth for Well BHP (Optional)......................................780
User-Specified Pressure Gradient For Reference Depth for Well BHP (Optional) ......782
Pressure Gradients for Calculation of Pressure Differences Between
Completions (Conditional)...................................................................................784
Alter Polymer Concentration (Optional) ..................................................................787
Alter Seawater Injection Volume Fraction (Optional)..............................................789
Well Scale Table Assignment (Optional) .................................................................791
Layer Scale Table Assignment (Optional)................................................................793
Well Scale Removal (Optional) ................................................................................795
Layer Scale Removal (Optional) ..............................................................................797
Data for Workover Action for Wells (Optional).......................................................798
Resetting Well Operating Constraint after Value Change (Optional) ......................801
Gas Lift Option (Optional) .......................................................................................803
Gas Lift Control (Optional) ......................................................................................806
Gas Lift Optimization (Optional)..............................................................................809
Well Element Geometry (Conditional).....................................................................812
Location of Well Completions (Conditional) ...........................................................815
Location of Vertical Well Completions (Conditional) .............................................824
Location of Vertical Well Completions for Hybrid Refined Grids (Conditional) ........831
User's Guide IMEX
Contents ix
Geometric Data for Deviated Well Completions (Conditional)............................... 839

Simplified Geometric Data for Deviated Well Completions (Conditional) ............. 843
Data for Hydraulic Fracture Proxy (HFP) Model (Conditional) .............................. 846
Special Relative Permeability Data for Well Completions (Conditional)................ 851
Group Production Constraints (Optional) ................................................................ 856
Group Injection Constraints (Optional).................................................................... 862
Gas Make-up Target for Group Gas Recycling (Optional) ...................................... 872
Maximum Make-up Gas Rate for Group Gas Recycling (Optional)........................ 874
Gas Producing Group for Group Recycling (Optional)............................................ 876
Maximum Re-injection Rate for Group Gas Recycling (Optional) ......................... 878
Fuel Consumption Fraction for Group Gas Recycling (Optional) ........................... 880
Maximum Fuel Consumption Rate for Group Gas Recycling (Optional)................ 882
Sales Rate Fraction for Group Gas Recycling (Optional) ........................................ 884
Maximum Sales Rate for Group Gas Recycling (Optional)..................................... 886
Water Make-up Target for Group Water Recycling (Optional) ............................... 888
Maximum Water Make-up Rate for Group Water Recycling (Optional)................. 890
Water Producing Group for Group Recycling (Optional) ........................................ 892
Maximum Re-injection Rate for Group Water Recycling (Optional) ...................... 894
Solvent Make-up Target for Group Solvent Recycling (Optional) .......................... 896
Maximum Make-up Solvent Rate for Group Solvent Recycling (Optional)............ 898
Solvent Producing Group for Group Recycling (Optional)...................................... 900
Maximum Re-injection Rate for Group Solvent Recycling (Optional).................... 902
Monitored Group Constraints (Optional) ................................................................. 904
Allow a Cycling Group to be Defined (Optional) .................................................... 908
Allow Cycling Group Data to be Revised (Optional) .............................................. 919
Allow Cycling Group Data to be Revised (Optional) .............................................. 921
Select Cycling Group Cycle Part to Start and End Cycling Group Control
(Optional) ............................................................................................................ 923
Defining Group Production or Injection as Going Through a Manifold (Optional)..... 925
Pressure Constraint Translation for Manifolds (Optional) ....................................... 927
Specification of Hydraulics Tables for Calculating Pressure Difference Between
Manifold and Surface (Optional)......................................................................... 929
Manifold Depth for Calculating Pressure Difference Between Manifold and
Surface (Optional) ............................................................................................... 931
Group Artificial Lift Quantity Value (Optional) ...................................................... 933
Well Artificial Lift Quantity Value (Optional) ........................................................ 935
Priority List for Automatic Drilling of Wells (Optional) ......................................... 937
Group Apportionment Options (Optional) ............................................................... 939
Apportionment Method for Meeting Group Targets (Optional) .............................. 941
Priority Formulae for Apportionment (Conditional) ................................................ 945
Guide Rates for Groups or Wells ............................................................................. 949
Flag for Accompanying Groups or Wells Not Under Group Control (Optional) ....... 952
Allows for a Set of Keywords to be Processed When a Specified Condition
(Trigger) is Satisfied (Optional) .......................................................................... 954
x Contents
User's Guide IMEX
Alter Well Constraint Value via a Multiplier (Optional) ..........................................974

Group Production Constraints Multiplier (Optional)................................................978
Group Injection Constraints Multipliers (Optional)..................................................982
Define Reporting Group (Optional)..........................................................................987
Well/Group On-time Fraction (Optional) .................................................................989
Hydraulic Pressure Table (Conditional) ...................................................................993
Set Number of Points for WHP Root Search (Optional) ........................................1001
Terminate Simulation (Required) ...........................................................................1003
Documentation, Templates and Execution
1005
Introduction.............................................................................................................1005
Documentation........................................................................................................1005
Templates................................................................................................................1006
Running IMEX .......................................................................................................1011
Sample Data Sets
1013
Notes on Sample Data Sets .....................................................................................1013

MXSPE001.dat: First SPE Comparative Solution Model .....................................1017
MXSPE002.dat: Three-Phase Coning Model ........................................................1020
MXWWM004.dat WAG with Simple Wellbore Model........................................1023
MXWWM005.dat: Voidage Replacement & Lease Line Model...........................1030
MXFRR002.dat: Dual Permeability Radial Coning Model...................................1033
MXGRO008.dat: Hybrid Refined Grid Model ......................................................1037
Appendix A
1041
Theoretical Outline .................................................................................................1041

Four Component Black-Oil Model with Dual Porosity .............................1041
Dual Permeability ......................................................................................1045
Subdomain Model......................................................................................1046
Matrix-Fracture Transfer Coefficient, sigma.............................................1046
Independent and Dependent Variables ......................................................1047
Appendix B
1051
Well Model .............................................................................................................1051

Total Mobility Weighted Injectors.............................................................1051
Unweighted Injectors.................................................................................1051
Well Injectivity ..........................................................................................1051
Producers ...................................................................................................1052
Wellhead Pressure Wellbore Model ..........................................................1052
User's Guide IMEX
Contents xi
Appendix C
1055
Calculation of Well Parameters.............................................................................. 1055
Appendix D
1059
Figure D-1a: Numbering of the Grid (Non-Corner Point) .................................... 1059

Figure D-1b: Numbering of the Grid (Corner Point) ............................................ 1060
Figure D-2: Cylindrical Reservoir......................................................................... 1061
Figure D-3: Calculation of Gravity Components for a Tilted Grid System .......... 1062
Figure D-4: Radial (Cylindrical) Coordinates ....................................................... 1063
Figure D-5a: Depth Measurement ......................................................................... 1064
Figure D-5b: 3-Dimensional Variable Thickness Grid System (8x3x2) ............... 1065
Figure D-6: Fault Throw Definition ...................................................................... 1066
Figure D-7: Typical Data Curves .......................................................................... 1067
Figure D-8: Typical Relative Permeability Curves ............................................... 1068
Appendix E
1069
Polymer Option ...................................................................................................... 1069

Factors and Mechanisms Affecting Polymer Floods................................. 1069
Polymer Types........................................................................................... 1075
Mass Conservation Equations ................................................................... 1076
Guidelines for Polymer Application.......................................................... 1077
References for Polymer Option................................................................. 1078
Pseudo-Miscible Option ......................................................................................... 1079
Mass Conservation Equations ................................................................... 1079
Pseudo-Miscible Considerations ............................................................... 1080
References for Pseudo-Miscible Option.................................................... 1085
Appendix F
1087
The Use of IMEX in Well Testing Applications.................................................... 1087
Appendix G
1095
Analytical Aquifers ................................................................................................ 1095

Mathematical Background ..................................................................................... 1096
Van-Everdingen and Hurst Method .......................................................... 1096
Infinite Aquifer with Constant Rate Solution............................................ 1096
Limited Aquifers with Constant Rate Solution ......................................... 1097
Carter-Tracy Method................................................................................. 1097
Fetkovich Method ..................................................................................... 1098
xii Contents
User's Guide IMEX
Practical Aspects of Using Analytical Aquifers .....................................................1099

a) Selecting Type of Aquifer ....................................................................1099
b) Selecting Method of Solution ...............................................................1099
Aquifer Parameters ....................................................................................1100
Model Defaults .......................................................................................................1101
Appendix H
1109
Running IMEX Coupled to the GAP Surface Network Model...............................1109

Remote Simulation Job Submission via SSH .........................................................1122
IMEX Results Accessible from RESOLVE Script .................................................1125
Operation on Well Layers via RESOLVE Script....................................................1131
Keyword Index
User's Guide IMEX
1133
Contents xiii
Introduction
Important Changes between IMEX 2010.10 and IMEX 2009.10

NOTES FOR IMEX 2010.10
Well Module
1. Completion Lumping Option
2. *WHYSTAB *AUTOWELL Monitor
3. *MXCNRPT Maximum number of Continue Repeats
4. *KRPERF *SORMAX Entry for Trapped Oil Hysteresis
Simulator
1. Improved Parallel Performance
2. Trapped Oil Hysteresis Option
3. IMEX-GAP Interface Improvements:
Handles Crossflow during Shutin (like *MODELSHUT)
4. Flux Sector Reporting Option
5. Entry of Permeability and Non Darcy Correction Factor in Recurrent Data
Completion Lumping Option
*LAYERCLUMP provides a way for the user to define a control lump - a set of layers
(perforations, completions) which possesses a name and a set of computed rate, cumulative,
and stream ratio values. To each control lump there is also an associated setting value.
*LAYERRLUMP provides a way for the user to lump well layers in a group called a report
lump. For report lumps, stream rates, cumulatives and production rate ratios are calculated
and the user can design triggers based on these quantities. A given layer can be assigned a
partial membership to report lump and the same layer can belong to more than one group.
*RM-LAYERCLUMP removes membership of a layer from the named control lump.
*RM-LAYERRLUMP removes membership of a layer from the named report lump.
*CLUMPSETTING assigns a control setting or flow restriction parameter to the named
control lumps. The flow restriction parameter is used as a multiplication factor in adjusting
the value of the well index for all the layers belonging to a given control lump.
User's Guide IMEX
Introduction 1
There is also a set of computed rate, cumulative, and stream ratio values for each control
lump. Triggers may be defined using these quantities specific to named control lumps;
actions can be any valid well data lines, including lines which alter the control lump setting
values. The syntax of the control lump based trigger statement is:
*TRIGGER trigger_name *ON_CTRLLUMP control_lump_name quantity > value
The syntax of the report lump based trigger statement is:
*TRIGGER trigger_name *ON_RPTLUMP report_lump_name quantity > value
See *LAYERCLUMP keyword for a list of quantities which can be used in control and report
lump triggers.
*WHYSTAB *AUTOWELL Monitor
*MONITOR WHYSTAB *AUTOWELL frequency can be used to monitor production
well wellbore hydraulics stability using the *AUTOWELL option, which shuts the well for a
specified time.
*MXCNRPT Maximum Number of Continue Repeats
Every well having *CONT *REPEAT (continue-repeat) specified through *OPERATE is
entitled to one repeat of the current time step if one of its operating constraints is violated.
During convergence of a time step, it is possible that different wells violate and then switch
their operating constraints consecutively after each repeat. Although the maximum number of
possible repeat cannot exceed the number of wells, it may lead to a significant increase of run
time if there are many wells. Quantity nmxrpt limits the number of such repeats per time step,
and keyword *MXCNRPT lets the user override its default value (of 1).
*KRPERF *SORMAX Entry for Trapped Oil Hysteresis
SORMAX on the *KRPERF keyword allow the user to change this saturation endpoint in
recurrent data. Sormax is the oil saturation in the two phase oil-water system at which oil first
becomes immobile during the boundary/primary imbibition process (the imbibition starting
from oil saturation equal to 1-Swcon). It only has an effect when the Krow hysteresis or
trapped oil hysteresis is modeled for the rock type of the completion.
Improved Parallel Performance
Improvements to the implementation of parallelization in IMEX have allowed the Simulator
to run more efficiently on larger number of threads on Shared Memory Processors.
Trapped Oil Hysteresis Option
Trapped oil hysteresis is now available for Pcow and Krow curves. To invoke the trapped oil
hysteresis option, it is only necessary to define the imbibition Pcow in an SWT table using
constant Pcow for water saturation values equal to and great than 1-Sormax. Once activated,
trapped oil hysteresis couples Pcow and Krow hysteresis. Keyword *HYSKRO is not
necessary and will be ignored. The endpoint array *SORMAX has been added to allow the
user to enter values of Sormax on a block by block basis. The keyword *DAMP-PCOW-TROIL
has been added to reduce the capillary pressure difference between imbibition and drainage curve
which might result in a pressure oscillation. Please see the Tutorial section: The Trapped Oil
Hysteresis Option for Pcow and Krow (Oil-Water System).
2 Introduction
User's Guide IMEX
IMEX-GAP Interface Improvements IMEX-GAP Interface Improvements:

Handles Crossflow during Shutin (like *MODELSHUT)
When wells are shutin by GAP but experience crossflow during the shutin, the resulting
pressure profiles around the well may change drastically if the productivity of well layers
vary drastically and crossflow is occurring. This feature allows IMEX to shutin a GAP
/RESOLVE controlled well using a *MODELSHUT type option, which allows crossflow in
the shutin well. GAP/RESOLVE controls whether the well is *MODELSHUT or simply
shutin. The user does not need to specify this IMEX keyword.
Flux Sector Reporting Option
*FLUX_SECTORNAMES and *FLUX_ISECTOR keywords introduces special sector definitions
which are used to calculate reservoir flow between these flux sectors. Flow into these
sectors and between these sectors can be reported to the output print file and to the simulation
results file as time series data. Flow can be further broken down into I-direction, J-direction
and K-direction flow components.
*FLUX_SECTORNAMES defines a list of flux sector names and their corresponding flux
sector numbers. *FLUX_ISECTOR assigns these flux sector numbers to cells using standard
array input options. Both keywords can only appear once in the data set. Please also see the
*WPRN *SECTOR, *WSRF *SECTOR, *OUTPRN *FLUX_SECTOR, and *OUTSRF
*FLUX_SECTOR keywords for output options.
Entry of Permeability and Non Darcy Correction Factor in Recurrent data
This option allows the input/redefinition of the permeability arrays *PERMI, *PERMJ and
*PERMK and the non Darcy correction factor *NDARCYCOR in recurrent data.
User's Guide IMEX
Introduction 3
Important Changes between IMEX 2009.10 and IMEX 2008.10

Well Module
1. *ITUBE1/*PTUBE1
2. *GCONI Stream *PMAINT *PMMAXR d1 d2
3. *GCONPMULT *RECYCLE
4. *LAYERHFP
5. *PERFRG keyword removed
Simulator
1. Improved Parallel Performance
2. *DRSDT
3. Oil wet Initialization Option
4. *RTYPE Subkeyword of Relative Permeability Scaling Arrays
5. IMEX-GAP Interface Improvements:
Layer Information Available for Scripting
Improved IPR Curve Generation
6. PTUBE1/*ITUBE1 Table Generation in Builder
*ITUBE1/*PTUBE1
PTUBE has been replaced by the more flexible PTUBE1, in addition ITUBE1 allows for the
definition of injection tables using CMG standard well head pressure input table format.
*PTUBE tables are still supported.
*GCONI Stream *PMAINT *PMMAXR d1 d2
Introduces the maximum surface rate of the specified injection stream that is available for the
pressure maintenance. It is the summation of two sources: a fraction (d1) of the total
produced (from the same group) and a make-up rate (d2). The values must be non-negative
real numbers. The fraction d1 is nondimensional and defaulted to 0. The make-up rate d2 is
(m3/day | bbl/day | cm3/min) for water and (m3/day | scf/day | cm3/min) for gas or solvent, and
is defaulted to 1.0e+20.
4 Introduction
User's Guide IMEX
*GCONPMULT *RECYCLE
This subkeyword specifies a recycling production target. This indicates that the production
wells connected to this group produce such that the phase injected by the injection wells
connected to this group as specified by *GAS, *WATER or *SOLVENT is reproduced
(recycled) out of the reservoir.
*LAYERHFP
*LAYERHFP allows the user to supply geometric and non Darcy information specifying
productivity indices which act as proxies for vertical fractures (assuming pseudo radial flow)
on a layer by layer basis.
*PERFRG keyword removed
This keyword is no longer required as the *PERF keyword can be used instead.
to run more efficiently on larger number of Shared Memory Processors.
*DRSDT
This option allows the input of the maximum rate of increase of solution gas ratio.
Oil Wet Initialization Option
An oil wet reservoir when initialized using the depth average initialization option now
produces realistic saturation profiles. This assumes that the reservoir was initially water wet
and over time changed to an oil wet condition.
*RTYPE Subkeyword of Relative Permeability Scaling Arrays
The relative permeability scaling arrays can now be made to operate on blocks which are
members of specified rocktypes (RTYPES).
IMEX-GAP Interface Improvements
The IMEX-GAP interface now is able to output well-layer level information to Resolve, this
enables the user to script in Resolve using well-layer information.
The IPR curve generation now optimally assigns pressure points in the IPR table to try to
capture curvature more accurately.
*PTUBE1/*ITUBE1 Table Generation in Builder
The new PTUBE1/ITUBE1 well head pressure tables are supported by the well head pressure
calculator program. This now includes generation of injection tables as well as production
tables which incorporate gas lift.
User's Guide IMEX
Introduction 5
Important Changes Between IMEX 2008.10 and IMEX 2007.10

Well Module
1. Cycling Groups
2. *GAPPOR *AUTODRILL stream identifier
Simulator
1. Improved Parallel performance
2. Improvements to the IMEX-GAP interface
3. NONDARCY GENERAL option
4. Fracture Cleanup Model
Cycling Groups
Allow the user to specify data for EOR/IOR processes which rely on alternating production
and/or injection cycles. Generally multiple cycles are required and each cycle consists of a
number of well defined parts. Examples include water alternating gas injection (WAG) or
cyclic steam injection.
*GCONCYCLE_START allows the user to begin to specify data for EOR/IOR processes
which rely on alternating production and/or injection cycles.
*GCONCYCR_START allows the user to revise data specified earlier with
*GCONCYCLE_START. The revisions will be applied immediately.
*GCONCYCLE allows the user to turn on or off group cycling control specified earlier with
*GCONCYCLE_START. The action indicated with GCONCYCLE takes effect immediately.
*CYCPRT_START allows the user to select the cycle starting part. CYCPRT_END allows
the user to select the cycle ending part. Group cycling control must be specified earlier with
*GCONCYCLE_START. The action indicated with CYCPRT_START and/or
CYCPRT_END take effect immediately.
GAPPOR Stream Specification
The group control may direct a group to switch target (value and stream) if there is a
maximum constraint (*GCONP *MAX) being violated. *AUTODRILL followed by the
stream identifier ensures that auto-drillable wells can be opened only when the group is
apportioned for the desired stream.
Improvements to the IMEX-GAP interface
LSF and SSH network job submission is now supported. The GAP down-time factor is now
passed to IMEX in order to calculate the IMEX on-time factor. Petroleum Experts is now
supporting the IMEX-GAP DLL so the advanced IPR calculations in Resolve are now
available in this DLL.
6 Introduction
User's Guide IMEX
NONDARCY GENERAL Option

The Non Darcy model now has a more general input method to input parameters for the
factor, in addition to the existing techniques. This input also allows non Darcy flow to be
modeled in all phases, not only the gas phase. In addition, input parameters are made available
to improve convergence of the Forchheimer number iteration (see Tutorial Section Modeling
Non Darcy Flow in Hydraulic Fractures Accurately Using a Grid Based Approach).
Fracture Cleanup Model
The fracture cleanup model is a group of options which allows the user to model the
production and clean-up of fracturing fluids used to create/extend a fracture followed by
production from the fracture.
Velocity dependent polymer-water mixture viscosity is modeled
(Keyword: PMIX VELTABLE)
A fracture width velocity correction term is added (Keyword: FRWIDTHCOR)
Block and direction dependent pressure gradient thresholds are modeled
(Keyword: PTHRESHI/PTHRESHJ/PTHRESHK)
Non equilibrium initial saturation overrides have been added (Keyword:
SWNEQ/SONEQ)
The pressure gradient threshold option can be initiated at a restart
See the Tutorial Section Using the Fracture Cleanup Model and review the keywords
mentioned above for more details.
User's Guide IMEX
Introduction 7

Well Module
1. Shut in Wells above Formation *MODELSHUT
2. Two New Drawdown Constraint Types *DWA, *DWB
3. Monitoring Well Head Pressure Stability *WHYSTAB
4. Voidage Replacement Production Target
5. Recycling Production Target *RECYCLE
6. Sector Pressure Maintenance Production Target *PMAINT
7. Shutting in Multiple Wells to meet Production Targets/Monitors *SHUTMOWS
8. Voidage Replacement Injection Targets including Makeup of Gas, Water or
Solvent *GMKUP, *WMKUP, *SMKUP
9. Sector Pressure Maintenance Injection Target *PMAINT
10. New Apportionment Methods for Meeting Group Targets *APPOR_METHOD,
*IP, *GUIDE, *INGUIDE, *PRIOR, *PRIOR-FORM
11. On-Time Factors Applying to Groups
Simulator
1. Improved Parallel performance
2. Improvements to the IMEX-GAP interface
3. Volatile Oil Model
4. Subdomain-Dual Permeability Model
5. Water Polymer Mixture Viscosity Table
6. Depletion Option
7. Improved Aquifer Printout
Shut in Wells above Formation
*MODELSHUT indicates that the listed wells identified by well_list will be closed above
formation with crossflow modelling of the constituent (open) layers, whenever the wells are
shut in due to a well operation.
Two New Drawdown Constraint Types
*DWA defines the maximum drawdown within all open layers,
Pd = max (Pblock l Pwelll )

l, open
which corresponds to the notion of imposing the maximum drawdown constraint to avoid
formation damage.
8 Introduction
User's Guide IMEX
*DWB specifies the average drawdown for all open layers:
Pd =
PI l (Pblock l Pwelll ) PI l
l, open
l,open
weighted by the total product/injectivity (PI) at the reservoir condition.

Monitoring Well Head Pressure Stability
A real well can operate only above a certain rate at which the bottom-hole pressure remains
in the stable region of the well-head pressure curve (a region in which the curve of WHP vs.
BHP has a positive slope). It can only be used for wells for which a method of computing
WHP has been introduced with the *PWELLBORE or *IWELLBORE keyword.
Voidage Replacement Production Target (*GCONP *VREP)
This indicates that the production wells connected to this group produce an amount of the
bottom-hole fluid in proportion to the total bottom-hole fluid injected into the reservoir by the
injection wells connected to this group.
Recycling Production Target (*GCONP *RECYCLE)
This indicates that the production wells connected to this group produce such that the phase
injected by the injection wells connected to this group as specified by *GAS, *WATER or
*SOLVENT is reproduced (recycled) out of the reservoir.
Sector Pressure Maintenance Production Target (*GCONP *PMAINT)
This subkeyword specifies that the group production rates shall be adjusted so as to maintain
the hydrocarbon volume weighted average pressure in a particular region/sector (*PMSECT)
at a desired level (*PMTARG).
Shutting in Multiple Wells to Meet Production Targets/Monitors (Subkeyword
*SHUTMOWS)
Action subkeyword indicating that if a maximum stock tank rate is exceeded for a group, then
a list of prioritized most offending wells (MOWS the wells with the higher rates of the
named surface stream) should be shut.
Voidage Replacement Injection Targets including Makeup of Gas, Water or Solvent
(GCONI *VREP *GMKUP/*WMKUP/*SMKUP)
A make-up stream can be added with *GMKUP, *WMKUP or *SMKUP to meet a total
voidage replacement fraction for each group under voidage replacement.
Sector Pressure Maintenance Injection Target
This subkeyword specifies that the group injection rates shall be adjusted so as to maintain
New Apportionment Methods for Meeting Group Targets
*APPOR-METHOD defines how to distribute a group target within its pool the collection
of all the contributing wells or groups. The pool for a targeted group consists of all its child
groups and the wells directly attached. . CMGs multi-level well management module
currently supports the following methods:
User's Guide IMEX
Introduction 9
Instantaneous potentials (*IP, default)

Guide rates (*GUIDE)
Internally generated guide rates (*INGUIDE)
Priority ranking (*PRIOR)
*PRIOR-FORM defines the priority formulae and numerical control parameters for the priority
ranking apportionment method set by *APPOR-METHOD *PRIOR to meet group targets.
On-Time Factors Applying to Groups
*ON-TIME specifies the fraction of time during which a well or group operates. Please see
the *ON-TIME keyword for a detailed description of how *ON-TIME operates with Groups.
Improvements to the IMEX-GAP interface
The Voidage replacement interface in RESOLVE is now supported. Scripting using IMEX
variables is supported and IMEX variables are now available for output in RESOLVE.
TRIGGERS may now be used in IMEX-GAP runs. Selected Simulator well control is now
enabled in IMEX-GAP runs. Please See Appendix H for details.
Volatile Oil Model
The Volatile oil model uses an extension of the *BLACKOIL option. Gas can exist in
solution in the oil phase and oil can exist in the gas phase. This option is similar to the
*GASWATER_WITH_CONDENSATE. It is more flexible in that the gas phase need not
always exist. This allows the model to be used in situations where oil can be above its bubble
point. The volatile oil model uses many of the same keywords as the *BLACKOIL model
and GASWATER_WITH_CONDENSATE model. In addition, it introduces a new PVT
table option *PVTVO, which includes both wet and dry gas properties. Please See the
Tutorial section for more details in using the volatile oil model.
Subdomain-Dual Permeability Model
The SD-DK model is an extension of the Subdomain model which allows the Simulator to
model gravity drainage and reinfiltration processes. Please see the *SUBDOMAIN,
*FRACVOL, *TRANSD and *SD_REINF keywords for details. Also see the Tutorial
section for a discussion on how to use the SD-DK model.
Water Polymer Mixture Viscosity Table
Previously, water viscosity in polymer model runs were analytical functions of pure water
viscosity and polymer concentration. Polymerwater mixture viscosity can now be defined
using tables of relative polymer concentration versus water viscosity (through a water
viscosity multiplying factor). Please see the *PMIX *TABLE keyword for more details.
10 Introduction
User's Guide IMEX
Depletion Option
The *DEPLETION option makes the input reference porosity (*POR) the actual input
porosity and makes the porosity a function of pressure depletion rather than pressure. The
*DEPLETION keyword in the data set indicates that all *CROCKTYPES using *CCPOR
(*CPROR), *CROCKTAB and *CROCKTABH use the Depletion Option. See the
*DEPLETION keyword in the Reservoir Description section for more details.
Improved Aquifer Printout
The normal printout of initial aquifer properties has been extended to include the ratio of
aquifer volume to reservoir volume. Aquifer volume includes both the volume associated
with the analytical aquifer and the volume associated with entirely water filled blocks in the
numerical model. Please see the *AQUIFER keyword in the Reservoir Description section
for details.
User's Guide IMEX
Introduction 11

1. Seawater/Scale Buildup Option
2. TRIGGER Action Option
3. Improved Parallelization Performance using *JACDOMS
4. Well Definition Order Relaxed/Improved
5. *WLISTSHUT/*WLISTOPEN Option
6. *WTMULT/*GCONIMULT/*GCONPMULT Options
7. Improvements to Tubing Head Pressure Calculations
8
*WOC_SW available with *DEPTH_AVE Initialization.
9. Reinfiltration Modeling for Fractured Reservoirs (Beta)

10. Binary File Reading Option
11. DPCONNECT
Seawater Injection/Scale Buildup Option
A Seawater Injection/Scale buildup option has been incorporated in IMEX. This feature is
actually made up of two independent options: 1) a Seawater Injection Tracking Option and 2)
a Scale Buildup/Well Damage model. The two models can be coupled to make well damage a
function of both water rate and the fraction of seawater in the produced water.
The Seawater Injection option adds a seawater flow equation to the normal Black Oil or Oil
Water Models in IMEX. This equation is solved simultaneously with the oil, water and gas
flow equations. It is an active model, and the injected water may have a viscosity different
from formation water. Thus it is possible to model seawater injection altering fluid flow
patterns within the reservoir. This model and the Scale Buildup/Well Damage model are
coupled through the *SCLDPS table (Scale Buildup/Well Damage Option) which relates
scale deposition to the volume fraction of seawater in the produced water. The Seawater
Option is controlled by the following new keywords:
1.
2.
3.
4.
5.
6.
*MODEL *BLACKOIL_SEAWATER
*MODEL *OILWATER_SEAWATER
*OUTPRN/*OUTSRF *GRID *SEAWATFRC
*CONVERGE *MAXRES *SEAWATER
*INCOMP SEAWATER
*ALTERFS
The Scale Buildup/Well Damage Model allows the user to model well (layer) productivity
damage based on the amount of water produced through each layer of a well. Optionally the
damage can be made a function of the fraction of seawater produced with formation water.
12 Introduction
User's Guide IMEX
This Option is controlled by the following new keywords:

1.
2.
3.
4.
5.
6.
*SCLDPS
*SCLDMG
*SCLTBL-WELL
*SCLTBL-LAYER
*SCLRMV-WELL
*SCLRMV-LAYER
Rock Fluid Data

Rock Fluid Data
The Scale Buildup/Damage model can be used with any Property model available in IMEX (for
example: API Tracking, Solvent, and Gas-water). It is only with the *BLACKOIL_SEAWATER
and *OILWATER_SEAWATER models that the active seawater equation can be used to make
scale formation a function of seawater fraction produced.
Trigger Action Option (Well and Recurrent Data)
This keyword allows for certain actions to be implemented when a specified condition or
trigger is satisfied. The actions are specified in the form of a block of valid keywords
encapsulated within the *TRIGGER and *END_TRIGGER keywords.
The trigger condition of each active trigger is tested at the bottom (end) of the time step (that
is after the time step is considered to be completed, the cumulatives have been updated and
the simulation is ready to proceed to the next time step). If the trigger condition is satisfied
then the list of associated actions are processed at the top (beginning) of the next time step.
The order of the triggers in the trigger list depends entirely on the sequence on which the
triggers are defined. The triggers are not sorted in any other way.
It is possible to nest triggers by including the definition of the inner trigger within the
keyword set of the outer trigger (that is before specification of the *END_TRIGGER
keyword of the outer trigger). Please see the *TRIGGER Keyword in the Well and
Recurrent Data Section for a complete description of the option.
*JACDOMS Parallelization Option (Numerical Methods Control)
*JACDOMS (-jacdoms on the command line) enables a new technique to control the
Jacobian domains used for parallelization. This technique is generally more efficient than
those previously used to control Jacobian Domains (controlled using the *DPLANES or
*DTYPE keywords) and is recommended.
More Flexible Well Definition Ordering (Well and Recurrent Data)
It is now possible to define a well, name it, and specify its group affiliation with a *WELL
keyword at one time, specify its completions with *PERF at a later time, and finally define
the wells type with *PRODUCER or *INJECTOR at a still later time and have the well
become active. Previously it was necessary to define a wells type (injector or producer)
immediately after its initial definition.
*WLISTSHUT, *WLISTOPEN (Well and Recurrent Data)
*WLISTSHUT and *WLISTOPEN provide a means to temporarily shut a large list of wells
and later re-open them without disturbing the pattern of shut and autodrillable wells that
existed when the *WLISTSHUT keyword was read.
User's Guide IMEX
Introduction 13
*WTMULT (Well and Recurrent Data)

*WTMULT allows modification of any previously specified well constraint value for well(s)
listed by well_numbers or 'well_names' by applying a multiplying factor.
*GCONIMULT, *GCONPMULT (Well and Recurrent Data)
*GCONIMULT is used to specify multipliers which modify existing group injection target
controls. *GCONPMULT is used to modify existing group production target controls with
the use of multipliers.
Mechanistic/Momentum Wellbore Pressure Drop Model Improvements
The Mechanistic Wellbore pressure drop calculation in IMEX (*HEAD-METHOD
*MODEL-MOMENTUM) has been made more rigorous. The updated Momentum option is
currently a full implementation of the Petalas-Aziz mechanistic model.
*WOC_SW in the *DEPTH_AVE Vertical Equilibrium Initialization Option
*WOC_SW allows the user to specify non-zero oil saturation below the water oil contact.
Small non-zero water zone oil saturation is often useful in avoiding convergence difficulties
if oil attempts to move into the water zone. Previously, when this option was required it was
necessary to use *BLOCK_CENTER initialization.
Reinfiltration Model for Fractured Reservoirs (Beta - no Builder Support or Manual
Entries)
The Subdomain-Dual Permeability-Reinfiltration (SD-DK) model is controlled by the following
Keywords,
*SUBDOMAIN turns on subdomain dual porosity model,
*FRACVOL controls subdomain block thickness,
*TRANSD controls SD-DK inter-subdomain connections, and
*SD_REINF determines if SD-DK inter-subdomain connections are used to
represent reinfiltration or gravity drainage.
Modeling capabilities include:
1. Standard Subdomain Model
If the *TRANSD keyword is not included in the model, *SD_REINF is not
required and the model reverts to the standard *SUBDOMAIN model. It is still an
excellent idea to use the new *FRACVOL keyword to minimize the error in the
capillary holdup.
2. Gravity Drainage between Subdomains
Using the *TRANSD keyword to set non-zero transmissibility multipliers between
domains allows the fluid in the bottom of a matrix subdomain to flow into the top
of the matrix subdomain directly below it without having to flow into the fracture
first. In essence this is a hybrid method which combines the Subdomain model
and the Dual Permeability model and gives similar results. As there is no capillary
holdup in the matrix, using *FRACVOL does not significantly improve the models
results. As *TRANSD is grid block dependent, gravity drainage between
subdomains can be modeled in user specified regions of the reservoir.
14 Introduction
User's Guide IMEX
3. Reinfiltration between Subdomains.

As in 2) the non-zero *TRANSD keyword enables inter-subdomain flow. The use
of *FRACVOL and *SD_REINF together enables the use of the bottom-most
subdomain block as a proxy for a horizontal fracture between the subdomains.
*FRACVOL is used to reduce the thickness of the bottom division of the
subdomain which reduces the amount of fluid stored in the fracture proxy block
and minimizes the error in capillary holdup.
*SD_REINF = 1.0 sets the bottom subdomain divisions capillary pressure to 0.0
(the fractures value) and introduces a real capillary discontinuity in the flow
between the two subdomains. *SD_REINF is grid block dependent.
4. Gravity Drainage and Reinfiltration between Subdomains
Actual drainage from Subdomain blocks is a combination of gravity drainage and
reinfiltration. This may be approximately accounted for by scaling back the bottom
subdomain divisions capillary pressure rather than setting it explicitly to zero.
This can be accomplished by setting *SD_REINF to a value between 0.0 and 1.0.
The fracture proxy blocks capillary pressure is scaled by 1.0 - *SD_REINF.
See the document SD-DK with Reinf.doc in the DOC directory of your IMEX
2006.10 install directory for more details.
Binary File Reading Option
Normally Builder writes data in text-format files. However, Builder is able to write some
grid definition and property data in binary form to a separate binary-format file. This option
is invoked in Builder via menu "File/Save As.../Array Saving Method/Binary File Format
(*.cmgbin)". The binary file is saved in the same folder as the main dataset file and given the
same root name but extension .cmgbin. Unlike the *INCLUDE facility which can involve
multiple include files, there is at most one binary format file associated with a main data file.
When the simulator detects subkeyword *BINARY_DATA in its initial data scan, it opens
the associated binary file which is assumed to have the same root name as the main data file
but extension .cmgbin. Each time the simulator encounters *BINARY_DATA during the
data loading pass, it locates that property in the binary file and reads one value for each block
in the grid, similar to the *ALL option. A mismatch between the text and binary parts of the
data set will result in an error.
The reading of binary data is much faster than text and so for large models can speed up
significantly data reading in the simulator. Text formatted data can differ slightly in value
from its associated binary data, so text and binary versions of the same data may give slightly
different simulation results.
*DPCONNECT Option
The DPCONNECT keyword allows the grid module to produce extra connection between
blocks in a dual porosity/permeability system when either the matrix or the fracture is
missing from one of the two blocks being connected.
User's Guide IMEX
Introduction 15
When enabled, a dual porosity or dual permeability block with a missing fracture will connect
the matrix to both the matrix and fracture of surrounding blocks. In addition a dual porosity
or dual permeability block with a missing matrix will connect the fracture to both the matrix
and fracture of surrounding blocks.
16 Introduction
User's Guide IMEX

1. Improvements have been made in frictional pressure drop calculations. Two new
template data sets have been added to demonstrate some of these features:
mxwwm028.dat and mxwwm029.dat
2. Manifold group options have been added. Four new templates have been added to
demonstrate some of these features: mxwwm030.dat, mxwwm031.dat,
mxwwm032.dat and mxwwm033.dat
3. Beta version of coupling to Petroleum Experts Gap surface facility model is now
available. Please see the installed Doc directory for more information.
4. The *TRANSF keyword for modelling fault transmissibilites has been added.
5. A beta 64 bit Linux version is now available for Red Hat Linux Enterprise 3 for
Intel EM64T enabled Xeon processors.
6. A beta 64 bit Windows version for Windows x64 for Intel EM64T enabled Xeon
processors is available.
7. For parallel IMEX, options are now available to use up to 64 processors for a
single run. Improvements have been made in load balancing with changed defaults
for Parasol. More non-solver portions of the code have been made parallel.
8. The maxsteps n command line option has been added.
In the Reservoir Description Section, the keyword *TRANSF has been added. This keyword
may only appear in this section and not in recurrent data. It allows the user to specify fault
transmissibility multipliers which are independent of the multipliers specified in the
*TRANSI, *TRANSJ and *TRANK keywords.
The use of the *MOD keyword has been altered. It is no longer permitted to have multiple
*MOD keywords following a keyword. However a single *MOD keyword may contain as
many mod definitions as necessary. For example:
*PERMI *CON 1000.00
*MOD 1:1 1:2 1:1 * 0.5
*MOD 2:2 1:2 1:2 * 0.25
is not permitted, while:

*PERMI *CON 1000.00
*MOD 1:1 1:2 1:1 * 0.5
2:2 1:2 1:2 * 0.25
is permitted.
Two major options have been added to IMEXs handling of groups and wells. The first is the
frictional pressure drop option; the second is the manifold group control option.
User's Guide IMEX
Introduction 17
Keywords associated with frictional pressure drop.

*HEAD-METHOD well_list *GRAV-FRIC (*MODEL-AG |MODEL-HOMOGENEOUS|
*MODEL-MOMENTUM | *MODEL-DRIFTFLUX) (*HEADRROUGH rrough)
(*HEADSURFT surft)
This keyword describes which frictional pressure drop calculation method is
used for wells in the well list.
*HEADITER well_list ( *ONE-STEP | *ITERATIVE)
This keyword specifies the calculation procedure for calculating the frictional
pressure drop. The *ITERATIVE method allows the model to compute
heads which are consistent with completion pressures. The *ONE-STEP
method does not iterate to calculate consistent heads and very often results in
unphysical results.
*BHPHEADINIT well_list ( *SEPARATE | *COMBINED )
This keyword when used determines whether well bottom hole pressures as
well as layer heads are made consistent with completion pressures. This
option is not normally required for good results as the HEADITER
*ITERATIVE option is quite robust on its own. Using the *COMBINED
option can result in significantly increased run times.
New keywords associated with the Manifold Model.
*MANIFOLD (*PROD | *GASI | *WATI | *SOLI) group_list (*ON | *OFF)
Determines which groups are manifold groups.
*PCON-MANIFOLD (*PROD | *GASI | *WATI | *SOLI) group_list (*ON | *OFF)
Determines whether rate constraints are translated into manifold pressure
constraints.
*GPHYDDEP (*PROD | *GASI | *WATI | *SOLI) group_list depths
This keyword adjusts the depth used to calculate the pressure drop between a
surface compressor and a subsea manifold.
*GPTABLE (*PROD | *GASI | *WATI | *SOLI) group_list
This keyword determines which table is used to calculate the pressure drop
between a surface compressor and a subsea manifold.
*GROUPALQ (*PROD | *GASI | *WATI | *SOLI) group_list
This keyword specifies the group ALQ property which is used in standard
VFPPROD and VFPINJ tables which are associated with flow from manifold
to compressor.
18 Introduction
User's Guide IMEX
*WELLALQ well_list
This keyword specifies the well ALQ property which is used in standard well
VFPPROD and VFPINJ tables.
Other keywords which are modified to account for manifold groups are:
*GCONI, *GCONP, *INCOMP and *IWELLBORE
User's Guide IMEX
Introduction 19

1. A 32 bit Linux version is now available.
2. A technique for modeling tilted water oil contacts has been developed using
PCOW_SHIFT (see tutorial section).
3. Multiple datum depths are supported
single run.
5. The checkonly option has been improved to read and process all of the well data
6. The 3 phase oil relative permeability when using Stones Second Model (the
default) is checked for consistency at low oil saturations. Specifically IMEX now
warns if non-zero Kro is observed when oil saturation equals zero.
7. Warning messages are written for inconsistent capillary pressure curves.
8. PVT consistency checks have been extended.
9. Warning messages are written when PBT table minimum and maximum depths
alter the WOC or GOC of a PVT region.
10. Two new well group monitor constraints have been added.
11. The maximum number of characters in a well name has been increased from 16 to
40. The number of characters in group names, sector names and leaseline names
have not changed and remain at 16.
12. The maximum number of characters per line in the data file has been increased
from 130 to 512.
NEW KEYWORDS AND OPTIONS ADDED TO IMEX 2004.10
*OUTPRN/OUTSRF *RES *PCWSHF *JFWSHIF
These keywords enable the output of capillary pressure and J function
shift/offsets.
Rock-Fluid Data
*PCOW_SHIFT/*JFW_SHIFT
This keyword enables the input of water capillary pressure or water J
function shifts. This allows the user to model tilted water contacts.
Initial Conditions
*DATUMDEPTH
The DATUMDEPTH keyword can now be applied separately to each PVT
region. Within each PVT region different calculation options may be used.
20 Introduction
User's Guide IMEX

*GCONM *MAXGAS
These keywords identify a maximum surface gas rate monitor for group
production. The *STOP, *SHUTALL, *SHUTMOW, *SHUTMOL,
*SHUTMOLDOWN, and *SHUTMOLUP actions are valid for this monitor.
The most offending well or layer for this constraint is deemed to be the one
with the highest GOR rather than the one with the highest gas rate.
*GCONM *MAXSTW
These keywords identify a maximum water rate monitor for group
with the highest WCUT rather than the one with the highest water rate.
User's Guide IMEX
Introduction 21

1. Peak storage requirements have been drastically reduced. For large 32-bit and
large 64-bit versions the peak storage requirements are approximately one-half of
what they where for 2002.10.
2. Support for very large 64-bit problems has been added and tested on problems up
to 112,000,000 grid blocks.
single run.
4. The sector output, both to the ascii file and the SR2, now includes average P/Z values
weighted according to both block pore volume and block hydrocarbon pore volume.
In order to use this feature, reservoir temperature must be defined (see *TRES)
5. Streamline output is enabled for display in results.
6. Gas condensate modeling capability has been added.
7. Scaling of relative permeability, capillary pressure and J function maximums (i.e.
Krw at irreducible oil, Krow at connate water, etc) has been included.
*OUTSRF GRID *STRMLN
This keyword enables the output of interblock velocities to allow Results3D
to display Streamlines. See the results manual for more information on
Streamline display. Due to the large size of the SR2 output, streamline
display is not enabled when the *ALL output option is used with OUTSRF
GRID. Streamline output must be explicitly requested.
*OUTPRN/OUTSRF *RES *KRWSCL *KROWSCL *PCWSCL *KRGSCL *KROGSCL
*PCGSCL
These keywords enable the output of initial relative permeability, capillary
pressure and J function maximums.
*OUTPRN/OUTSRF *GRID *KRWSCL *KROWSCL *PCWSCL *KRGSCL *KROGSCL
*PCGSCL *RS *BO *EG *RV *DPP
These keywords enable the output of relative permeability, capillary pressure
and J function maximums and the output of Rs, Bo, Eg, oil content Rv (for
condensates) and dew point pressures (for condensates) in recurrent output. The
relative permeability table maximums are only output at times other than the
initial time if they are explicitly requested on the OUTPRN GRID or OUTSRF
GRID keywords, the *ALL output option will not enable their output.
22 Introduction
User's Guide IMEX
*PSPLIT *ON
This keyword enables the output of surface production splits for black oil and
condensate models. In a black oil model the gas production is split into gas
at surface from solution gas in the reservoir and gas at surface from free gas
in the reservoir.
In a condensate model, in addition to the splitting of the surface gas
production, oil production is also split. Oil is split into oil from gas in the
reservoir and oil from mobile oil in the reservoir.
*ISECTOR
*SECTORNAMES
These keywords enable an alternate method of entering sector information.
Use of these keywords may make sector data conversion to IMEX from other
simulators more straightforward.
Once read into MODELBUILDER, these keywords are translated into
SECTORARRAY keywords.
*MODEL *GASWATER_WITH_CONDENSATE
This keyword enables the gas water with condensate model.
*PVTCOND
This keyword signals the input of the main condensate PVT table. Saturated
gas and condensate properties are entered in this table.
*BOT
This keyword enables the input of under saturated Bo. This allows the user
to directly input Bo as a function of pressure and bubble point pressure.
*VOT
This keyword enables the input of under saturated oil viscosity. This allows
the user to directly input oil viscosity as a function of pressure and bubble
point pressure.
*EGUST/*BGUST/*ZGUST
These keywords enable the input of Gas Eg/Bg/Zg for pressures above the
dew point when using the GASWATER_WITH_CONDENSATE option.
*VGUST
This keyword enables the input of Gas viscosity for pressures above the dew
point when using the GASWATER_WITH_CONDENSATE option.
User's Guide IMEX
Introduction 23
Rock-Fluid Data
*KRWIRO *KROCW *PCWMAX *JFWMAX *KRGCL *KROGCG *PCGMAX
*JFGMAX
These array keywords enable the input of relative permeability table
maximums on a block by block basis in initial and recurrent data.
The user can now enter the values of Krw at irreducible oil, Krow at connate
water, Pcow (or water J Function) at connate water, Krg at connate liquid, Krog
at connate gas and Pcog (or gas J function) at connate liquid for each block.
This can be done both in the rock-fluid data section and in the recurrent data
section.
Initial Conditions
*PDEW
*PDEWT
These keywords enable the input of initial block dew point pressures or
initial dew point pressure versus depth tables for each PVT region. When
using the GASWATER_WITH_CONDENSATE option initial dew point
pressures must be defined.
Numerical
*NORM *PDW
*MAXCHANGE *PDW
These keywords enable the input of time stepping control based on the
changes in dew point pressure during the previous time step. These
keywords can only be used with the GASWATER_WITH_CONDENSATE
option.
*CONVERGE *MAXRES (*OIL|*WATER|*GAS|*OTHER|*SOLVENT|*POLYMER|
*LIGHTOIL)
These keywords enable the control of maximum residual based convergence
tolerances by individual phase or component. This allows more precise
control of a run. In addition, when a higher level of output information is
selected (see *WPRN *ITER) each phases maximum residual error is
displayed.
*PTUBE *CONDENSATE
This keyword enables the use of condensate tubing head pressure tables.
These tables can be used with the GASWATER_WITH_CONDENSATE
option.
24 Introduction
User's Guide IMEX

The recommended procedure for running on a dual processor Pentium III or Pentium 4 is to
use the two command-line arguments -doms parasol described in Numerical below.
The -doms switch enables parallelization of the Jacobian building and some other nonsolver portions of the code.
The -parasol switch enables the use of the parallel linear solver, Parasol, instead of the
usual non-parallel solver, AIMSOL.
The use of either of these parallel IMEX options requires that you have a license with
Parallel IMEX enabled.
Parasol will in general give slightly different answers than AIMSOL, as it is a different
iterative solver.
There are a number of additional options related to Parasol and the parallel Jacobian building.
These are described below and in the Numerical Methods Control section of this manual.
The -aimsol command line switch overrides PARASOL related keywords and forces the
use of AIMSOL.
This version has a correction for the *SCONNECT Keyword when used with dual porosity
and dual permeability models.
The use of *MODSI units has been corrected in the aquifer model.
The *MAXRES convergence tolerance has been made more consistent when used with the
Polymer option. This will correct problems encountered in convergence when dispersion
coefficients are used.
The 2002.10 release changes the time unit of *DISPI, *DISPJ and *DISPK from seconds to
the unit defined by the users choice of *INUNIT. The user must change data accordingly.
A problem with WHP controlled wells getting unphysically large pressure derivatives has
been corrected.
The use of the *ONTIME feature with *WHP controlled wells has been corrected.
If the head is changed for a temporarily shut-in well, then, if it is reopened during the same
timestep it is reset to its original value.
If a well has its constraint type changed during apportionment, its bottom hole pressure is reinitialized. This produces more consistent results.
For pseudo-miscible runs, there may be a slight drift in answers. Smoothing for So-Sorms
has been made more consistent.
Handling of oil phase disappearance has been made more consistent.
For non-physical pressures (less than zero), if they occur in a block which contains a well, the
corresponding well is temporarily shut in.
The handling of non-physical pressures during convergence has been improved.
User's Guide IMEX
Introduction 25
The crossflow handling of injectors has been improved to correctly account for crossflowing
solution gas.
Special release 64 bit versions of IMEX have been created for Itanium WinXP 64-bit and
Power 3, 4 AIX 5.1 IBM RS6000s.
The *MODEL *LIGHTOIL Option is undergoing major revision and is not currently
available. Please contact CMG Support for more information.
Numerical
Command line arguments:
-doms
which is equivalent to
*DPLANES
and overrides the DTYPE/DPLANES keywords in the dataset, and
-parasol
which is equivalent to the
*SOLVER *PARASOL
*PPATTERN 2
keywords
-aimsol
which overrides any parasol keywords in the Numerical Methods Control section
and forces the use of AIMSOL.
Non-solver parallelization of IMEX:
*DTYPE
Complete storage grid array input of Jacobian domain numbers.
*DPLANES imxdom
imxdom target number of planes per domain Planes are chosen in the
dimension with the largest number of non-trivial planes. imxdom is the
number of corresponding non-trivial planes in this direction per domain. If
imxdom not specified and DPLANES is, then the default is 4.
*PNTHRDS
Number of threads to be used - if parallel Jacobian building and Parasol are
not specified, then defaulted to one. If parallel Jacobian building or Parasol
is specified, then it is defaulted to the min of (the number of processors in
current machine, 2).
If PNTHRDS is set to a number greater than the number of processors,
performance will degrade.
If PNTHRDS is set to greater than two then the solver PPATTERN should be
changed in order to load balance properly, otherwise poor performance is
likely to occur.
26 Introduction
User's Guide IMEX
Keywords related to PARASOL:

*SOLVER (*AIMSOL | *PARASOL)
Choose which solver to use, AIMSOL or PARASOL.
Default if no *SOLVER keyword is *SOLVER *AIMSOL.
*PNSUBD nclas2
Choose the number of PARASOL classes used for vector operations in the
GMRES iteration. Default is 2.
*CHECKRB (*ON | *OFF)
When ON, red-black ordering is abandoned for a PARASOL class in which
the fraction of red blocks is too small.
Default is OFF.
*PDEGAA idegaa
Factorization degree to be used within PARASOL classes. Default is the
value of *SDEGREE (1 for red-black, rcmrb; 2 for natural, rcm).
*PDEGAB idegab
Factorization degree to be used between PARASOL classes. Defaults to
idegaa + 1.
*PPATTERN ( ipatrn | *PARTITION | *PPARTITION | *GPARTITION | *APARTITION )
PPATTERN sets the basic partitioning of the reservoir into non-connected
regions and separators that makes possible the parallelization of the linear
solution. Default is *PPATTERN 2.
Please see the Numerical Methods Control section for a complete description
of the *PPATTERN keyword.
There are two major additions to the well data for this release: The introduction of reporting
groups (the *REPORTING-GROUP keyword) and the addition of a bottom-hole fluid (BHF)
group production rate target and monitor.
The purpose of the reporting groups is to allow reporting of results for well aggregates which
are very flexibly defined, without the restrictions which apply to the hierarchical groups
through which group controls are imposed.
*REPORTING-GROUP reporting_group_name well_list weight_list
*REPORTING-GROUP allows the user to define a set of wells with
differing membership weights. No Group Control can be specified for a
reporting group, but there are no restrictions placed on well membership in
reporting groups. The weight_list allows the reporting group to handle
element of symmetry output.
User's Guide IMEX
Introduction 27
The BHF group target allows group production strategies to be set during prediction runs
which may better allow reservoir pressure to be maintained while respecting maximum rate
limits upon injectors when used in combination with the injection voidage replacement target.
*GCONP
*GCONP
*GCONM
*GUIDEP
group_name_1 group_name_n
*MAX
*BHF
value
action
*TARGET
*BHF
value
action
*GCONP *MAX *BHF and *GCONP *TARGET *BHF : allow
the user to control bottom-hole fluid group production rates.
*MINBHF
value
( *STOP | *SHUTALL)
*GCONM *MINBHF: allows the user to monitor group bottom-hole
fluid group production rates.
*BHF
group_or_well_list
guide_rates
*GUIDEP *BHF: allows the user to specify guide rates proportional to
bottom-hole fluid rates.
There is also a minor keyword addition in this section. The keyword *NWFHYS has been
added to give users extra control over some aspects of the numerical method used to compute
the bottom-hole pressure which yields a specified tubing head pressure. In nearly all cases
the default will work well and the user need not be concerned with this keyword. Please see
the relevant manual pages for more detail.
*NWHYFS
well_list
Integer_list
28 Introduction
User's Guide IMEX

*SBDZ
IMEX has added subsidence to the grid_list for *OUTPRN *GRID.
*SBDZ
IMEX has added subsidence to the grid_list for *OUTSRF *GRID.
See the Tutorial Section for more information
The user should notice that the display of well rates has been sped up greatly in
ResultsGraph, this is due to a ResultsGraph optimization which makes use of extra
information the simulator writes to the SR2 binary files.
The *LIST/*NOLIST options for displaying echoed data has been altered. The default under
the *LIST option is to display the first 20 lines of data, the remainder of the data is not
echoed. This default action can be overridden by placing the keyword *NOLISTLIM in the
Input/Output control section.
When output of reservoir condition well variables is enabled, the voidage replacement ratio
defined by the group control option is output to the text output file and to the SR2 binary file
for display using Results Graph.
The grid module routines have been sped up significantly in this release. The improvements
will be most evident in large corner-point models where initialization can now take place
using approximately one-third the CPU time.
The grid model can now automatically handle the accidental user refinement of null blocks.
Previous versions of the Grid Module would signal an error if this occurred.
The handling of pinchouts has been made consistent between corner point and non-corner
point grids. Pinchout out connections that may have been previously missed will now be
correctly made.
Water properties can now be defined per PVT region. Oil viscosity change with pressure
(*CVO) can now be defined per PVT region. The input structure is both backwards
compatible and more flexible than previous versions. The tutorial section has been altered to
reflect these changes, please see Defining Multiple PVT Regions in the Tutorial Section.
All oil and water properties per PVT region are echoed in the simulator output file (.out).
In addition, it is now possible to define an oil compressibility for each PVT region using
*CO. It is no longer necessary to include a *COT table for each PVT region or include oil
compressibility as a function of bubble point pressure in each PVT table to accomplish this.
User's Guide IMEX
Introduction 29
Rock-Fluid Model
Capillary pressure hysteresis in the Pcog table and Krow hysteresis has been added to the
simulator. See the keywords below and the *SGT/*SLT keywords for a detailed description
of the options.
*HYSKRO
Krow hysteresis using Carlsons model (non-wetting phase).
*EPSPCG
Pcog hysteresis using Killoughs model.
*RPT *IMBIBITION (*PCOW *PCOG *BOTH)
The user can specify which curve (imbibition or drainage) to begin the
simulation on. The specification is done independently for Pcow and Pcog.
The *SWTKRTHR and *SLTKRTHR keywords are no longer sub-keywords of the
*SMOOTH keyword (for *SWT and *SLT/*SGT). They are now sub-keywords of *SWT
and *SLT/*SGT. This allows the user to change the threshold saturation for endpoint
determination without using the smoothing options.
Initialization
The initialization input has been made more flexible when using multiple PVT regions. Now
multiple PVT region Contacts (such as WOC or GOC or WGC) can be placed on a single
*DWOC, *DGOC or *DWGC keyword. In addition, multiple PVT region reference depths
and pressures can be placed on a single *REFDEPTH or *REFPRES keyword. Please see
Defining Multiple PVT Regions in the Tutorial Section for more details.
Contacts and reference pressures are now echoed in the simulation output file (.out)
Numerical
*AIM *STAB *AND-THRESH is now the default adaptive implicit method used. This was
changed from *AIM *STAB.
Neighbours of implicit blocks have their IMPES stability checked once per timestep,
however; neighbours of blocks that have just switched from IMPES to implicit on that
timestep are no longer checked. The *AND_THRESH sub-keyword ensures that all IMPES
blocks have their maximum saturation and pressure changes checked every iteration for
IMPES to implicit switching.
The result of this change is generally a more stable simulation.
*GEOA and *KHA
The anisotropic Peaceman well model has been added to the simulator. This
model accounts automatically for permeability anisotrophy and gridblock
aspect ratio. It is available on the well model based well index calculations
(*GEO => *GEOA, and *KH => *KHA). The geometry keyword remains
as is, but the geometric factor C (normally 0.37) is ignored as it is calculated
internally.
30 Introduction
User's Guide IMEX
This feature is especially useful when used with the *LAYERXYZ and
*LAYERIJK deviated well options.
*LAYERIJK
The new *LAYERIJK keyword allows the user to override, on a layer basis,
the well direction input on the *GEOMETRY keyword.
The *LAYERXYZ allows the user to enter trajectory information into the
simulator through Builder. In addition, *LAYERXYZ information can be
synthesized from Builder without requiring trajectory data.
However, for simple well trajectories, the *LAYERIJK keyword allows the
user to model some aspects of well deviation without using builder or
requiring trajectory files.
*GMKUPTO *WMKUPTO *SMKUPTO
*GMKMAXR *WMKMAXR *SMKMAXR
*GRECYMAXR *WRECYMAXR *SRECYMAXR
*GPRODGROUP*WPRODGROUP
*SPRODGROUP *GSALESFRAC *GSALESMAXR
*GFUELFRAC *GFUELMAXR
Enhancements to the group recycling options have been made in this version
of IMEX. In addition to reducing the amount of gas injected due to fuel and
sales gas, options to separately specify maximum recycling rates and total
injection rates has been implemented for gas, water, and solvent recycling.
See the *GCONI keyword in this manual as it describes in detail the recycling calculation. In
addition, see the manual entries describing the recycling options listed above.
*BHPDEPTH *BHPGRAD *LAYERGRAD
These new keywords allow the user to specify a depth to which a wells
bottom hole pressure is referred to.
The use of quotes in well names is now supported in a consistent manner. To use a double
quote in a well name, use a single quote as a well string delimiter. To use a single quote in a
well name, use a double quote as a well string delimiter.
User's Guide IMEX
Introduction 31

FORGAS Integration
Previously, if any well had any change (e.g. ALTER), even a non-FORGAS well, then all
FORGAS wells were reset for the following timestep to the current IMEX specification,
over-writing the FORGAS values. Now, even if a FORGAS well is ALTERd, except for the
time where it is initially defined, the *ALTER will be ignored and the value from FORGAS
will be used.
The IMEX calculated values for a and b for the IPR calculation for FORGAS are now based,
for wells perfd in more than one grid block, on the surface mobility weighted average delta
Ps (or delta P squared for quadratic pressure) rather than the value at the reference layer.
Also, the reservoir pressure passed to FORGAS was formerly the reservoir pressure in the
reference layer. Now it is the surface mobility weighted average of (reservoir pressure head). This new approach, in some cases, significantly reduces oscillations in rates.
Sectors are now viewable in RESULTS. When probing a block in RESULTS, the sectors the
block is in will be displayed. In addition, It is possible to select viewable regions by sector in
both 2D and 3D views.
Command Line Options:
IMEX has the new command line option
-log
Directs screen input to a log file which has the root file name of the
data set and the extension ".log". Screen input is read from a file
which has the root name of the data set and the extension ".in". If a
file name has not been defined using the f command line option
and the log option is used, it can be entered in the ".in" file. In
addition, the user *INTERRUPT *INTERACTIVE choice is also
entered in the ".in" file when the log option is used. A carriage
return in the ".in" file where the simulator is expecting to read the
interrupt option defaults to *RESTART-STOP.
*NDBETCR, *WSRFTN, *GSRFTN, *CRSETN, *SWCON, *SWCRIT, *SOIRW,

*SORW, *SGCON, *SGCRIT, *SLCON, *SOIRG, *SORG
IMEX has added the following variables to the res_list for both *OUTPRN
*RES and *OUTSRF *RES. *NDBETCR displays the non-Darcy correction
factor, *WSRFTN and *GSRFTN display the water-oil and gas-liquid
surface tension arrays respectively. *CRSETN displays the compaction rock
type of each block. The remaining keywords display water-oil and gas-liquid
relative permeability table end point arrays.
32 Introduction
User's Guide IMEX
*PCOW, *PCOG, *RFG, *SWCON, *SWCRIT, *SOIRW, *SORW, *SGCON, *SGCRIT,

*SLCON, *SOIRG, *SORG
IMEX has added the following variables to the grid_list for both *OUTPRN
*GRID and OUTSRF *GRID. *PCOW and PCOG display the capillary
pressures of each block, *RFG displays the non-Darcy flow reduction factor.
The remaining keywords display water-oil and gas-liquid relative permeability
table end point arrays.
*WELL-SECTOR
This keyword specifies that well in sector information will be printed out.
Well in sector information consists of two tables. The first is ordered by
sectors (sector table) and lists all active wells within a sector (an active well
has at least one perforation open). The second table is ordered by well (well
table) and lists all sectors associated with each active perforation of every well.
Wells which have all perforations closed or have not had any perforations
defined will not be listed in either table. Individual perforations that are no
longer active will not be listed in the well table. Perforations which have had
their production/injection zeroed due to use of the *XFLOW-MODEL
*ZERO-FLOW keyword will not be listed in the well table for as long as
their production/injection is zeroed. Active perforations of wells that are
shut in will be listed in the well table, but shut in wells are not listed in
the sector table.
The table information is printed out every timestep to a separate file with
extension sct, the file will be created if when scanning the data IMEX finds
the *OUTPRN *WELL-SECTOR keywords. The amount of printout can be
controlled by using the *NONE sub-keyword in recurrent data.
Sector calculations in some instances were taking excessive amount of run-time. This was
due to an unfortunate choice of algorithm for sector lookup during sector calculations. This
algorithm was revised, resulting in remarkable improvements in run-time. For example, on a
field problem with 159 sectors, total run-time was reduced by a factor of 2.56 with an
increase in storage allocated of only 0.4%.
The compaction option has been revised to remove several limitations. Compaction can be
used with dual porosity and dual permeability models. In addition the following changes
were made:
1. When values for the third column in a rock table are blank or negative, horizontal
permeability multipliers are defaulted to one.
2. When values of the fourth column in a rock table are blank or negative, vertical
permeability multipliers are defaulted to be the same as the corresponding
horizontal permeability multipliers.
User's Guide IMEX
Introduction 33
3. When a value of a permeability multiplier is greater than or equal to zero and less
than 1.0E-04, it is replaced by a value of 1.0E-04.
4. When a value of a porosity multiplier is less than 1.0E-02, it will be replaced by a
value of 1.0E-02.Small values of porosity may result in poor performance of the
simulator.
5. If there is only one rock type and there is no keyword *CTYPE used, all the grid
blocks are assigned to have that rock type.
6. There is no longer a limit to the number of compaction rock types which may be
used in a data set.
7. There is no longer a limit to the number of rows in a table permitted under the
*CROCKTAB or *CROCKTABH keywords.
8. There is no limitation in the number of hysteresis curves for a given rock type.
However, the first row of each sub-table under the keyword *CROCKTABH must
be identical to the corresponding row on the main path from which the hysteresis
path starts as discussed in the user manual.
9. The *CROCKTYPE region definitions may exist for region numbers not specified
for any given block using the *CTYPE keyword. However, if a block is set to a
particular region type number using *CTYPE, then there must exist a
corresponding *CROCKTYPE definition.
10. The compaction rock region types (input using *CTYPE) may now be output using
the sub keyword *CRSETN with *OUTPRN *RES and/or *OUTSRF *RES.
A dilation option was added, see the Reservoir Description section and the Tutorial section
for further information.
A pseudo-capillary pressure option was added to the dual porosity and dual permeability
option. See the keyword *TRANSFER for a complete description of the option.
The *PINCHOUTARRAY keyword was added to the reservoir description section. This
allows for an alternative method of pinchout definition.
The use of surface tension input on *PVT tables has been extended to work with multiple
PVT regions. The surface tension is used to correct Pcog with respect to pressure. A single
reference pressure is input for the surface tension correction, but this reference pressure
*REFPST is used to calculate a separate reference surface tension for each table.
The use of oil compressibility (Co) tables for oil above the bubble point at any pressure has
been supported in IMEX for some time. However, the conversion of Bos above the bubble
point to Cos is not trivial as IMEX replaces the original Cos with integrated averages. We
have added the ability of automatically converting Bos to Cos to ModelBuilder. The user
may now use ModelBuilder and enter either Bos or Cos above the bubble point.
34 Introduction
User's Guide IMEX
Rock-Fluid Model
*SWCON, *SWCRIT, *SOIRW, *SORW, *SGCON, *SGCRIT, *SLCON, *SORG
The end point scaling option had been completely rewritten and redesigned.
Instead of the four scaling points: 1) connate water (*SWCON previously
*SWC), 2) critical gas (*SGCRIT previously *SGC), 3) residual oil in the
oil-water system (*SORW), and 4) residual oil in gas-liquid system
(*SORG). We now scale with respect to:
Connate Water Saturation
(*SWCON)
Critical Water Saturation
(*SWCRIT)
Connate Gas Saturation
(*SGCON)
Critical Gas Saturation
(*SGCRIT)
Irreducible Oil Saturation (oil-water system)
(*SOIRW)
Residual Oil Saturation (oil-water system)
(*SORW)
Connate Liquid Saturation (gas-liquid system)
(*SLCON)
Residual Oil Saturation (gas-liquid system)
(*SORG)
There is a complete description of the new option in the Rock-Fluid section
under the keyword *SWCON.
Use of the new scaling produces more consistent scaling of the relative
permeability tables and initialization.
The use of the new end point scaling option may alter previous results. For
the sake of compatibility, we have allowed the older four point end point
scaling method to be used (see *RPT *SCALING-OLD).
A Leverett J Function option has been implemented into IMEX. This option
allows the user to replace water-oil and/or gas-liquid capillary pressures with
Leverett J functions. Capillary pressures are calculated from the J function
tables, input surface tension, and account for the varying permeability and
porosity of each block. In addition the effect of pressure on Pcog can be
accounted for by entering surface tension as a function of pressure on PVT
tables. See the keyword *JFUNC, *SRFTNW, *SRFTNG, *PVT, *REFPST
for details on the use of this option.
Non-Darcy gas flow in the reservoir can now be modeled in addition to nonDarcy flow into the well (see *TURB on the *PERF card). The non-Darcy
reservoir gas flow model uses the Forchheimer equation with the
Forchheimer number calculated using any of 1) Geertsmas 1974 correlation
(*NONDARCY *GEERTSMA), 2) Frederick and Graves first correlation
(*NONDARCY *FG1) and 3) Frederick and Graves second correlation
(*NONDARCY *FG2). The Forchheimer number for gas can be further
modified by the scaling parameter defined on the *NDARCYCOR keyword.
See *NONDARCY and *NDARCYCOR for more details.
User's Guide IMEX
Introduction 35
Initialization
The Initialization section has no obvious changes, but the new end point
scaling option and the Leverett J Function option both can alter initialization
results.
IMEX now allows the user to perforate null blocks. Previous versions of
IMEX stopped and reported errors when perforations went though null
blocks. This new feature allows perforations in null blocks, the perforations
are closed automatically (*NULL-PREF). This feature is on by default.
IMEX allows the user to enter the layer X, Y and Z coordinates of the
entrance and exit location of a well perforation running through a grid block.
With this information IMEX will calculate an accurate deviated well
productivity/injectivity index (*LAYERXYZ).
The *KRPERF keyword allows the user more flexible control of perforation
relative permeability arrays.
The *GCOMP and *GCONM keywords now allow the user to shut all wells
in a group when group constraints are violated.
Well reference layers can be defined anywhere in a branching well.
The *PERF keyword now can handle all refined grid situations including
hybrid grids. The *PERFRG, *PERFVHY and *PERFHY keywords are
valid but no longer required.
The *ON-TIME information of each well is now written to the SR2 file
system. This now allows RESULTS to display instantaneous rates as well as
average rates.
The *FULLY-MIXED wellbore crossflow model is now the default for all
wells.
36 Introduction
User's Guide IMEX

User block addresses (UBA) are used throughout the output file. No longer are block
addresses printed out in packed storage or complete storage format. UBA: I, J, K format is
used consistently throughout the printout.
IMEX has several new command line options.
`-wait'
Wait for License

If all available licenses are being used, this argument keeps the
process in a sleep mode until a license is available (up to 72 hrs).
`-dd'
Data Directory
Simulator output files are created in the directory where the data
files are specified. This can be different from the current directory
where the executable resides. Include files are assumed to be in the
same directory as data.
`-wd'
Working Directory
Simulator output files are created in the specified working directory.
Data files reside either in the current directory or in a directory
specified by the -f command line argument. Include files are
assumed to be in the same directory as data.
*PVTSETN
IMEX has added the following variable to the res_list for both *OUTPRN
*RES and *OUTSRF *RES.
*FLUXSC, *FLUXRC, *VELOCSC, *VELOCRC,
IMEX has added the following variables to the grid_list for OUTSRF *GRID.
The above keywords allow RESULTS to plot velocity and flow vectors on top
of displayed properties. The above can also be displayed for each direction,
but their primary purpose is for the generation of flux/velocity vectors.
*POROS, *PERM, *KRSETN
IMEX has added the following variables to the grid_list both *OUTPRN
*GRID and OUTSRF *GRID. *PERM displays permeability at each desired
output time, permeability is now a function of pressure and so varies with time.
User's Guide IMEX
Introduction 37
*WSRF *SECTOR
Lease line output to the SR2 system has been added. It is controlled by the
*SECTOR frequency of SR2 output.
Note: Grid Module keywords appearing in the Well or Recurrent Data section between two
*DATE or *TIME cards must be grouped together. Failure to do so could allow the Grid
Module to read only part of the data. The cards in question are *TRANSI, *TRANSJ,
*TRANSK, *TRANLI, *TRANLJ, *TRANLK, and *REFINE (RANGE).
The Grid Module has generally been sped up for most cases and now has improved tolerance
of odd corner point grids. Refined corner point grids are now allowed.
In addition, the algorithms for handling the connections made by sloping faults has been
rewritten and improved.
The method for handling transmissibility multiplier inheritance to refined grids has been
improved.
The handling of reservoir porosity versus pressure has been greatly extended in the IMEX
simulator. Multiple rock regions have been added as well as the use of pressure tables for
both porosity and permeability (vertical and horizontal) multipliers.
In addition the modelling of porosity and permeability hysteresis and irreversibility vs. pressure
allows IMEX to model compaction. See the keywords *CROCKTYPE, *CROCKTAB,
*CROCKTABH, *IRREVERS, *CCPOR, *CPRPOR and *CTYPE for a complete explanation.
IMEX includes significant additions to the Component Property section.
A gas-water (2-equation subset) has been added to IMEX. This allows a simpler data input
format to model gas-water problems as well as reduced execution times to run gas-water
problems. It is no longer necessary to enter fictitious oil properties to model gas-water
systems. Only gas and water properties need be entered. Gas-water component properties
keywords are listed below:
*MODEL *GASWATER
This keyword signals IMEX to use the 2-equation gas-water model subset of
the black-oil equations.
*PVTG
This keyword is used to input a gas PVT table (for use with GASWATER
option). Gas formation volume factor (Bg, Eg, or Zg) and gas viscosity are
tabulated against pressure.
Rock-Fluid Model
*SWT, *SLT, *SGT
The gas-water option triggers slightly different input in each of the above
tables. The tables no longer require input of either Krow or Krog. Pcow is
replaced by Pcgw. Pcog is not entered.
38 Introduction
User's Guide IMEX
The gas relative permeability hysteresis option and the Pcow hysteresis options have been
reviewed and improved. MXSMO024.dat has been added to the template directory to
highlight the Pcow hysteresis option. This new template duplicates the radial test example in
J.E. Killough's classic paper "Reservoir Simulation with History-Dependent Saturation
Functions", SPEJ, Feb. 1976, pp. 37.
Initialization
The Initialization section has no obvious changes, but the gas-water option allows the user to
use the option:
*VERTICAL *BLOCK_CENTER *WATER_GAS
when nonzero Pcgw is entered in the *SWT table. Previous versions of
IMEX forced the user to use:
*VERTICAL *DEPTH_AVE *WATER_GAS *TRANZONE *ON
to model gas-water capillary pressure. It is still required to use the
*TRANZONE *ON option with the *GASWATER option when
*DEPTH_AVE initialization is required.
IMEX includes a significant modification to the well module simulator interface. This
modification allows the well module better access to simulator information. This in turn
allows the well module to more accurately calculate well and group operating constraints.
As a consequence of the above, the *CONT *REPEAT subkeywords on operating and
monitor constraints are now operational. In addition, the well gravitational head calculation
has been moved to a position within the code which ensures the well module and the
simulator use identical heads.
Improvements have been made to how the well module extrapolates tubing head pressure
tables; more consistent results will now be obtained.
The well index printout (available when using *WPRN *WELL *ALL) has been greatly
enhanced. In addition to well index, the actual layer and total well fluid productivity/injectivity
is displayed.
As well as being able to display well productivity the user is now able to directly input fluid
phase or total well productivity/injectivity or productivity/injectivity multipliers using the
*SETPI keyword listed below.
*SETPI
SETPI can be used to alter the well productivity to measured values or by a
multiplicative factor correcting the last (*MULT) or the fundamental
(*MULTO) productivity.
User's Guide IMEX
Introduction 39
*SETPI
(*OIL)
well_list
(*WAT)
(*GAS)
(*LIQ)
(*TOT)
(*MULT)
(*MULTO)
Each time a *SETPI keyword is read, wells on the well list have their
fundamental productivity corrected to match the user input value. The
correction is only applied when the *SETPI keyword is read (i.e. when the
test was done). Well productivity will normally change as fluids move in the
vicinity of the well. However, productivities can be corrected at any later
time by using the *SETPI keyword again.
40 Introduction
User's Guide IMEX

For the 98.05 release the interface between the simulator and well management module has been
improved so that the action of repeating a timestep in response to a well constraint violation works
more smoothly than in past releases. This is now the strategy of choice if, for example, it is
extremely important that a certain well's bottomhole pressure never fall below a certain value.
Since certain constraint checks and constraint actions may be handled slightly differently in the
new version of IMEX, some data sets may run differently than with earlier versions. The user
should examine output carefully when comparing runs made with earlier versions of IMEX.
The default mode of the WHP well constraint is now *IMPLICIT instead of the former
default of *INITIALIZE. IMEX users have for the most part preferred the *IMPLICIT
mode. Please see the documentation for the *OPERATE keyword for more information.
Two new well-related keywords have been introduced:
*WORKOVER-DATA:
This keyword is intended to allow the user more direct control over the
parameters determining the workover action which can be imposed in
response to certain well monitor violations. It is intended to replace the entry
of these parameters after the *WORKOVER action subkeyword of the
*MONITOR keyword, which in the past was the only way to enter these
parameters; the old way, however, is still supported and documented. Please
see the manual pages for the *MONITOR and the *WORKOVER-DATA
keywords for more information.
*MRC-RESET:
This keyword allows the user to specify on a well- by-well basis whether a
well should be tested for and set to operate upon its most restrictive
constraint after a constraint value has been changed, or whether it should
continue with its current operating constraint and switch only if a constraint
is violated at the end of a timestep. The default is to check and switch to the
most restrictive constraint whenever a constraint value is changed through
data (through the *OPERATE, *ALTER, or *TARGET keywords), which
can save the simulator the sometimes difficult task of trying to cause a
timestep to converge with an unrealistic constraint; however the option of not
resetting is allowed in case that the user wishes to enforce operation upon a
certain constraint for the timestep. Please see the manual page for the
*MRC-RESET keyword for more information.
User's Guide IMEX
Introduction 41

IMEX 98.00 has several new command line options.
`-f input_data'
Defines the input data file.
`-r input_restart'
Defines the restart irf file.
`-checkonly'
Runs the model in check only mode. An IMEX

license is not required for a check only run specified
from the command line.
`-dimsum'
Prints detailed information on model dimensioning at

run time.
`-onestep'
Runs the model for one timestep only.
*MAX_GRIDS
The IMEX 98.00 model now has a *DIM *MAX_GRIDS subkeyword to
change the default number of refined grids.
*SR2PREC
The SR2 binary file is by default written out in double precision mode.
Single precision output is also possible. Single precision halves the size of
most SR2 files. Restart information is slightly effected by the loss of
precision.
*OUTPRN *GRID *APIGRAV
The API tracking model allows the printout of light oil volume fraction for
each block using the *OUTPRN *GRID *APIGRAV print option.
*LSOLVER
*DEBUG *LSOLVER *ON turns on the debug output of the number of
solver failures.
Both multilevel refinements and refinements defined in recurrent data have been added to
IMEX 98.00. This represents a major improvement in how the user may use IMEX to model
complex reservoirs.
A more robust handling of faults has been included as well as a more general implementation
of the corner point grid option.
The analytical aquifer model has been reworked, introducing a new option (Fetkovitch aquifer)
and improved Carter-Tracy influence function extrapolation and more realistic defaults.
42 Introduction
User's Guide IMEX
A feature to input corner point tolerances has been included to allow the user to tune whether
blocks are in contact with each other. In addition, this feature can be used to define a
distance within which a blocks corner is assumed to be pinched out.
Sectors can now be input as an array to further simplify input.
The user can now explicitly define 'special connections' between pairs of blocks along with a
transmissibility for the connection.
*REFINE
The *REFINE keyword may also occur as part of recurrent data. Refined
blocks may appear later in a simulation. *REFINE must be used to set up
refinements defined in recurrent data prior to using the *PERFRG keyword
which refers to them.
*RANGE
The *RANGE keyword can now refer to single multilevel refinement blocks
as well as to groups of fundamental blocks.
*AQMETHOD
The Fetkovitch aquifer calculation has been added to the analytical aquifer
calculation (*AQMETHOD *FETKOVITCH).
The Carter-Tracy aquifer calculation has been modified to 1) use more
realistic default values and 2) allow more realistic extrapolation of the
aquifer influence functions (*AQMETHOD *CARTER-TRACY).
A compatibility option *AQMETHOD *OLD has been included in this
release to allow the users to use the same *AQPROP defaulting as IMEX
97.00 and to use the same influence function extrapolation as IMEX 97.00.
*SECTORARRAY
It is now possible, in array input format, to define block which are in sectors.
This allows a streamlined input of sector information.
*SCONNECT
The *SCONNECT keyword can set up a flow connection between any two
reservoir blocks. Transmissibility is defined explicitly on the *SCONNECT
card.
*CORNER-TOL
The tolerance setting how close to each other corner point blocks have to be
before they are considered to be in contact is now user defined.
IMEX 98.00 includes significant additions to the Component Property section.
The phase appearance/disappearance algorithms have been rewritten from scratch to properly
account for the disappearance of multiple phases within a grid block (e.g. water and gas
within a solvent filled block).
User's Guide IMEX
Introduction 43
An API tracking option was developed (jointly with Petrobras) to model reservoirs with
simple compositional gradients.
A gas saturation threshold was added to the minimum solvent saturation to control onset/loss
of miscibility in the four component miscible model.
Both the solvent model's gas saturation threshold and minimum solvent saturation have had
smoothing applied to the cutoff values, this causes a smoother transition from miscible to
immiscible conditions (and the reverse) as well as improved numerical performance.
Previous restrictions which limited the *OILWATER option to use a single PVT region have
been removed. *GORINT is no longer valid when used with the *OILWATER option.
*MINSS
min_sol_sat'
(*SGTHRESH `sgval')
(*SMOOTHEND (*ON | OFF))
This keyword controls onset/loss of miscibility at low solvent saturations

(min_sol_sat) and low gas saturations (sgthresh). The discontinuity in fluid
properties this causes, by default, is smoothed to improve convergence.
*API-INT or *API-INTOW
These keywords signal IMEX to use the API tracking model with free gas
(*API-INT) or without free gas (*API-INTOW).
*PVTAPI
This keyword is used to input a gas PVT table (for use with the API tracking
option).
*APIGRAD
This keyword is used to input an oil PVT table for each oil density (gravity)
specified within a PVT region (for use with the API tracking option).
*GORINT
*GORINT is no longer used with the *MODEL *OILWATER option.
*GORINT continues to be used with the *MISCG model (miscible model
with chase gas injection).
Numerical Methods
*NORM *PBUB *MAXCHANGE *PBUB
The *NORM *PB and *MAXCHANGE *PB have been changed to *NORM
*PBUB and *MAXCHANGE PBUB for compatibility with CMG's
GRIDBUILDER. Please alter data to reflect this change.
Initialization
The Initialization section has changes which reflect the new API tracking option and the
*SWINIT option.
44 Introduction
User's Guide IMEX
*API *APIT
The API tracking option allows for two methods to specify the initial light oil
volume fraction in the reservoir. API can either be defined as a grid array or
as a function of depth for each PVT region.
*SWINIT
The *SWINIT option allows the user to specify a water saturation and still go
through the *BLOCK_CENTER gravity capillary equilibrium option.
*SWINIT can be different from the connate saturation. Equilibrium is
achieved by scaling the individual blocks Pcow curve to achieve equilibrium
at a specified water saturation.
*GOC_SW *WOC_SW
These keywords allow the user to explicitly define the water saturation above
the GOC and below the WOC for each PVT region. These values override
the normal defaults assigned when using the *BLOCK_CENTER gravity
capillary initialization options and the values assigned using *SWINIT.
*RANGE and *REFINE
The capability to define new refined grids in the Recurrent Data section has
been added to IMEX. Thus the keywords *RANGE and *REFINE may
appear in the well (recurrent) data. Please refer to the Reservoir Description
section for information on the usage of these cards.
*PERFRG
The ability to perforate in multilevel refined blocks has been added to the
*PERFRG keyword.
The *PERFRG card now can handle perfs which exist at in layers at different
levels of refinement. For example the following is now possible.
Fundamental
*PERFRG
Refine1
(5x5x1)
j1
Refine2
(3x3x1)
i
2
j
2
k
2
i1
k1
i2
j2
Well Index
k2
1.0
1.0
1.0
In addition, wells that are accidentally perforated in pinched out or nulled out
blocks are now flagged in a more informative manner. A list of valid perfs
(with i,j,k locations) is presented to the user, along with a separate list for
invalid perfs. It is now a simple process to produce a valid perf list.
User's Guide IMEX
Introduction 45
*WELLINIT
The *WELLINIT keyword can now also be defined on a well by well basis.
In addition to the previous usage,
*WELLINIT
(*ITER)
(*CHANGE)
(*TIMESTEP)
the format,
*WELLINIT
well_list
(*ITER)
(*CHANGE)
(*TIMESTEP)
can be used.
*ALTER, *TARGET, *GUIDEI, *GUIDEP, *GLIFT, *ON-TIME, *ALTERCP
These keywords listed allow the user a simpler method for assigning all
values in a well list to one value. Previous versions of IMEX allowed the
user to employ well lists with wildcarding, but values (rates, etc.) assigned to
each well had to be entered explicitly (even if one value is used for all wells).
Example
*ALTER 'PRODNW*'
14*1000.0
<----- Input List (14 entries)
Thus the number of wells in the wildcard list had to be known, making input
more time consuming.
If the user now specifies a single value in the list input IMEX will now set all
wells in the list to this value. This applies to all types of well lists (not just
those using wildcards).
*ALTER 'PRODNW*'
1000.0
<----- single entry, all wells

in list alter primary
constraint to 1000.0
It is not necessary to know the number of wells in the well list to apply the
single value to the list.
OLD OPTIONS NO LONGER SUPPORTED BY IMEX 98.00
*DIM
The IMEX 98.00 model no longer supports the *DIM *MEMORY_MODEL
keyword. The run time dimensioning now makes this backward
compatibility feature unnecessary.
46 Introduction
User's Guide IMEX

*DIM
The IMEX 97.00 black-oil model is run time dimensioned. That is, the
simulator allocates memory that is required at run time. The multiple copies
of simulators for the users specific RAM requirements are no longer
necessary. In addition special versions are now not required, as any number
of wells and blocks are handled.
In most cases, the action taken by run time dimensioning will be transparent
to the user and no extra dimensioning control will be required. However, in
the case of complex models with pinched out layers, the run time
dimensioner may require additional user guidance in order to minimize
storage allocated.
Due to this need we have introduced the *DIM keyword which enables the
user to override automatic settings and calculations and to explicitly specify
model dimensions.
*INTERRUPT, *WRST_COLD, *STOP_COLD
The interrupt keyword now has two extra options. The first allows the user
to write an unscheduled restart record and continue on with the run. The
second is only for PC workstations and immediately terminates the run.
*TNEXT
The *WRST keyword and the *WPRN (*WELL *GRID *SECTOR)
keywords all now allow the use of the *TNEXT keyword. The *TNEXT
keyword specifies that the simulator output action (e.g. write a restart) will
occur on the date (time) specified on the next *DATE (*TIME) card only.
*OUTPRN, *GRID, *SWC, *SORW, *SGC, *SORG
The critical/connate/residual saturations can now be respecified in the
Recurrent Data section. These variables have been added to the grid output
arrays and the SR2 output via the *OUTSRF *GRID card.
*OUTPRN, *RES, *POR
The porosity printout in the initial condition output no longer includes the
effects of net-to-gross ratio. The porosity printed out is the gross porosity.
*FORGAS
The *FORGAS card is no longer required to use the IMEX/FORGAS
coupled model. FORGAS or IMEX-only specification is made interactively
at the beginning of a FORGAS/IMEX run.
User's Guide IMEX
Introduction 47
*DEBUG, *XFLOW, *PARAMS

IMEX 97.00 incorporates a fully coupled wellbore crossflow model. THE
*DEBUG *XFLOW keyword enables the user to obtain detailed information
on how the crossflow model is operating. The *DEBUG *PARAMS
keyword enables the user to view the integers which determine the 'Virtual'
common block dimensions of the run time dimensioned IMEX.
The Grid Module within IMEX has been significantly improved this year. Of particular
interest to the users of refined grids is a large reduction in CPU time required to initialize
large refined regions. The initialization of large refined regions of blocks has been sped up
by as much as ten times.
*TRANLI, *TRANSLJ, *TRANSLK
IMEX 97.00 now allows the user to specify interblock transmissibility on all
faces (not just a single face in each direction). This feature allows the user to
more easily specify interblock flow between refined and unrefined regions.
Earlier versions of IMEX did not properly handle the use of aquifers
connected to specified regions. This has been corrected in IMEX 97.00.
A positive total hydrocarbon compressibility check has been added to the PVT table reader.
This check ensures physically meaningful results when gas comes out of solution.
Therefore Bg*dRs / dP - dBo/dP > 0.0 must be true, or in terms of table entries i and i+1:
Bg(i+1) * (Rs(i+1) - Rs(i)) / (P(i+1) - P(i)) > (Bo(i+1) - Bo(i)) / (P(i+1) - P(i))
A warning is printed if this check is violated. However the run will continue.
Rock-Fluid Properties
New checks on Swc, Sorw, Sgc and Sorg have been included to minimize possible user error.
*NOSWC
The *NOSWC option has been completely rewritten to more accurately
account for fluid relative permeabilities measured without connate water
present.
*KROIL, *SEGREGATED
A fourth three-phase oil relative permeability option has been included.
*RTYPE, *SWC, *SGC, *SORW, SORG
The rock type (*RTYPE) and connate/critical/residual saturations can now be
entered in the Recurrent Data section. This feature can be used to model
hysteresis effects.
48 Introduction
User's Guide IMEX
Numerical Methods
The AIMSOL solver has been improved to handle more efficiently options which employ
higher order factorizations (*SDEGREE 2). The CPU time penalty for using a more accurate
factorization has been substantially reduced.
*NORM *PB, *MAXCHANGE *PB
It is now possible to enter separate desired changes (*NORM) and maximum
changes (*MAXCHANGE) for bubble point pressure. If not input the
bubble point pressure change will default to the pressure change.
*AIM *STAB *ALL-BLOCKS, *AIM *STAB *AND-THRESH
Two new *AIM *STAB subkeywords have been added to allow the user
more control over the stability based explicit-implicit block switching
algorithm. Normally only neighbours of implicit blocks are checked for
stability (C.F.L.) violations.
The *ALL-BLOCKS option checks every explicit block. This requires a
significant amount of extra CPU time and may increase run times by as much
as ten percent.
The *AND-THRESH option adds nothing to the CPU time and in addition to
the stability checking performs threshold checking on all blocks. This
increases the rate at which explicit blocks turn implicit but also may improve
run timestepping behavior.
MULTILATERAL WELL OPTION
*FLOW-FROM, *FLOW-TO, *CLOSED, *REFLAYER
It is now possible to specify well trajectories for multilateral wells under the
*PERF or *PERFRG keywords. For each perforation, following the
location, relative permeability and status information, the optional keyword
*FLOW-TO for producers or *FLOW-FROM for injectors can be used to
assign "parent-child" relationships between perforation layers. This
information serves to define the well trajectory. In addition to open or
automatic status designation for a perforation, a third choice of closed is
introduced with the keyword *CLOSED. Unlike the two previous choices a
"perforation" with a closed status has no flow associated. Finally the
keyword *REFLAYER may be used to designate the reference layer for
bottomhole calculations. Please see the pages for *PERF and *PERFRG for
more information.
User's Guide IMEX
Introduction 49
WILDCARDS IN WELL LISTS

A significant new feature has been added to all keywords which acts upon well lists (e.g.
*ALTER 'well1' 'well2'). Wild cards have been incorporated in the following manner:
* replaces any number of characters at the end of a well name or can be used on its
own to represent all wells (e.g. *ALTER '*' or *ALTER 'wel*').
? replaces any single character anywhere in the well name (e.g. *ALTER '?ell1').
The two wild cards can be combined on any list and when wild cards are used the well list
generated is printed out for the user to check.
E.g. if the wells in the model are 'well1' 'well2' and 'prod1', then, '*' would list all 3, 'p*'
would list 'prod1'. '?ell?' would list 'well1' and 'well2'.
CROSSFLOW MODEL
*XFLOW-MODEL 'well list'
*SOURCE-SINK
*ZERO-FLOW
*FULLY-MIXED
A fully featured implicit crossflow model has been added into IMEX 97.00.
This model when used in *FULLY-MIXED mode allows the user to model
crossflow in a well. *ZERO-FLOW allows the user to shut in backflowing
layers (the default), while *SOURCE-SINK reverts the model to the IMEX
96.00 default which does not properly handle crossflow or backflow.
MISC OPTIONS
*HEAD-METHOD 'well list'
*GRAVITY
*ZERO-HEAD
In flowing wells it is appropriate to calculate the density of the fluids in the

wellbore based on mobilities of fluids in the well blocks. However for
pumping wells, except for a small amount of liquid collected at the bottom of
the well, the head in the well is effectively zero (the head of gas). In this
case we offer the user the *ZERO-HEAD option. The default *GRAVITY
option calculates wellhead as was previously done in IMEX 96.00.
*OPERATE *PENALTY
*WGR
*NODAMP
*DAMP
*PHASE2
The ability to use water-gas ratio as a penalty has been added along with
different methods for calculating the penalty. The *PHASE2 most accurately
honors the penalty rate but the *NODAMP and *DAMP methods may more
accurately mimic field operations.
50 Introduction
User's Guide IMEX
MOBILITY WEIGHTED WELL PRESSURES

IMEX will print out mobility weighted well pressure, mobility weighted datum pressure and
mobility weighted drawdown to aid in history matching. In order to obtain this report, the user
must request well output at reservoir conditions. See '*OUTPRN *WELL *RESERVOIR'.
The mobility weighted datum pressure is written to the SR2 file system, if detailed layer
output or block pressure is specified for output. See '*OUTSRF *WELL'
AUTODRILLING OF WELLS
In the 96.00 release the automatic opening of wells possessing the *AUTODRILL status was
triggered as an action taken when a group minimum oil or minimum gas constraint was
violated. This had the unfortunate effect of preempting the use, in these situations, of minimum
oil or gas monitors to terminate the simulation. In the 97.00 version, the opening of wells with
*AUTODRILL status is triggered when *AUTODRILL *ON is specified for a group or groups
using the *GAPPOR (group apportionment) keyword and a group listed fails to meet a target
during the target apportionment. The *AUTO-OPEN action for the *GCONM keyword is no
longer supported; this keyword structure must be converted by the user to the *GAPPOR form.
Please see the discussion of the *GAPPOR, *AUTODRILL, and *DRILLQ keywords for more
information.
GUIDE RATE APPORTIONMENT
In earlier releases of IMEX when guide rates were specified to determine target rate
apportionment, production targets other than oil rates were not properly handled. This has
been corrected in version 97.00; guide rate apportionment is done correctly for all allowed
target rates.
TRANSFER OF CUMULATIVES IN HIERARCHY REDEFINITION
In earlier releases if a group structure was introduced after a model had been run for some
time with no group structure defined, the FIELD cumulative was reset to zero when the
FIELD became part of the group hierarchy. In version 97.00, when the group structure is
redefined, the newly defined top-level group acquires the cumulative production of the old
top-level group.
IDENTIFICATION OF REFERENCE LAYER AFTER REPERFORATION
In earlier releases when a multilayer was reperforated in such a way that some layers were
reperforated, some eliminated, and some new layers were added, the reference layer (the one
in which the well's bottomhole pressure is defined) was chosen as the layer first perforated
historically, regardless of its position in the current set of perforated layers. For example, if a
well were first perforated in layers 1, 3, and 4, in that order, and later reperforated in layers 2,
3, and 4, earlier releases would pick layer 3 as the reference layer. Version 97.00 always
picks the first layer perforated in the most recent set of perforation lines as the reference
layer; in the example, version 97.00 would pick layer 2 as the reference layer. Please see the
discussion of the *PERF keyword for more information.
User's Guide IMEX
Introduction 51

*MONITOR *BACKFLOW *AUTOLAYER for injectors is no longer valid. Use the default
*XFLOW-MODEL *ZERO-FLOW to accomplish this function.
In addition, we recommend that *MONITOR *BACKFLOW *AUTOLAYER not be used for
producers, instead the XFLOW-MODEL should be used.
The *AUTO-OPEN action for the *MINOIL and *MINGAS monitors under the *GCONM
keyword is no longer supported; this keyword structure must be converted by the user to the
*GAPPOR form. Please see the discussion of the *GAPPOR, *AUTODRILL, and
*DRILLQ keywords for more information.
52 Introduction
User's Guide IMEX

IMEX 96.00 has several enhancements that make the input output controls more flexible and
user friendly. Display of the screen diary file can be changed to include additional well
information. IMEX can now be run with the name of the data set specified through a
command line argument (i.e. mx064.exe -f data1.dat). The root names of the output file,
input .irf and .mrf files will automatically be the root name of the data file.
*OUTSRF, *WELL, *DOWNHOLE, *LAYER, *BLOCKP
Well layer records are no longer written to the SR2 files *WELL at surface
and reservoir conditions. In order to reduce the size of .mrf files, only well
records at surface conditions are written. The user can still write layer
information at either reservoir and/or surface conditions using the keywords
shown here.
*XDR
The binary (data) file may be written in external data representation (XDR)
format as well as the binary format native to your platform. Use of XDR
allows the SR2 binary file(s) to be written on one platform and read on
another. For example, the SR2 files can be generated on a UNIX workstation
server and then accessed with RESULTS or the Report Writer on a PC. If
the SR2 is in XDR format, then the keyword "XDR" will appear near the top
of the index file (IRF). XDR is a nonstandard option, and the source code is
shipped with XDR disabled. Most executables shipped by CMG will have it
enabled.
*OUTDIARY, *WELLSTATUS
These keywords specify that the number of producing wells, injecting wells,
and shut in wells and the maximum changes of saturations and pressure will
be written to the output-diary-file. The oil/gas/water production/injection
rates will be written with up to 6 significant figures. The water-cut will be
written with up to 3 significant figures.
*INTERRUPT, *RESTART-STOP, *STOP, *INTERACTIVE
These keywords indicate that when the user interrupts the simulation run by
simultaneously pressing the "CTRL-C" keys or by typing "kill -2" followed
by the process identification number (PID), one of the following user
specified actions can take place:
User's Guide IMEX
1.
Continue the simulation run
2.
Finish the current timestep, write a restart record and stop the
simulation.
3.
Flush buffers, close files and stop the simulation run immediately.
Introduction 53
Use of this keyword prevents corruption of the .irf and .mrf files when
aborting a simulation run and optionally writes a restart record before
stopping.
*PVCUTOFF
This option allows the user to adjust the pore volume cut off value and
controls the level at which a block will be set null due to a small pore
volume. This option ensures that blocks with small pore volumes can be
systematically removed from the simulation. Such small pore volume blocks
can hinder convergence and should not remain in a simulation.
*NETGROSS
This option allows the input of net-to-gross ratios which are used to modify
the porosities and permeabilities in the I and J directions. The net-to-gross
ratios are used as multiplication modifiers.
A number of options for smoothing both user input relative permeability and capillary
pressure data. In addition, these data can also be generated using standard power law
correlations.
*SMOOTHEND, *QUAD, *CUBIC, *POWERQ, *POWERC
This optional keyword indicates the type of smoothing that is to be used at
the end points for the interval between kr = 0 and the first nonzero value of kr
for both krw and krow (krg and krog). The keyword may be followed by
subkeywords that specify the type of smoothing to use.
For the subkeyword *QUAD, a quadratic smoothing is used. For *CUBIC, a
cubic smoothing is used. If no smoothing has been specified, then linear
interpolation is used. For *POWERQ and *POWERC end point smoothing
is done using power law if the exponent is between 1.5 and 4.0. Otherwise,
for *POWERQ, quadratic is used and for *POWERC, cubic is used for end
point smoothing.
Please see the *SWT or *SGT (*SLT) entry in this manual for more details.
*SMOOTH, *ALL, *ALLINT, *ALLIN, *ALLPL
This option smoothes the entire relative permeability table and generates a
new table consisting of smoothed, equally spaced values. One or all
saturation dependent properties in the table may be smoothed based on user
controlled specifications. The data may be smoothed using a power law
correlation, linear interpolation or a combination of the two. Please see the
*SWT or *SGT (*SLT) entry in this manual for more details.
54 Introduction
User's Guide IMEX

*RELAX, *ACC, *GAS
The *RELAX *GAS enables the gas-saturation under-relaxation option.
This option can lead to faster convergence for simulations where the overall
pressure in the reservoir is declining and gas is being liberated due to the
pressure falling below the bubble point. This option prevents the gas from
disappearing in a grid block for three Newtonian iterations. If IMEX predicts
that gas in a grid block will disappear for more than three iterations, only
then will the gas be allowed to disappear.
The *RELAX *ACC option performs over-relaxation of the accumulation
term. This normally improves model performance.
Well Management
Well management has several improvements and bug fixes. Some new options have also
been added.
*ON-TIME
This new option allows the input of fractions for all wells or selected wells
which specify the fraction of the simulated time during which the well is online and actually in operation. If a well is assigned an on-time fraction less
than one, the rates reported (the "averaged" rates) are less than the rates
which apply when the well is actually in operation (the "instantaneous"
rates). The user should enter well constraints (such as BHP, WHP and rate
constraints) which correspond to INSTANTANEOUS rates; reported BHP
and WHP values will correspond to the instantaneous rates. Reported rates
and cumulative volumes will, however, be the averaged values. If the *ONTIME keyword does not appear in the data set, the default value of one is
applied for the on-time fraction and the simulation is unaffected by the
introduction of the option.
*MONITOR, *MIN or *MAX, *WHP or *BHP
It is now possible to enter maximum BHP and WHP values as monitored
constraints for producing wells; please see the entry for the *MONITOR
keyword in this manual for more details.
It is also now possible to specify maximum lift gas rates for wells under the
*GLIFT *OPT option; please see the *GLIFT entry for details.
When entering wellbore hydraulics tables under the *PTUBE keyword, it is
now possible to direct the simulator to supply missing values by interpolation
or extrapolation from other values in the same row of the table by entering
*EXTP in the place which the missing number would normally occupy;
please see the *PTUBE entry in this manual for more details.
User's Guide IMEX
Introduction 55
*MINOIL, *MINGAS, *AUTO-OPEN

An *AUTO-OPEN action has been added to the *MINGAS and *MINOIL
constraints under the *GCONM group monitoring keyword. This action
allows wells in a particular group which have *AUTODRILL status to be
opened automatically when the group's gas or oil production rate falls below
the specified value. Please see the *GCONM entry in this manual for more
details.
*TURB, *QUAD
This option is used for calculating the drawdown pressure for high gas rate
wells. These keywords specify the use of quadratic inflow performance and
a rate dependent turbulent skin factor. Please see the *PERF entry in manual
for more details.
*WELLINIT, *CHANGE, *ITER
This option allows the user to specify whether the bottomhole pressure value
for all wells running on a rate constraint, drawdown constraint or an
implicitly imposed wellhead pressure constraint should be reinitialized 1)
only after well changes or 2) after each Newtonian iteration. This keyword
also determines whether wells operating under a *WHP *IMPLICIT
constraint are checked for shutting in at the Newtonian or timestep level.
Use of the *ITER option, especially for high permeability cases, usually
results in rapid convergence of the well equations by reducing the total
number of Newtonian iterations. For cases with low permeability and/or
high compressibility, this option may reduce the total number of Newtonian
iterations and reduce the CPU time.
The *UNDER-RELAX *GAS option is no longer supported by IMEX 95.00. Use the
*RELAX *GAS option instead.
56 Introduction
User's Guide IMEX

IMEX 95.00 now allows corner point input. Several different methods for entering corner
point data are allowed.
*GRID, *CORNER
This option allows the entry of corner point grids. Corner point grids are
made up of blocks defined by their eight corner points. Each corner point is
described by giving its x, y and z coordinate values, which gives its location
in the reservoir.
*COORD
This options causes the reading of information defining the "x"- and "y"coordinate locations for all corner points defining a *CORNER grid. Since
the corner points must lie on vertical lines, there being exactly (ni+1) * (nj+1)
such lines, and since definition of a line requires the specification of two
points, each requiring the specification of three coordinates, *COORD
expects to read 2 * 3 * (ni+1) * (nj+1) values.
*XCORN
This option will cause the reading of all the (ni+1) * (nj+1) * (nk+1) "x"coordinate values required to define the lateral locations of all points defining
a *CORNER grid.
*YCORN
This option will cause the reading of all the (ni+1) * (nj+1) * (nk+1) "y"coordinate values required to define the lateral locations of all points defining
a *CORNER grid.
*ZCORN
This option causes the reading of all depths ("z"-coordinates) of all the corner
points required to define the grid.
*CORNERS
This options causes the processing of 3*(8*ni*nj*nk) values, with the first
group of 8*ni*nj*nk values giving all the "x"-coordinates of all corner
points, the second group giving all the "y"-coordinates, and the third group
giving all the "z"-coordinates.
User's Guide IMEX
Introduction 57
*KROIL
This option allows the specification of different methods for the evaluation
and determination of Kro, the 3-phase oil relative permeability. These
methods are Stone's first model; Stone's second model as modified by Aziz
and Settari; and the linear isoperm method.
*STONE1
This specifies the use of Stone's first model for computing Kro. The Krow and
Krog values may be looked up as functions of either Sw and Sg or as functions
of 1-So and 1-Swc-So.
*STONE2
This specifies the use of Stone's second model for computing Kro. The Krow
and Krog values may be looked up as functions of either Sw and Sg or as
functions of 1-So and 1-Swc-So.
*LINEAR_ISOPERM
This specifies the use of the LINEAR ISOPERM model as proposed by L.E.
Baker. The method allows the specification of straight-line segments along
which Kro assumes a constant value.
Well Management
*GLIFT, *RATE, *GLR, *OPT
Gas lift rates can now be specified in three ways:
1.
Input lift gas rates directly using the *GLIFT *OPT *RATE keywords.
2.
Input a gas-liquid producing ratio using the *GLIFT *GLR keywords.

The lift gas injection rates will be calculated from the current timestep's production data. However, the lift gas rates will not be used for
the calculation of production rates until the next timestep.
Lift_gas_inj = GLR_input * Total_liquid_est - GOR_est * Oil_rate_est
3.
Gas injection rates will be calculated automatically using the values

specified in the *GLOPT keyword.
These rates can be modified at specified dates and times in the Well and
Recurrent Data section.
58 Introduction
User's Guide IMEX
*GLCONTROL
This option allows the specification of gas lift constraints for a given set of
production wells. The specified operating constraints are checked at every
timestep. If one of the constraint is violated, then the well or group of
production wells will be put on gas lift. The operating constraints that can be
specified are: wellhead pressure, bottomhole pressure or stock tank oil rate
and maximum water-cut. A hydraulics table can also be specified, which is
used for wellbore hydraulics calculations when the well or wells are switched
to gas lift. The operating constraints can be changed at specified times or
dates in the Well and Recurrent Data section.
*GLOPT, *MAXGAS, *RECYCLE, *GLOPT-TIME
This option will automatically optimize gas lift injection rates to provide
maximum oil production rates for the specified wells. The total lift gas
available can either be input directly using the *MAXGAS keyword or
indirectly from the field's gas production rate through the *RECYCLE
keyword. In either case, a portion of the available lift gas can be specified
for compressor consumption. This consumption rate is proportional to the
total lift gas injected. An iterative procedure is used to determine the
optimum lift gas allocations to each well. A minimum gas-cost factor, below
which gas lift injection uneconomic, can also be specified. This can be used
to ensure that large volumes of lift gas are not allocated for small incremental
oil volumes. The values specified with this option can be modified at dates
and times specified in the Well and Recurrent Data section.
The *WELLDPS option and the *DP_PSEUDO option are not supported by IMEX 95.00.
Fluid Properties
*DIFLIB
This option allows for the input of differential-liberation tables. These tables
are converted by IMEX to PVT tables (corrected for separator conditions).
*LIGHTOIL
Use a black-oil model, modelling flow of oil, water, and gas. Separate Bo, Rs
and two parameters alpha and beta are used to convert stock tank rates to
reservoir rates. Alpha is the gas shrinkage factor and beta is the liquid drop
out from the gas at stock-tank conditions. This option is useful for those
fluids which behave like an oil at reservoir conditions, but are so light that at
surface conditions liquids drop out of the gas. This option uses the method
outlined in Rovere, A., Sabathier, J.C. and Codreanu, D.B., "Modification of
a Black-Oil Model for Simulation of Volatile Oil Reservoirs".
User's Guide IMEX
Introduction 59
*PVTLO
This keyword indicates PVT input for the light oil option.
Initial Conditions
The keywords for initial conditions specifications have been completely re-written. However
IMEX will accept all the old keywords as before. A depth averaged (*DEPTH_AVE)
initialization option has also been added. A description of all the initialization options can be
found in Section 8 under the heading Notes on Initial Conditions.
Well Management
The group-well structure can now be changed at any time. Wells can now be moved from
one group to another.
Dimensioner
The IMEX 94.00 dimensioner has been improved to dimension the common block tightly.
60 Introduction
User's Guide IMEX
Important Changes Between IMEX 93.00 and 92.00

WELL GROUP CONTROL MODULE ADDED TO IMEX 93.00
IMEX 93.00 has the ability to specify group control for wells in a fairly comprehensive
manner. The following features are now available in IMEX.
1. Up to three group levels can be specified.
2. Groups can have target rates. GOR and water-cuts can be monitored and
appropriate actions can be taken.
3. Wells can be opened (drilled) automatically to meet target constraints. This
drilling can be done automatically based on instantaneous production/injection
potentials or can be done in a user specified sequence.
4. Well layers can be shut in and opened automatically depending on the current
GOR and water-cut.
5. Wells can be shut in and opened automatically depending on the current GOR and
water-cut.
6. In previous versions of IMEX, voidage replacement was based on surface rates of
producing wells. It is now based on production at reservoir conditions and thus
gives much more accurate results.
7. In addition to surface rates, bottomhole rates can now be specified as well
constraints.
8. *GLIFT -- This keyword allows for the specification of lift gas rates for producing
wells.
LIMITATIONS OF WELL GROUP CONTROL
This version of well management has the following limitations:
1. A maximum of three group levels are allowed.
2. The top most group level ('FIELD') cannot have wells attached to it. Only other
groups.
3. Groups to which wells are attached cannot have other groups attached to it.
4. New wells may be attached to a group at any time, however a well cannot be
attached to more than one group. Furthermore a well cannot be moved from one
group to another.
5. Group controlled injection fluids are limited to gas, water or solvent only. Oil
injection cannot be controlled by groups.
6. The well management module can automatically shut in and reopen well layers
when GOR or WCUT exceed a certain limit. When this option is used the layers
are sorted according to their depths, in order to open/close layers. Thus if the well
is perforated horizontally, the behavior of this option may be unpredictable.
User's Guide IMEX
Introduction 61
DATA INCOMPATIBILITIES BETWEEN 93.00 AND PREVIOUS VERSIONS OF

IMEX
Although most of the well data are compatible with previous versions of IMEX, there are
some incompatibilities.
1. For the pseudo-miscible option (solvent injection) the previous versions of IMEX
monitored the gas+solvent-oil ratio when *MONITOR *GOR was used. In this
version of IMEX *MONITOR *GOR refers to gas-oil ratio only. If it is desired to
monitor the gas+solvent-oil ratio the *MONITOR *GSOR should be used.
2. Minimum rate operating constraints have been replaced by minimum rate
monitoring constraints. If a minimum rate well constraint is specified, it is
converted automatically to a monitoring constraint and a warning message is
printed.
3. Well names now have to be unique i.e. it is no longer possible to have two or more
wells with the same name. This is because wells can now be referred to by their
names only. Well numbers are optional.
4. *PERFV can no longer be used to perforate wells located in hybrid grids. The
*PERFVHY or *PERFRG keywords should be used to perforate wells located in
hybrid grid blocks.
5. The *GATHER keyword and the related keywords *PTARG and *ITARG are now
obsolete and cannot be used. The *GROUP and *GCONP, *GCONI and
*GCONM keywords must be used instead.
6. Subkeywords *SRFIN, *SRFOUT, *RESTARTIN and *RESTARTOUT under the
*FILENAMES keyword are now obsolete and will not be recognized. They have
been replaced with the SR2 file system keywords: *INDEX-OUT, *MAINRESULTS-OUT, *REWIND-OUT, *INDEX-IN, *MAIN-RESULTS-IN and
*REWIND-IN.
Water-gas ratio is now printed for gas-water reservoirs instead of water-cut.
*MODSI
This option allows for the input/output of pressure in Kg/cm.
*SECTOR
The *SECTOR subkeyword in the *WPRN and *WSRF keywords now
controls the sector and lease line printing frequency to the output and SR2
files.
*OUTDIARY
Printing to the diary-file (normally the screen) can now be controlled by the
use of *OUTDIARY keyword.
62 Introduction
User's Guide IMEX
SECTOR OUPUT
The sector output has been revised and considerably enhanced. This option now prints a
table of sectors with aquifer influx, and production/injection from each sector, in addition to
all the previous quantities.
*AQLEAK
In previous versions of IMEX, the water OUTFLOW from the reservoir into the
aquifer was not modelled when the *AQUIFER (analytical aquifer model) was
used. If the reservoir pressure climbs above the aquifer pressure, this
implementation can give incorrect results. Thus an option (*AQLEAK) was
added to model water outflow into the aquifer when the reservoir pressure rises.
Fluid Properties
*COT
This keyword allows the user to specify oil compressibility as a function of
bubble point pressure and block pressure above the bubble point pressure.
Initial Conditions
*GASCAP
In previous versions of IMEX, when *VERTICAL *ON option is used, the
gas-cap is initialized not to have any oil EVEN if the immobile liquid
saturation from the gas-liquid relative-permeability table is GREATER than
the connate water saturation from the water-oil relative-permeability table.
An option (*GASCAP *OIL) is now available to initialize the gas-cap with
residual oil.
The SR2 File System
SR2 FILE SYSTEM
The output for graphics and restarts is now handled in a completely different way. The
graphics and restart information is now output to a set of up to three files.
The new file system called the SR2 file system combines the graphics and restart information
for greater space efficiency and smaller output files during run time.
The SR2 file system consists of two or three files. These are the index-results-file (IRF), the
main-results-file (MRF) and the rewindable-results-file (RRF). The RRF file exists only if
the simulator is directed to rewind the restart record. In such a situation, only the IRF and the
MRF files will exist.
The two or three files belonging to the SR2 file system work together. Thus an IRF will be
useless without a MRF and vice-versa. The IRF is an ASCII file which is an index to the
MRF and the RRF i.e. the IRF is a user readable ASCII file which details the contents of the
binary MRF and RRF files. The MRF and RRF files are both binary files and hence cannot
be read by the user. However a summary of the contents of the MRF and RRF files can be
obtained by examining the IRF.
User's Guide IMEX
Introduction 63
SRF FILE
Since the SR2 file system is more flexible and space efficient, the specifications of the SR2
are completely different to those of the SRF. For those user's who rely on the SRF for their
graphics and other output processing, a SR2 to SRF converter is provided with this release.
Using this converter any set of SR2 files can be converted to a SRF file.
STRUCTURE OF RESTART RECORD
Due to the new SR2 file system, the writing of the restart record has been completely changed.
Thus it is not possible to use restart records generated by previous versions of IMEX.
64 Introduction
User's Guide IMEX
Introduction to IMEX
IMEX is a three-phase black-oil simulator with gravity and capillary terms. Grid systems
may be Cartesian, cylindrical, or variable depth/variable thickness. Two dimensional and
three dimensional configurations are possible with any of these grid systems. Gas phase
appearance/disappearance is handled by variable substitution.
Some of the novel features of IMEX are:
Adaptive Implicit Formulation
IMEX can be run in explicit, fully implicit and adaptive implicit modes. In many cases only
a small number of grid blocks need to be solved fully implicitly; most blocks can be solved
by the explicit method. The adaptive implicit option accomplishes this and is useful for
coning problems where high flow rates occur near the wellbore, or in stratified reservoirs
with very thin layers.
By using the adaptive implicit option, a savings of one third to one half of the execution time
may occur because timesteps are as large as those obtained using the fully implicit method.
Options are available so that a user can select a fixed pattern of fully implicit blocks. IMEX
also selects these blocks dynamically, based on user specified thresholds or on matrix
switching criteria.
Dual Porosity/Dual Permeability
The dual porosity option allows the discretization of matrix blocks in either a nested format
called the multiple-interacting continua (MINC) approach or a layered format called the
subdomain method. The dual porosity approach idealizes the fractured reservoir as consisting
of two parts: the primary porosity and the secondary porosity.
The primary porosity (the matrix) represents the small intergranular pores in the rock matrix.
The secondary porosity (the fractures) consists of fractures, joints, and/or vugs. The dual
porosity approach is characterized by representing one reservoir volume with two continua.
The fractures, having small storativities, are the primary conduits of fluid flow, whereas the
rock matrices have low fluid conductivities but larger storativities.
A simple dual porosity model using a one matrix/one fracture system may be specified.
Shape factors based on the work of Warren and Root, or Gilman and Kazemi may be used.
In this case the matrix-fracture transfer is assumed to be under semi-steady state flow.
To achieve adequate representation of the matrix-fracture transfer mechanism, it may be
necessary to subdivide the matrix volume into smaller volume elements. The MINC
approach achieves this by subdividing the matrix into a series of nested volume elements.
The discretization is based on the assumptions that the elemental surfaces are equipotential
surfaces, and that the variation of the primary variables across one fracture block is small.
The MINC model is a good representation of transient matrix-fracture flow. Gravity effects,
however, are not handled properly which may become a serious disadvantage when the matrix
block height is large. A model which improves the representation of gravity effects and fluid
phase segregation is the subdomain method. In this model, fluid and pressure distribution in the
matrix are calculated, but gravity segregation is assumed to exist in the fracture.
User's Guide IMEX
Introduction 65
Besides the MINC model and the subdomain model, the dual permeability model allows fluid
communication between the matrices. This option is useful when fracture connectivities are
oriented in one preferred direction, or when matrix-matrix connectivity is important, such as
in gas-oil gravity drainage processes.
Pseudo-Miscible Option
A pseudo-miscible option models first-contact and multiple-contact miscible processes. It
can model solvent in the water phase and also chase gas.
Polymer Option
A polymer option models rock adsorption, polymer inaccessible pore volume, resistance
factors, viscosity mixing, and dispersion.
API Tracking
An API tracking option models reservoirs with compositional gradients. Oil is divided into
two components. Oil properties are functions of composition as well as pressure and bubble
point pressure.
Faulted Reservoir Option
A fault option models stratified reservoirs with one or more faults. This option accurately
accounts for flows between offset layers and enables simulation to be performed without
unrealistic averaging of properties. Sloping faults which are nonparallel to the co-ordinate
axes are also modelled.
Fully Implicit Wells
Wells are solved in a very robust fashion. The bottomhole pressure and the block variables
for the blocks where the well is completed are solved fully implicitly. If a well is completed
in more than one layer, its bottomhole pressure is solved in a fully coupled manner, i.e. all
completions are accounted for. This eliminates convergence problems for wells with multiple
completions in highly stratified reservoirs. Also, a comprehensive well control facility is
available. An extensive list of constraints (maximum, minimum bottomhole or wellhead
pressures, rates, GOR, etc.) can be entered. As a constraint is violated, a new constraint can
be selected according to the user's specifications.
Matrix Solution Method
IMEX uses a state-of-the-art solution routine based on incomplete Gaussian elimination as a
preconditioning step to GMRES acceleration. AIMSOL has been developed especially for
adaptive implicit Jacobian matrices. For more information see the AIMSOL Technical Manual.
For most applications the defaults control values selected by IMEX will enable AIMSOL to
perform efficiently. Thus, IMEX users do not require detailed knowledge of matrix solution
methods.
66 Introduction
User's Guide IMEX
Local Grid
A facility for local grid refinement is included in IMEX. The user specifies a region of the
reservoir which is to be subdivided. All interblock connections and transmissibilities are
automatically calculated by IMEX. This option can be used to study near well effects in
field-scale simulation and eliminates the need for well pseudo-functions. Static fractures can
also be efficiently modelled with this technique. All extra terms are correctly handled by the
matrix solution routine.
Pinched Out Layers
Pinch outs are efficiently modelled by using the pinch out option. This option removes
pinched out blocks from the active block list, and automatically couples together layers above
and below the pinch out.
Reservoir Initialization
Initial reservoir conditions can be established with given gas-oil and oil-water contacts.
Flexible Grid System
Several grid options are available with IMEX: Cartesian coordinates, cylindrical coordinates,
variable thickness/variable depth and corner point grids. Two dimensional and three
dimensional systems are possible with any one of these options.
Variable Bubble Point
A rigorous variable bubble point formulation using variable substitution is used.
Undersaturated regions, and mixing of oils with different PVT data are properly modelled.
Aquifer Models
Aquifers are modelled by either adding boundary cells which contain only water or by the use
of the analytic aquifer model proposed by either Carter and Tracy or Fetkovitch. The first
method is useful in the situation where the aquifer dimensions and location are well known
and its inclusion in the reservoir can be achieved by a relatively small number of additional
blocks. The later methods are more useful for large to infinite aquifers where an approximate
calculation of water influx into the reservoir is desired, but their representation through the
addition of boundary reservoir blocks is not feasible.
Input/Output Units
SI, field, laboratory or modified SI units can be specified.
Portability
IMEX is written in standard FORTRAN 77 and has been run on various hardware platforms.
These include: IBM main frames, CDC, CRAY, Honeywell, DEC, Prime, Gould, Unisys,
Apollo, SUN, HP, FPS and IBM-PC compatible 386 and 486, Pentium, Pentium Pro, and
Pentium II computers.
User's Guide IMEX
Introduction 67
Graphics
CMG's graphics system RESULTS, uses the SR2 file system for post-processing of
simulation output.
RESULTS can also be used for input data preparation and grid design.
Run Time Dimensioning
IMEX is written to automatically dimension itself to your problem size. The input file is prescanned so that the model dimensioning parameters can be found. IMEX then assigns storage
based on the scanned model parameters.
Crossflow in the Well
IMEX can use an advanced fully mixed crossflow model to describe well crossflow/backflow
phenomena.
Condensate Modelling
IMEX can be used to model gas reservoirs with condensate. Oil is permitted to exist in the
gas phase in the *GASWATER_WITH_CONDENSATE option (the gas phase must always
exist and the oil phase must initially not be present). The *VOLATILE_OIL model can
model the most general case allowing both saturated and undersaturated oil and gas. Any of
the oil, gas, and water phases may initially be present in the reservoir.
Parallellized Jacobian Building and Matrix Solver
The Jacobian Building and Solver have been rewritten to allow IMEX to run in parallel on
some platforms. Please refer to the Numerical Method Control section of the user manual for
more information.
Wellbore Friction Model
A wellbore friction model has been introduced within IMEX. This includes several powerful
frictional pressure drop correlations and is robust enough to handle complex scenarios (such
as multilateral wells).
Seawater Injection/Scale Buildup Model
The tracking of seawater injection within the reservoir has been added and coupled to a scale
deposition/well productivity damage model. This allows IMEX to model the affects of
seawater and formation water mixing when calculating scale deposition at producing wells.
68 Introduction
User's Guide IMEX
Tutorial
Introduction
The Tutorial section is a guide for the novice user of the keyword input system and does not
replace the User's Guide. Only specific keywords and topics are discussed in this Tutorial
section. The user manual contains a detailed description of each keyword, while the Tutorial
section tackles "how-to" questions you may have when building a data set.
IMEX uses the data set that you create initially and then creates three or four other files.
Each IMEX run creates an output-file (OUT), an index-results-file (IRF) and a main-resultsfile (MRF). In addition a rewindable-results-file (RRF) may or may not be created depending
on the options selected by the user.
.OUT (output)
.IRF
DATA SET
IMEX
(index)
.MRF (main)
.RRF
(rewindable optional)
If a restart run is desired, then several existing files are needed and another three are
generated. This is illustrated in the diagram:
DATA SET
.OUT (output)
INPUT.IRF
.IRF
(index)
IMEX
INPUT.MRF
.MRF (main)
INPUT.RRF
(optional)
.RRF (rewindable optional)
User's Guide IMEX
Tutorial 69
Data Groups in the Keyword Input System

There are several points to remember when you build a data set using the keyword input
system:
a) There are seven different data groups in the keyword input system.
b) The groups must follow a certain input order:
Rock-Fluid Data
Initial Conditions
c) The keywords belonging to each group cannot appear in other groups, unless it is
specifically written. Usually, this happens with recurrent data from other sections
which may be changed in the Well and Recurrent Data section.
d) Also pay attention to the order that some keywords, within a group, are entered.
70 Tutorial
User's Guide IMEX
How to Document Your Data Set

Documenting your data set is done with the keywords:
a) *TITLE1
b) *TITLE2
c) *TITLE3, and
d) *CASEID
They must be located in the Input/Output Control section.
These keywords are optional and may be excluded from the data set; however they are very
useful for documenting files and distinguishing similar data sets from each other. At least
one title should be used. All titles and the case identification must be enclosed within single
quotes.
*TITLE1 and *CASEID are both used in the SR2 file system, which is used to create
graphics of the simulation. *TITLE1 may be as long as 40 characters, but both *TITLE2 and
*TITLE3 are allowed up to 80 characters each. The case identification is limited to 8
characters.
You may also use two keyword indicators or '**' to insert comments throughout your data set.
Comments may appear anywhere in your data set.
Example:
*TITLE1
'Simulation Run #1 - 1989-01-23'
*TITLE2
'Dual Porosity Problem using the MINC option'
*TITLE3
'This is a 12 x 12 x 10 Cartesian grid system'
*CASEID 'RUN1'
** You may add additional information here or
** anywhere if the title lines did not allow
** enough room for documenting this data set.
** You may also use the comments to describe
** your data as you enter it.
User's Guide IMEX
Tutorial 71
How to Do a Restart
WHAT ARE RESTART FILES?
Restart files are a set of SR2 files. These include at least an index-results-file (IRF) and a
main-results-file (MRF). The set may or may not include a rewindable-restart-file depending
on the options used by the user. Restart information (usually called a restart record) about the
simulation variables is written to these files at the frequency chosen by the user. Restarts can
be done from only those timesteps at which a restart record is written. Writing a restart
record is optional.
WHY WOULD YOU NEED TO DO RESTART?
You may want to do restarts for the following reasons:
a) To do sensitivity studies or history matching,
b) To change well specifications,
c) To perform a short simulation run to see if the results are satisfactory, before
running bigger, longer jobs, and
d) To save execution time in subsequent runs. For instance, you have completed a
simulation run and the preliminary results look good. Now you want to do
prediction runs.
Because you have created restart records with the initial run, you may select a timestep from
the middle of your run and 'restart' the simulation. The simulator does not need to start at the
beginning; it continues execution from the timestep you have chosen.
HOW TO DO A RESTART
Restart records are optional and do not have to be written. But if you do plan on doing
restarts, you need to create one with your initial run.
Use *WRST and *RESTART to create your restart file. They must be located in the
Input/Output Control section of your data set. *WRST may, however, appear in the Well and
Recurrent Data section when well changes are done.
*WRST indicates the frequency of writing to the restart record. *RESTART indicates that
the current simulation run is a restart run. If you wish to start at the last timestep, then leave
the rest of the line after *RESTART empty. This is the default. Otherwise, enter a timestep
number.
Example:
*RESTART 30
*WRST
10
To do a restart run:
a) Do not change any of the original reservoir data, but do add *RESTART to the
Input/Output Control section of your data set.
b) Increase the maximum number of timesteps, if necessary, or leave out
*MAXSTEPS altogether.
72 Tutorial
User's Guide IMEX
c) You will need input files (which were created during the initial run), including an
input IRF file, an input MRF file and possibly an input RRF file.
d) You will also need names for the new set of output files. Be sure that all the input
files were created from the same simulation run.
User's Guide IMEX
Tutorial 73
Controlling Contents of the Output File

To control the contents of the output file, use:
a) *WPRN, and
b) *OUTPRN
These keywords may appear in the Input/Output Control section or the parameters may be
changed later on in the data set in the Well and Recurrent Data section.
*WPRN indicates how often to write grid block data, sector & lease plane data, well data, and
numerical method control data, such as Newton iterations and timestep convergence
behavior.
If no grid, sector or well information is desired in the output file, then the frequency may be
set to zero.
Example:
*WPRN *WELL
0
*WPRN *GRID
0
*WPRN *SECTOR 0
If either of these is left out of the data set, then the default is to print the *GRID and
*SECTOR information at every *TIME or *DATE keyword in the Well and Recurrent Data
section. For *WELL the default is to print information at every timestep. These defaults can
produce a very large output file, which can fill up the available disk space on your computer
very quickly.
*OUTPRN limits what well data, grid data, reservoir data, and how many property tables are
printed. You may actually list the grid data types that you want.
Well data is treated differently. You may print out everything possible or print a well
summary only. To print out information per layer for all wells use *OUTPRN *WELL
*LAYER. To print out production at reservoir conditions use *OUTPRN *WELL
*RESERVOIR. To print out a one line summary for each well use *OUTPRN *WELL
*BRIEF. The latter is the default.
The use of the *TNEXT subkeyword of *WPRN is an efficient method of specifying output
on a specific date. *TNEXT turns on printing at the next date card, then resets itself to not
print after that date.
74 Tutorial
User's Guide IMEX
Controlling Contents of the Graphics File (SR2)

To control the contents of the SR2 file system, use:
a) *WSRF, and
b) *OUTSRF
These keywords may appear in the Input/Output Control section or the parameters may be
changed later in the Well and Recurrent Data section of the input data file.
*WSRF indicates how often to write grid block, sector and well data. This information may
be written at a certain frequency.
If no grid or well information is desired in the SR2 files, then the frequency is set to zero.
This may be used to cut down the size of a very large file. You may, however, change this in
subsequent well changes.
If *WSRF is left out of the data set, then the default action taken is to print well information,
at every timestep. *GRID and *SECTOR are printed when *TIME is found in the Well and
Recurrent Data section. The amount of information can be controlled by the *OUTSRF
keyword.
The keyword combination *WSRF *GRID 0 ensures that grid information is never written to
the simulation results file. This can be changed in the well data at subsequent well changes.
*OUTSRF limits what grid data are printed. You may also ask to have special variables
printed at given grid block locations. Separate lists of variables are available for grid
information and well information.
The use of the *TNEXT subkeyword of *WSRF is an efficient method of specifying output
on a specific date. *TNEXT turns on SR2 writing at the next date card, then resets itself to
not print after that date.
User's Guide IMEX
Tutorial 75
Describing Your Grid System

To describe your grid system, you need:
a) *GRID
b) *KDIR
c) *DI
d) *DJ
e) *DK
Use either
f) *DEPTH
or
g) *PAYDEPTH
or
h) *DTOP, and
i)
*DIP
Among these keywords, only *KDIR and *DIP are totally optional and may be left out of the
data set.
The keywords listed above must appear in the Reservoir Description section and must appear
in the data set before the *NULL and *POR keywords.
*GRID describes the type of grid system that is being used. There are 4 choices: regular
Cartesian, variable depth/variable thickness, and radial-angular cylindrical and corner point
grid. Each of these choices requires the number of grid blocks in the I (x or r) direction, in
the J (y or theta) direction, and in the K (z) direction.
Example:
*GRID
*GRID
*GRID
*GRID
*CART
*VARI
*RADIAL
*CORNER
10 10 6
10 10 6
10 1 15
10 10 6
The first describes a regular Cartesian grid that is 10x10x6. The second describes a variable
depth/variable thickness grid that is also 10x10x6. The third example describes a radial-angular
cylindrical system for a coning study. It is 10x1x15. The last sets up a corner point grid.
The keywords *DI, *DJ, and *DK are required keywords (except for *CORNER). You enter
the dimensions of the grid blocks using these three keywords. You must use the array
reading options with these keywords.
76 Tutorial
User's Guide IMEX
Example:
*GRID *CART 10 10 12
*DI *CON 100.0
*DJ *CON 100.0
*DK *KVAR
25.0 2*50.0 3*40.0 75.0 3*40 2*50
where the grid system is a regular Cartesian grid system. Each of the 10 grid blocks in the I
direction is 100.00 meters wide. Each of the 10 grid blocks in the J direction is 100.0 meters
wide and each layer in the K direction has the same thickness but the thicknesses differ between
layers. Please note that your data starts with the bottommost layer when using *KDIR *UP.
User's Guide IMEX
Tutorial 77
Describing Refined Grid

To describe the location of refined grid, use *REFINE. You may also use the hybrid grid
option with *REFINE. Hybrid grids are cylindrical refinements within a Cartesian grid
system. *REFINE must appear in the Reservoir Description section and must appear in the
data set before the keywords *NULL and *POR.
*REFINE requires the number of refined blocks the fundamental grid blocks will be split up
into, in each direction where refinement is desired. *HYBRID allows you to split a Cartesian
well block into cylindrical cells. You must use *RANGE to indicate the location of the
fundamental grid blocks where refinement will occur.
For example, you want to split block (1,1,3) in a 10x10x3 regular Cartesian grid system into 2
refined grid blocks in the I direction, 3 blocks in the J direction and 2 in the K direction. The
keyword looks like this:
Example:
*REFINE 2
3
2
*RANGE 1:1 1:1 3:3
You are allowed to split up a fundamental block into and including 4 refined blocks. If you
want to split different areas into different configurations, then you may use subsequent
*REFINE keywords, being sure to keep track of the correct locations of those fundamental
blocks.
Note that *HYBRID areas that are split into different configurations adjacent to each other
have certain restrictions (see the *REFINE keyword for details). In some cases areas with
different degrees of refinement must be separated by at least one fundamental (unrefined)
grid block. Also note that grid refinement is not allowed when the dual-porosity option is
being used.
The variable thickness, variable depth option may be used with refined grid. However, the
thicknesses of individual refined blocks are assumed to be equal within each individual
fundamental grid block.
Now that you have stated that you want to use refined grid, you must use *RG for any
properties that may differ from the properties of the corresponding fundamental blocks.
Otherwise, the properties of the fundamental blocks are inherited.
78 Tutorial
User's Guide IMEX
Using Dual Porosity/Dual Permeability

The simulation of naturally fractured reservoirs can be conducted through the use of dual
porosity models. In these models, the reservoir is discretized into two collocated continua
(two sets of grid blocks located in the same space), one called the matrix and the other called
the fracture. The matrix continuum is assumed to be comprised of matrix blocks which are
separated spatially by fractures. The dimensions of these matrix blocks can be variable
throughout the reservoir and are a function of the fracture spacing, orientation and width.
Within the computational grid, the matrix- fracture transfer is represented by a single flow
term. The flow of fluids through the reservoir occurs through the fracture system, which is
the primary conduit for fluid flow. Wells in a fractured block are always connected to the
fracture.
IMEX contains four different types of dual-porosity models that can be used. When using
any of the dual porosity options, both matrix and fracture properties must be specified. This
is done by using the *MATRIX or *FRACTURE modifier after the main keyword.
Example:
Specify matrix and fracture porosities and permeabilities. Matrix porosity is

0.10 and fracture porosity is 0.05. Matrix permeability is 50.0 md areally with
kv/kh = 0.10. Fracture permeability is 500 md areally with kv/kh = 0.10.
In the Reservoir Description section enter the following:

*POR *MATRIX
*POR *FRACTURE
*PERMI *MATRIX
*PERMJ *MATRIX
*PERMK *MATRIX
*PERMI *FRACTURE
*PERMJ *FRACTURE
*PERMK *FRACTURE
*CON
0.10
*CON
0.05
*CON 50.0
*CON 50.0
*CON
5.0
*CON 500.0
*CON 500.0
*CON 50.0
In a similar manner, other properties such as assigning relative permeability tables to matrix
and fracture blocks or specifying blocks as IMPES or fully implicit can be accomplished.
Example:
Initially set matrix blocks to IMPES and fracture blocks to fully implicit.
In the Well and Recurrent Data section after the first *DATE keyword, enter the following:
*AIMSET
*AIMSET
*MATRIX
*FRACTURE
*CON 0
*CON 1
In addition to specifying the type of dual porosity option to use, the method of calculating the
shape factor must also be specified. IMEX has two methods available for calculating the
shape factor. They are based either on the Warren and Root model or the Gilman-Kazemi
model. Please refer to the appropriate section of the User's Guide for further reading on the
two methods for calculating shape factor. To invoke the Warren and Root method, specify
*SHAPE *WR after specifying the type of dual porosity option to use in the Reservoir
Description section of the data set. Similarly, to invoke the Gilman-Kazemi method, specify
*SHAPE *GK after specifying the type of dual porosity option to use in the Reservoir
Description section of the data set.
User's Guide IMEX
Tutorial 79
In order to calculate shape factors, IMEX requires information on the geometry of the fractures.
This is done by specifying the fracture spacing in the three principal directions (I, J and K
directions). The fracture spacing can vary spatially throughout the reservoir. Sections of the
reservoir can be dual porosity and sections can be set to single porosity by setting the fracture
porosity to zero. Specifying the fracture spacing in the three principal directions is done by
using the *DIFRAC, *DJFRAC and *DKFRAC keywords in the Reservoir Description section
of the data set. Note that specifying a very large fracture spacing indicates that the shape factor,
sigma, will be small and hence the matrix fracture transfer will be reduced. Specifying small
values of fracture spacing indicates the block contains many fractures and consequently, the
matrix-fracture transfer will be increased. Specifying a fracture spacing of zero indicates that
there are no fracture planes perpendicular to those axes. Note that the fracture spacing affects
only the matrix-fracture transfer term and not the number of connections between matrix and
fracture. In order to specify a region of the reservoir to be single porosity, it is necessary to set
the grid block fracture porosity to zero. Note that in order to null out matrix or fracture blocks,
the porosities must be set to zero. It is not possible to use:
*NULL
-or*NULL
*MATRIX
*CON 0
*FRACTURE
*CON 0
This is because *NULL applies to spatial grid blocks (in this case both matrix and fracture
blocks) and not to matrix and fracture blocks separately (active blocks).
Example:
A reservoir model consists of a 10x10x5 grid. All layers are fractured with
fracture spacings of 50 ft in the I, J directions and 10 ft in the K direction.
In the Reservoir Description section specify:

*DIFRAC
*DJFRAC
*DKFRAC
*CON
*CON
*CON
50.0
50.0
10.0
Layer 3 and 4 have been found to contain no fractures (they can be represented as a single
porosity system). Modify the data accordingly. Assume that the fracture porosity is 0.05.
*POR
1:10
1:10
FRACTURE *IJK
1:10 1:5 0.05
1:10 3.4 0.0
The four different dual porosity options are described below. Only one type of dual porosity
model can be specified in a data set.
a) Standard Dual Porosity
Fluid flow through the reservoir takes place through the fracture network. The
matrix blocks essentially act as source and sink terms. This model is invoked by
specifying the keyword *DUALPOR in the Reservoir Description section.
Example: Use standard dual porosity model with the Gilman-Kazemi method of
calculating the shape factor.
In the Reservoir Description section of the data set, enter the following:
*DUALPOR
*SHAPE *GK
80 Tutorial
User's Guide IMEX
b) Dual Permeability
This model differs from the standard dual porosity model in that each matrix block is
connected to both the fracture blocks and the surrounding matrix blocks. Now, fluid
flow occurs through both the fracture network as well as through the matrix blocks.
The dual permeability option can be important for cases where there is capillary
continuity. For these cases, dual permeability is often important for reservoirs with
free gas, oil and large variations in depth. Usually, only the vertical k direction
matrix-matrix transfer is important. The dual permeability option is invoked by
specifying *DUALPERM in the Reservoir Description section. Matrix-matrix flow
can be reduced or set to zero through the use of matrix transmissibility multipliers.
Example:
Use the dual permeability model with the Gilman-Kazemi method

of calculating the shape factor. Also set matrix-matrix transfer in
vertical direction only.
In the Reservoir Description section of the data set, enter the following:
*DUALPERM
*SHAPE
*GK
*TRANSI *MATRIX
*TRANSJ *MATRIX
*TRANSK *MATRIX
*CON
*CON
*CON
0.0
0.0
1.0
c) Subdomain
This model is a variation of the standard dual porosity option. It allows the user to
refine the matrix blocks in the vertical direction in order to more accurately represent
the gravity drainage process from matrix to fracture which is essentially vertical.
Each matrix block is refined by the number of subdivisions specified in order to
more accurately represent the fluid pressures and saturations within the matrix
blocks. This resolution is not needed in the fracture, therefore only the matrix blocks
are refined. The subdomain option is invoked by specifying *SUBDOMAIN ndiv in
the Reservoir Description section, where ndiv is the number of refined layers for
each matrix block. Note that the value for ndiv is constant throughout the reservoir.
Example:
Modify the dual porosity example case above to subdomain.

Assume the number of vertical subdivisions are 3.
In the Reservoir Description section:

*SUBDOMAIN 3
*SHAPE *WR
Assume the reservoir model consists of a 10x10x5 grid. All layers are fractured
with fracture spacings of 50 ft in the I, J directions and 10 ft in the K direction.
Add the following in the Reservoir Description section:
*DIFRAC *IJK
1:10 1:10 1:5
*DJFRAC *IJK
1:10 1:10 1:5
*DKFRAC *IJK
1:10 1:10 1:5
User's Guide IMEX
50.0
50.0
10.0
Tutorial 81
d) MINC
Another extension of the standard dual porosity concept is the multiple interacting
continua (MINC) approach proposed by Pruess and Narasimhan. The main
advantage of MINC is the representation of the matrix-fracture transfer calculation
using a nested discretization of the matrix blocks. This allows a very efficient
representation of the transient fluid regime, which is often neglected, in the standard
dual porosity model. The nested discretization is one-dimensional and can represent
the pressure, viscous and capillary forces. The gravity force however is not
considered in this nested, one-dimensional matrix refinement. The MINC option is
invoked by specifying *MINC ndiv in the Reservoir Description section, where ndiv
is the number of nested layers for each matrix block. Note that the value for ndiv is
constant throughout the reservoir.
Example:
Modify the dual porosity example case above to subdomain.

Assume the number of nested subdivisions is 3.
In the Reservoir Description section:

*MINC 3
*SHAPE *WR
Assume the reservoir model consists of a 10x10x5 grid. All layers are fractured
with fracture spacings of 50 ft in the I, J directions and 10 ft in the K direction.
Add the following in the Reservoir Description section:
*DIFRAC *IJK
1:10 1:10 1:5 50.0
*DJFRAC *IJK
1:10 1:10 1:5 50.0
*DKFRAC *IJK
1:10 1:10 1:5 10.0
Summary: To invoke the dual porosity/dual permeability options you may use:
a) *DUALPOR
b) *MINC
c) *SUBDOMAIN
d) *DUALPERM
e) *SHAPE
f) *DIFRAC
g) *DJFRAC, and
h) *DKFRAC
These keywords must appear in the Reservoir Description section. For the
different options that are available, only one may be used in any given data set.
If any of these are used, locate them before *NULL and *POR.
82 Tutorial
User's Guide IMEX
Input of Null Blocks

There are two ways to indicate the presence of null blocks within a given grid system:
a) *NULL and
b) *POR
Both must appear in the Reservoir Description section. *NULL is optional, while *POR must
appear in your data set, in any case. *POR is the keyword used for specifying porosities.
With *NULL, null blocks are indicated by the number 0; active blocks are indicated by the
number 1. In the example below, all blocks except blocks 1 to 4 in the I direction, 1 to 3 in J
direction and blocks 1 to 3 in the K direction, are active.
You may use the *IJK array reading option for this example:
Example:
*NULL *IJK
1:10 1:10 1:3 1
1:4 1:3 1:3 0
Observe that the second line overrides the first line. *NULL is optional and if it is not
present, it is assumed that all blocks are active.
The designation by *NULL always overrides the designation in *POR. If a nonzero porosity
is assigned to a block that has already been designated a null block by *NULL, then the
*POR assignment is ignored.
Null blocks using *POR are indicated by a zero porosity, while active blocks are assigned
nonzero values.
The aforementioned example may also be represented by *POR:
Example:
*POR *IJK
1:10 1:10 1:3 0.3
1:4 1:3 1:3 0.0
DUAL POROSITY/DUAL PERMEABILITY CASE

In the case of a dual porosity/dual permeability model, the input of porosity values requires
input CASE for the matrix and the fracture. Data for the matrix must be entered first and then
the data for the fracture. This procedure is similarly expected for other data.
Example:
*POR
1:10
1:4
*POR
1:10
8
*MOD
8
*MATRIX *IJK
1:10 1:3 0.3
1:3 1:3 0.0
*FRACTURE *IJK
1:10 1:3 0.0
7:9 1:2 0.4
7:8
User's Guide IMEX
1 = 0.45
Tutorial 83
The example also illustrates the use of *MOD, which modifies the grid property of some
blocks, from a porosity of 0.40 to a new porosity of 0.45.
In dual porosity, null blocks imply that both the matrix and the fracture have zero porosity.
In general, either the matrix porosity or the fracture porosity may be set to zero and the other
nonzero.
84 Tutorial
User's Guide IMEX
Using the Aquifer Option (see Appendix G)

The equations that are used in IMEX to model an aquifer are given in Appendix G and in
Carter, R.D. and Tracy, G.W., "An Improved Method for Calculating Water Influx," Trans.,
AIME, vol. 219, (1960), pp. 415-417. These equations are an approximation to the equations
of Everdingen and Hurst. Please refer to van Everdingen, A.F. and W. Hurst, "The
Application of the Laplace Transform to Flow Problems in Reservoirs," AIME, Dec. 1949,
pp. 305-324B and Chapter 9 in "Fundamentals of Reservoir Engineering" by L.P. Dake.
In order to use the aquifer option the following keywords could be used:
a) *AQUIFER
b) *AQPROP
c) *AQMETHOD
d) *AQLEAK
e) *AQFUNC
These keywords are expected in the Reservoir Description section and are completely
optional. If these options are used, they must be placed after *NULL and *POR.
The keyword *AQUIFER is required to activate the aquifer option. The location of the
aquifer is specified through the subkeywords *BOUNDARY *BOTTOM and *REGION.
*AQUIFER *REGION i1:i2 j1:j2 k1:k2 *IDIR will indicate that the aquifer is connected to
the grid blocks specified by the i1:i2 j1:j2 k1:k3. *IDIR specifies that the aquifer is
connected to faces perpendicular to the I direction. For a grid block i,j,k the aquifer is
connected to the face bordering this grid block and grid block i+1,j,k .
*AQUIFER *BOUNDARY is used to connect the aquifer to all the grid blocks at the
boundary of the reservoir.
*AQUIFER *BOTTOM is used to connect the aquifer to all grid blocks at the bottom of the
reservoir.
The properties of the aquifer can be specified using the keyword *AQPROP followed by the
values of the aquifer thickness, porosity, permeability, radius, and angle and the ratio of the
aquifers external radius to the reservoirs effective radius (Fetkovich aquifer only).
*AQPROP
thickness porosity permeability radius angle (R-ratio)
The simulator calculates default values for the thickness, radius and angle. In order to use
these default values please input 0.0 at the appropriate location.
Carter and Tracy's equations require the usage of a dimensionless pressure influence function
P(td) as a function of the dimensionless time td. This function can be specified using the
keyword *AQFUNC. The default dimensionless pressure function used is the one given by
van Everdingen and Hurst for a constant terminal rate solution and an infinite radial aquifer.
Influence functions for limited extent aquifers can also be found in the same reference.
Please note that Carter and Tracy's formulation requires the usage of a dimensionless pressure
influence function P(td) instead of the dimensionless rate influence function Q(td) that are
normally used to calculate flux rates for terminal pressure problems.
User's Guide IMEX
Tutorial 85
If a Fetkovitch aquifer is specified, the simulator calculates water influx from the aquifer
using a formulation based on work by Fetkovitch (see for instance, M.J. Fetkovitch, "A
Simplified Approach to Water Influx Calculations - Finite Aquifer Systems," JPT, July 1971,
pp. 814-828). This approach is able to model finite aquifers without having to generate
dimensionless pressure functions. Only a "R-ratio" parameter is required.
The *AQLEAK option allows for the analytical aquifer model to have negative influx or
outflow from the reservoir when the reservoir pressure exceeds the aquifer pressure. Since
the outflow from the reservoir is modelled analytically it is possible to have water outflow
from a grid block attached to the aquifer that has no mobile water. Thus great care should be
exercised when using *AQLEAK *ON. It is strongly recommended that aquifers be
modelled using grid blocks, for data sets where the reservoir pressure is expected to increase
significantly during the course of the simulation.
86 Tutorial
User's Guide IMEX
Using the Pseudo-Miscible Option

Define the pseudo-miscible model by using either of *MODEL *MISCG, which denotes a
pseudo-miscible model, including chase gas or *MODEL *MISNCG, which denotes a
pseudo-miscible model, not including chase gas. These keywords must appear in the
Component Properties section of your data set.
The following keywords are additional required keywords to be included in your data set
when you use the pseudo-miscible option.
You need:
a) *PVTS
b) *DENSITY *SOLVENT
c) *OMEGASG
d) *MINSS
e) *GORINT (*MODEL *MISCG only)
These keywords must be located in the Component Properties section.
f) *SWT, where capillary pressure and the irreducible oil saturation is required input.
*SWT must appear in the Rock-Fluid Data section of your data set.
g) *PBS; this keyword must appear in the Initial Conditions section of your data set.
h) *INCOMP *SOLVENT *GLOBAL
i)
*OPERATE *SOLVENT, and
j)
*ALTER, perhaps, to change the solvent rate at subsequent well changes. h), i)
and j) must appear in the Well and Recurrent Data section of your data set.
User's Guide IMEX
Tutorial 87
Using the API Tracking Option

Define the API tracking model by using either of *MODEL *API-INT, which denotes a three
phase model, or *MODEL *API-INTOW, which denotes a two phase (oil-water model).
These keywords must appear in the Component Properties section of your data set.
when you use the API tracking option.
You need:
a) *API or *APIT (INITIAL CONDITIONS)
b) *PVTAPI (COMPONENT PROPERTIES)
c) *APIGRAD (COMPONENT PROPERTIES)
d) *APIGRAV (INPUT / OUTPUT CONTROL)
When using the API tracking option, the oil is assumed to be comprised of two components, a
light component and a heavy component.
The volume fraction of light oil at STC is defined as Vl, the volume fraction of heavy oil is
Vh = (1-Vl).
The density of pure light oil at STC is denol(STC), and the density of pure heavy oil is
denoh(STC). The density denoh(STC) is the density of the oil mixture at Vl=0.0. This is the
first *APIGRAD table which must be defined. The last *APIGRAD table must be at defined
for pure light oil (Vl=1.0). The value of *APIGRAD for this table is therefore denol(STC).
The density of the mixture deno(STC) is:
deno(STC) = denoh(STC)*Vh + denol(STC)*Vl
The solution gas ratio of the mixture Rs is:
Rs = Rs(mix)
where Rs(mix) is obtained by interpolation of *APIGRAD tables whose oil mixture densities
are just lower and just higher than the density defined by the equation for deno(STC) above.
If only two APIGRAD tables are used for a PVT region, this is equivalent to:
Rs(mix)= Rsh*Vh + Rsl*Vl
where Rsh is the Rs of the heavy component and Rsl is the Rs of the light component.
Oil mixture formation volume factor oil compressibility (co) and viscosity is derived in the
same manner.
Using the *API keyword or the *APIT keyword, light oil volume fraction (at STC) is
initialized. Using the *PVTAPI keyword a gas PVT property table is input for each PVT
region (note: separate gas and liquid tables are used in this option).
The *APIGRAD tables follow the *PVTAPI gas table for each PVT region. Each APIGRAD
table is associated with a oil mixture density for a PVT region. Oil mixture densities must be
entered from heaviest (Vl=0) to lightest (Vl=1.0). Multiple PVT Regions may be used with
this option.
88 Tutorial
User's Guide IMEX
Using the Rock Compaction/Dilation Model Option

when you use the rock compaction model option. These keywords must be located in the
Reservoir Description section (see the keywords *CROCKTYPE, *CROCKTAB,
*CROCKTABH, *CCPOR, *CCPRPOR, *CIRREVERS, and *CTYPE for information).
It is always necessary to define both *CPOR and *PRPOR when the analytical aquifer model
is used. The aquifer compressibility is not defined by any of the options available on the
keywords following *CROCKTYPE, only *CPOR defines aquifer compressibility.
When a region of the reservoir is not explicitly assigned a value of *CTYPE while other
regions are assigned *CTYPE values, it is assumed the unassigned regions compressibility
and reference pressure are defined by the *CPOR and *CPRPOR keywords.
Please note that the *INT option used with input tables is not operational with the
*CROCKTAB and *CROCKTABH tables.
You need:
a) For the case of constant rock compressibility which varies by *CTYPE region
*CROCKTYPE
*CCPOR
*CPRPOR
For example:
*CROCKTYPE 1
*CCPOR
3.0e-6
*CPRPOR 14.7
*CROCKTYPE 2
*CCPOR
*MATRIX
*CPRPOR *MATRIX
*CCPOR
*FRACTURE
*CPRPOR *FRACTURE
(Note: for rock region 1,

fracture properties defaulted to
matrix properties)
3.0e-6
14.7
5.0e-6
14.7
or
b) For the case of reversible nonconstant rock compressibility or the case when
permeability varies with pressure. (This option can only be applied on single
porosity, dual porosity and dual permeability models).
*CROCKTYPE
*CROCKTAB
For example:
*CROCKTYPE
*CROCKTAB
**
**pressure
1000
2000
3000
4000
User's Guide IMEX
1
porosity multiplier
1.002956
1.005956
1.008956
1.011956
permeability
multiplier (hor.)
1.014445
1.029286
1.044309
1.059518
Tutorial 89
The above porosity multipliers are computed by using the following equation:
porosity multiplier = 1 + cpor * (p - prpor)
(1)
where:
=
=
cpor
prpor
3.0e-6 1/psi
14.7 psi
p is pressure on the table.

and permeability multipliers are determined by the modified equation (2) (Espinoza, C.E. A
New Formulation for Numerical Simulation of Compaction, Sensitivity Studies for Steam
Injection, SPE 12246, 1983.):
perm.multiplier = [ ( por.mult ) * * m ]
[(( 1 / por _ input 1) / ( 1 / por _ input por.mult )) ** 2 ]
(2)
where:
por.mult
por_input
=
=
porosity multiplier
initial porosity which equals 0.3 in the examples
m is an adjustable parameter that depends on the rock type. Assuming its value to
be 4 in the examples.
In the example above because separate horizontal and vertical permeability multipliers were
not entered, vertical permeability multipliers default to horizontal permeability multipliers.
or
c) For the case of irreversible rock compressibility and permeability multipliers.
(This option can only be applied on single porosity, dual porosity and dual
permeability models).
*CROCKTYPE
*CROCKTAB
*CIRREVERS
For example:
*CROCTYPE 1
*CROCKTAB
*CIRREVERS
**
**pressure
1000
2000
3000
4000
90 Tutorial
porosity multiplier
1.002956
1.005956
1.008956
1.011956
permeability
multiplier (hor.)
1.014445
1.029286
1.044309
1.059518
User's Guide IMEX
or
d) For the case of rock compressibility and permeability multipliers which undergo
hysteresis. (This option can only be applied on single porosity, dual porosity and
dual permeability models).
*CROCKTYPE
*CROCKTAB
*CROCKTABH
For example:
*CROCTYPE 1
*CROCKTAB
**
**pressure
1000
2000
3000
4000
*CROCKTABH
1000
2000
3000
4000
*CROCKTABH
2000
3000
4000
*CROCKTABH
3000
4000
porosity multiplier
1.002956
1.005956
1.008956
1.011956
permeability
multiplier (hor.)
1.014445
1.029286
1.044309
1.059518
1.002956
1.004962
1.006968
1.008974
1.014445
1.024348
1.034333
1.044400
1.005956
1.007968
1.009980
1.029286
1.039341
1.049480
1.008956
1.010974
1.044309
1.054519
The above porosity multipliers on hysteresis branches are also computed on the basis of
equation (3) as below:
porosity multiplier = [1 + cpor * (p - prpor)] * [1 + cpor' * (p - Pr)]
(3)
where:
cpor
prpor
cpor
=
=
=
3.0e-6 1/psi
14.7 psi
2.0e-6 1/psi
Pr is rebound pressure (first pressure on a hysteresis branch)

P is pressure on the table
User's Guide IMEX
Tutorial 91
A graph of hysteresis branches based on the above data is shown below.
porosity multiplier
1.014
1.012
rebound pressure
1.01
1.008
1.006
1.004
1.002
0
1000
2000
3000
4000
5000
pressure (psi)
and
permeability multiplier
1.07
1.06
1.05
1.04
1.03
1.02
1.01
0
1000
2000
3000
4000
5000
pressure (psi)
Note that the first point in all the hysteresis tables must match exactly a point in the main
table. Failure to ensure this is true could cause numerical difficulties.
When the *CTROCKTAB table is quite nonlinear it is important to define a sufficient
number of hysteresis branches using *CROCKTABH. If the *CROCKTABH data is sparse it
is possible that there will be a porosity discontinuity at the rebound pressure. This can be
minimized by increasing the number of *CROCKTABH tables and can be eliminated entirely
by defining a *CROCKTABH table for every pressure point in the main table.
92 Tutorial
User's Guide IMEX
e) When more than one rock type are used

*CTYPE
For example: A 10x10x4 grid blocks reservoir consists of 4 layers and each layer has one
rock type except the first layer which has two rock types. Distribution of rock types on grid
blocks of the reservoir is given as follows assuming 4 crocktypes:
*CTYPE
*CTYPE
*CTYPE
*CTYPE
*CTYPE
1:5
6:10
1:10
1:10
1:10
1:10
1:10
1:10
1:10
1:10
1
1
2
3
4
1
2
2
3
4
When the rock dilation model option is used, a number of additional keywords should be
included in a data set. Those keywords must be located in the RESERVOIR DESCRIPTION
section (see the keywords *CROCKTABE, *CROCKTABD, *CROCKTABU,
*CROCKTABC, *CROCKTABR for more details).
It is also noted that those above keywords can not be combined with any other keywords of
the rock compaction model (*CROCKTAB, *CROCKTABH and *CIRREVERS) and vice
versa for one rock type.
In one reservoir, rock compaction model and rock dilation model can not be used
simultaneously. However, if there are more than one *CROCKTYPE (reservoir) in a
simulation, each reservoir can use one of the rock models if those *CROCKTYPE's
(reservoirs) are not in contact to each other.
Similar to the rock compaction model, it is always necessary to define both *CPOR and
*PRPOR when the analytical aquifer model is used.
*CROCKTYPE and *CTYPE keywords are also used as in the case of the rock compaction
model to define rock type numbers and rock type regions.
A following sample data for rock dilation model that is based on formulae by Beattle, C.I,
Boberg, T.C. and McNab, G.S.: "Reservoir Simulation of Cyclic Steam Stimulation in the Cold
Lake Oil Sands", SPE Reservoir Engineering, May 1991, pp. 200-206 and STARS Users
Guide Version 98 is under keywords *CROCKTABE, *CROCKTABD, *CROCKTABU,
*CROCKTABC and *CROCKTABR.
Parameters that were used in the formulae to create the data are:
pbase
pdila
ppact
crd
cp
fr
phi0
pmin
m
User's Guide IMEX
=
=
=
=
=
=
=
=
=
14.7 psi
400.0 psi
200.0 psi
0.0007 (1/psi)
0.0001 (1/psi)
0.1
0.3
0.001 psi
1.0
Initial pressure at which elasticity begins

Pressure at which dilation begins
Pressure at which recompaction begins
Dilation rock compressibility
Elastic rock compressibility
Residual dilation fraction
Initial porosity
Minimum pressure that recompaction ends
An adjustable parameter used in permeability
multiplier formula
Tutorial 93
Please see the below figure for locations of the above corresponding pressures.
Multiplier
C
cp
D
E
cr
crd
cp
G
AB:
BC:
DE:
EF:
FG:
B
A
Elastic path
Dilation path
Unloading path
Recompaction path
Reloading path
cp,cr,crd: rock compressibility
cp
Pressure
pmin pbase
ppact
pdila
pend
A Reservoir Rock Dilation Model Based on Formulae
Also noted that when using the formulae to create data for rock dilation model, the elastic
path, unloading path and reloading path have the same value of rock compressibility.
*CROCKTABE
** PRESSURE
14.7000
100.0000
150.0000
200.0000
250.0000
300.0000
350.0000
400.0000
POR_MULT
1.0000000
1.0085665
1.0136219
1.0187027
1.0238090
1.0289409
1.0340985
1.0392819
HOR_PERM_MULT
1.0000000
1.0160131
1.0255614
1.0352320
1.0450268
1.0549481
1.0649980
1.0751787
VER_PERM_MULT
1.0000000
1.0000000
1.0000000
1.0000000
1.0000000
1.0000000
1.0000000
1.0000000
Porosity multipliers in the *CRCOCKTABE table are computed by a formula:

por_mult = exp(cp*(p pbase))
(1)
where:
p (psi) is pressure in the first column of *CROCKTABE table.
94 Tutorial
User's Guide IMEX
Horizontal permeability multipliers are given by Espinoza, C.E.: "A New Formulation for
Numerical Simulation of Compaction, Sensitivity Studies for Steam Injection", SPE 12246,
1983.
hor_perm_mult = [(por_mult)**m]*[((1/phi0 1)/
(1/phi0 por_mult))**2]
(2)
*CROCKTABD
** PRESSURE
400.0000
500.0000
600.0000
800.0000
1000.0000
1200.0000
1300.0000
1400.0000
1500.0000
POR_MULT
1.0392819
1.1146383
1.1954587
1.3751049
1.5817471
1.8194423
1.9513667
2.0928568
2.2446060
HOR_PERM_MULT
1.0751787
1.2327992
1.4240450
1.9523725
2.8069041
4.3221774
5.5628500
7.4048444
10.3099231
VER_PERM_MULT
1.0000000
1.0000000
1.0000000
1.0000000
1.0000000
1.0000000
1.0000000
1.0000000
1.0000000
Porosity multipliers in the *CROCKTABD table are given by:

por_mult = exp(cp*(pdila-pbase))*exp(crd*(pd(i) pdila))
(3)
where:
pu(i) is pressure at i row in the *CROCKTABU table.
Horizontal permeability multipliers are computed by using the formulae (2).
*CROCKTABU
** PRESSURE
700.0000
600.0000
500.0000
450.0000
400.0000
300.0000
250.0000
200.0000
POR_MULT
1.2821393
1.2693818
1.2567512
1.2504832
1.2442463
1.2318659
1.2257219
1.2196086
HOR_PERM_MULT
1.6591104
1.6223585
1.5867360
1.5693339
1.5521972
1.5186989
1.5023269
1.4861998
VER_PERM_MULT
1.0000000
1.0000000
1.0000000
1.0000000
1.0000000
1.0000000
1.0000000
1.0000000
Porosity multipliers in this *CROCKTABU table are computed by:

por_mult = exp(cp*(pdila - pbase))*exp(crd*(p pdila))*
exp(cp*(pu(i) pu(1)))
(4)
where:
pu(i) is pressure at i row in the *CROCKTABU table.
pu(1) is the first pressure in the *CROCKTABU table.
In the equation (4), unloading path has the same rock compressibility as that of the elastic path.
Horizontal permeability multipliers are computed similarly by using the formulae (2).
User's Guide IMEX
Tutorial 95
*CROCKTABC
** PRESSURE
200.0000
150.0003
100.0005
50.0008
0.0010
POR_MULT
1.2196086
1.1671184
1.1168873
1.0688181
1.0228177
HOR_PERM_MULT
1.4861998
1.3541457
1.2377947
1.1347675
1.0431193
VER_PERM_MULT
1.0000000
1.0000000
1.0000000
1.0000000
1.0000000
Porosity multipliers in the *CROCKTABC table are computed by:

phii = phi0*exp(-cp*pbase) + fr*(phi0*exp(cp*(pdila-pbase)))*
(exp(crd*(pu(1)-pdila)) 1)
(5)
cr = (ln(phii/ (phi0*exp(cp*(pdila-pbase))*
(exp(crd*pu(1)-pdila)))-cp*(ppact-pu(1)))/ppact
(6)
por_mult = phii * exp(-cr*pc(i))/phi0
(7)
where:
pc(i) is recompaction pressure which is determined by dividing equally interval pressure
(0.001, ppact) by a number of intervals such as 5 as in this example.
Horizontal permeability multipliers were obtained by using equation (2).
*CROCKTABR
** PRESSURE
0.0010
88.0112
176.0214
264.0316
352.0418
440.0520
POR_MULT
1.0228177
1.0318593
1.0409808
1.0501829
1.0594664
1.0688319
HOR_PERM_MULT
1.0431193
1.0606251
1.0785331
1.0968556
1.1156055
1.1347960
VER_PERM_MULT
1.0000000
1.0000000
1.0000000
1.0000000
1.0000000
1.0000000
In the above table, porosity multiplier is computed as follows:

i)
Pressure at the intersection point between reloading path and dilation path:
Assuming that the reloading path has the same rock compressibility as that of the
elastic path. Pressure at the intersection point between reloading path and dilation
path is computed as:
pend = (ln(phi0*exp(cp*(pdila pbase))/por_cend) +
cp*pc_end crd*pdila)/(cp crd)
where:
pend
por_cend
pc_end
:
:
:
(8)
pressure at the intersection point

porosity at the last pressure pc_end in the
*CROCKTABC table
last pressure in the *CROCKTABC table (= 0.001 psi as
shown above)
Reloading pressure pl(i) in the CROCKTABR table is determined by dividing the

pressure interval (0.001,pend) by a number of intervals such as 6 as in the example.
96 Tutorial
User's Guide IMEX
ii) Porosity multiplier is obtained as:

por_mult = por_cend*exp(cp*(pl(i) pc_end)))/phi0
(9)
In the equation (9), reloading path has the same rock compressibility as that of the elastic path.
Horizontal permeability multipliers in the *CROCKTABR table are computed through
equation (2).
User's Guide IMEX
Tutorial 97
Using the Polymer Option

Define the polymer option by either *MODEL *POLY which models the flow of oil, water,
gas, and polymer, or by *MODEL *POLYOW which denotes a polymer model with no gas
flow or no variation in solution gas. These keywords must be located in the Component
Properties section.
The following keywords are additional keywords to be included in your data set when you
use the polymer option.
You need:
a) *PADSORP (*MODEL *POLY only)
b) *PMIX (*MODEL *POLY only)
c) *PVISC
d) *PREFCONC
e) *PPERM
These keywords must be located in the Component Properties section of your data set.
f) *DISPI
g) *DISPJ
h) *DISPK
These must appear in the Rock-Fluid Data section.
i)
*INCOMP *WATER, followed by the initial polymer concentration
j)
*ALTERCP
These keywords must appear in the Well and Recurrent Data section of your data set.
98 Tutorial
User's Guide IMEX
Problems with Small Timesteps or Long Execution Times

When problems of a numerical nature are encountered it is extremely helpful to rerun the
simulation with
*WPRN *ITER *MATRIX
which turns on the matrix convergence as well as the Newtonian iteration convergence
diagnostics.
Convergence failure may result due to:
a) Inner (linear solver) iteration convergence failure
b) Newtonian convergence failure resulting in timestep cuts, or
c) Material balance error.
If you find in the output file that the "iteration routine fails to converge" frequently, then try
these remedies:
1. Take smaller timesteps. This is done by setting a smaller maximum timestep size
with *DTMAX or reducing the desired changes per timestep with *NORM
*PRESS and/or *NORM *SATUR.
2. Increase the number of specified iterations by using the keyword *ITERMAX, or
3. Increase the degree of factorization by using *SDEGREE. Please note that this
remedy increases storage requirements.
If the iteration routine fails to converge on the first or second Newton iterations, but
converges on at least the last one then it is not a serious problem.
Newtonian iteration convergence failure results in timestep cuts and are caused by maximum
changes, in the primary variables, which exceed the specified nominal changes by a factor more
than two per timestep. Nonphysical values such as negative pressures and/or saturations may
be encountered or the specified maximum number of Newtonian iterations is exceeded.
If the problem is caused by maximum changes, it is not a major problem IF it does not occur
often. If large numbers of time-cuts occur, then you can try the following remedies:
4. Check the rock and PVT curves for nonlinearities, discontinuities or inflection
points. The curves should be smooth. Also note whether IMEX warned you of
negative total compressibility in your PVT table. If negative total compressibility
was observed, correct this.
5. Check that grid and other properties are properly specified.
6. Check the well constraint specification. It is good practice to always specify a
maximum bottomhole pressure for each injector and a minimum bottomhole
pressure for each producer.
7. Increase the specified number of Newton's cycle using the keyword
*NEWTONCYC if nonconvergence is caused by the maximum number of
Newtonian iterations being exceeded.
User's Guide IMEX
Tutorial 99
If the maximum number of iterations is reached due to an oscillation in values as exhibited by

an oscillation in maximum changes and by messages in the output file that gas is on or off in
the grid blocks, then smoothing nonlinearities (4) or reducing the timestep size (1) are better
solutions.
8. Increase the number of timestep cuts available by using the keyword *NCUTS.
9. Set some regions of the reservoir or the entire reservoir to fully implicit. The
default switching criterion, *AIM *STAB checks for switching from explicit to
implicit only if the grid block is a neighbor of an implicit block. Thus if there are
regions of the reservoir where there are dramatic changes taking place and these
regions are not adjacent to wells, then set the problem regions to implicit.
It is also possible to use the '*AIM *STAB *AND-THRESH' option to check blocks adjacent
to wells using stability constraints and all explicit blocks using threshold limits.
10. Ensure that is the adaptive implicit method is being used that *AIMWELL
*WELLNN is employed. The default for AIMWELL is to only set well blocks
implicit.
11. Ensure that reservoir block pore volumes are of a reasonable size. Very small
block can cause serious oscillations when connected to large blocks. The use of
*PVCUTOFF to null out small blocks can in many cases eliminate this problem.
Examples of such situations include:
a) When vertical equilibrium initialization is not used. In some cases, this may result
in large changes in pressure and saturation initially, even if all the wells are shut in.
Run fully implicit when this happens.
b) When there is gas cap. The bottom layer of the gas cap can be set implicit if there
is strong cusping - at least in the regions where the cusping occurs.
c) Where blocks have extremely high permeability, small changes in pressure make
very large changes in saturation. Set blocks to fully implicit in these regions.
Using 0.1 kPa as the pressure convergence tolerance is recommended for high
permeability areas.
Material balance errors can be caused by convergence tolerances being too large compared to
the nominal change at each timestep. Check to make sure that the tolerances are about one
order of magnitude less than the nominal values.
In most cases, the default values for desired changes, *NORM *PRESS and *NORM *SATUR,
and the tolerances, *CONVERGE *PRESS and *CONVERGE *SATUR, are adequate.
However, in cases where you are trying to simulate fractures or horizontal wells, it is best to use
smaller values. For coning problems, smaller values of desired changes are also recommended.
Definitions of some iteration variables found in the output file (when *WPRN *ITER is used)
include:
100 Tutorial
User's Guide IMEX
kcyc
Newton cycle counter
dpmx
maximum change in pressure over a timestep
iblk
grid block number
dpbmax
maximum change in bubble point pressure over a timestep
dsomax
maximum change in oil saturation over a timestep
dsgmax
maximum change in gas saturation over a timestep
dswmax
maximum change in water saturation over a timestep
iconv
convergence check counter
nitr
current iteration in subroutine jsolve
nimp
number of implicit blocks
omega
if first omega is small, e.g. 1.0E-3, then the iterative routine has
difficulty converging
rms
sum of the squares of the residuals
rmsi
initial rms
rms/rmsi
linear solver has converged if rms/rmsi less than 1.0E-6 or when all
current updates are less than the convergence tolerance * relative
tolerance
User's Guide IMEX
Tutorial 101
Problems with Solver Convergence

Larger problems (50,000 blocks plus) or problems with high permeability and low porosity
may suffer from poor solver convergence and poor material balance.
A possible solution for this behavior is to increase both the number of orthogonalizations the
solver performs and the number of iterations the solver performs.
The default is currently:
*NORTH
*ITERMAX
30
40
A possible combination which could be used to improve solver convergence is:

*NORTH
*ITERMAX
80
80
It is critical to ensure that the number of orthogonalizations is as large as storage (memory)

and CPU time restrictions allow.
The user must increase the internally allocated storage to access *NORTH > 30 (say 40), by
adding the following keywords in the Run Time Dimensioning section.
*DIM SOLVER_DIMENSIONING 'ON'
*DIM MAX_ORTHOGONALIZATIONS 40
It is critical to ensure that the number of orthogonalizations is as large as storage (memory)
and CPU time restrictions allow.
Values of 40 or higher for *NORTH are recommended as long as the extra memory
requirements for high values of *NORTH do not access virtual memory during the run, which
in turn would slow the model run times significantly.
If increasing the number of orthogonalizations has no effect try increasing the degree of the
factorization from one to two (*SDEGREE 1 *SDEGREE 2).
Another option which has been shown to improve difficult problems, is to increase the solver
convergence tolerance using the *PRECC keyword. Increase the solver tolerance to 1.00E-6.
(*PRECC 1.00E-6)
102 Tutorial
User's Guide IMEX
Simulating a Gas-Water or Gas Reservoir

To simulate dry gas in a gas-water or single-phase gas reservoir:
1. Use *MODEL *GASWATER.
2. Enter gas PVT properties using the *PVTG table, do not enter any oil phase
properties such as *DENSITY *OIL, *CO, *COT, *CVO, *SO, *PB, or *PBT.
These keywords must be located in the Component Properties section of the data set.
3. Enter only Sw (column 1), Krw (column 2) and optionally Pcgw (column 3) in the
*SWT table. Kro is not entered.
4. Enter *Sl or *Sg (Column 1) and Krg (column 2) in the *SLT (or *SGT) table. Krog
and Pcog are not entered.
These keywords must be located in the Rock-Fluid Data section of the data set.
5. Establish the gas-water contact with the following steps:
a. Use *VERTICAL *BLOCK_CENTER *WATER_GAS, or
b. Use *VERTICAL *DEPTH_AVE *WATER_GAS, if Pcgw is zero, or
c. USE *VERTICAL *DEPTH_AVE *WATER_GAS *TRANZONE, if
Pcgw is not zero.
d. Set gas-water contact using *DWGC.
e. Do not define initial oil saturation, or bubble point pressure.
These keywords must be located in the Initial Conditions section of the data set.
6. Ensure that the irreducible oil saturations Soirw, and Soirg are 0.0 with the following
steps:
a) Input a water relative permeability table. The first *SWT saturation
entry is Swcon, the last *SWT saturation entry is (1-Soirw). Ensure the last
*SWT entry is 1.0 to ensure Soirw is 0.0.
b) In your gas table it is necessary to ensure that Soirg calculated from the
table is zero.
There are four possible ways to enter the liquid-gas relative permeability table, and thus four
ways to ensure that Soirg is equal to zero. The four are listed below.
1. Using a *SGT table, not using the *NOSWC option: the final gas saturation entry
must be (1-Swcon).
2. Using a *SLT table, not using the *NOSWC option: the first liquid saturation entry
must be Swcon.
3. Using the *SGT table, using the *NOSWC option: the final gas saturation entry
must be 1.0.
4. Using the *SLT table, using the *NOSWC option: the first liquid saturation must
be zero.
These keywords must be located in the Rock-Fluid Data section of your data set.
User's Guide IMEX
Tutorial 103
Simulating a Gas-Water with Condensate Reservoir

To simulate a wet gas and water reservoir:
1. Use *MODEL *GASWATER_WITH_CONDENSATE.
2. Enter saturated gas and oil phase PVT properties using the *PVTCOND table, do
not enter any oil phase properties which are required for pressures other than the
bubble point or those which define the bubble point, such as *CO, *COT, *CVO,
*PB, or *PBT. Optionally, gas properties for undersaturated gas may be entered
using the *EGUST (*BGUST or *ZGUST) and *VGUST tables
These keywords must be located in the Component Properties section of the data set.
3. Enter Sw (column 1), Krw (column 2), Krow (column 3) and optionally liquid-gas
capillary pressure Pcgl (column 4) in the *SWT table.
4. Enter *Sl or *Sg (Column 1), Krg (column 2), and Krog (column 3) in the *SLT (or
*SGT) table. Pcog is not entered.
5. Pcgl is output in the Pcog output arrays for both ascii and SR2 output. The Pcow
output array is not used.
These keywords must be located in the Rock-Fluid Data section of the data set.
6. Establish the gas-water contact with the following steps:
a) Use *VERTICAL *BLOCK_CENTER *WATER_GAS, or
b) Use *VERTICAL *DEPTH_AVE *WATER_GAS, if Pcgl is zero, or
c) Use*VERTICAL *DEPTH_AVE *WATER_GAS *TRANZONE, if Pcgl is
not zero.
d) Set gas-water contact using *DWGC.
e) Do not define initial oil saturation, or bubble point pressure.
f) Define the initial Dew Point using the *PDEW or *PDEWT keywords.
These keywords must be located in the Initial Conditions Data section of the data set.
7. Enable output options which aid in model analysis with the following steps:
a) *OUTPRN/*OUTSRF *GRID *RV *DPP outputs Oil Content and Dew
Point Pressure to the output and SR2.
b) *PSPLIT *ON adds well production split information to the well output
(Output file and SR2 file). The surface oil production is made up of oil
from condensate and black oil. The surface gas production is made up of
free gas and solution gas.
These keywords must be located in the Input/Output Control section of the data set.
8. Enable numerical control options if necessary:
a) *NORM *PDW allows you to control timestep size based on dew point
pressure changes.
b) *MAXCHANGE *PDW allows you to control maximum timestep size
allowed based on dew point pressure changes.
These keywords must be located in the Numerical Methods Control section of the data set.
104 Tutorial
User's Guide IMEX
Simulating a Tilted Water-Oil/Water-Gas Contact

To simulate a tilted water contact it is necessary to properly define the grid array Water Cap
Pres Shift/Offset which corresponds to the IMEX keyword PCOW_SHIFT.
Water Cap Pres Shift/Offset can be used to add an offset to each blocks water-oil or watergas capillary pressure. This is used to set up a stable offset to each blocks water-oil contact.
The PCOW_SHIFT option can only be used only if the *DEPTH_AVE initialization method
is in effect.
In order to use this technique to model a tilted contact (1) each PVT regions WOC_PC must
either be defaulted (to zero) or explicitly set to zero and (2) the water-oil /water-gas capillary
pressure curves must extend to a water saturation of 1.00.
This tutorial assumes that PVT region averaged estimates of reservoir pressure and bubble
point pressure can be used to calculate average PVT properties in a region. The average PVT
properties are used to convert between contact heights and capillary pressure offsets.
This tutorial also assumes that that the effect of water and oil compressibility on the initial
water and oil density can be ignored or accounted for by slightly altering the average value of
Bw or Bo.
A digitized WOC map should opened in BUILDER. This map is then used to calculate the
input array Water Cap Pres Shift/Offset. The single map should be used to calculate the
Water Cap PresShift/Offset for all layers.
An estimate of each PVT regions average pressure P(est) must also be made in order to
approximate an average water formation volume factor Bw(est). The water formation volume
factor should also approximately account for water compressibility (i.e. Bw(est) = Bwi*(1Cw*(P(est)-Prw)). Normally the input reference pressure REFPRES in a PVT region is a good
estimate of P(est).
If the model being initialized is a Black Oil Model, an estimate of each PVT regions average
bubble point pressure Pb(est) must be made in order to estimate oil properties. Normally the
input bubble Pressure table PBT in a PVT region can provide a starting point in determining
Pb(est).
If the model being initialized is a Gas Water model, P(est) is also used to estimate the gas
formation volume factor Bg(est).
Note: In a Gas Water model the input array is still Water Cap Pres Shift/Offset and the
IMEX input keyword is stilled PCOW_SHIFT.
Procedure for Black Oil Models:
1. Open the WOC map you wish to use in BUILDER.
2. Define the models PVT regions. Normally a single PVT region would be used.
The input requires the user to define a WOC depth (DWOC) which is the PVT
regions water-oil contact. This water-oil contact will be used as a reference value
for the shifted water-oil contact calculation (the WOC when PCOW_SHIFT=0.0).
The value of the reference WOC must lie between the highest and lowest WOC in
the input WOC map to minimize the offset from the reference.
User's Guide IMEX
Tutorial 105
3. Use Pb(est) in each PVT region to determine a value of Rs(est) and Bo(est). Bo(est)
may need to approximately account for oil compressibility if the reservoir is
significantly above its bubble point. Bo(est) = Bo(Pb(est))*(1-Co(p(P(est)-Pb(est)).
4. Use P(est) in each PVT region to determine Bw(est). Bw(est) may need to be
altered to approximately account for water compressibility if the reservoir pressure
is significantly different from Prw. Bw(est) = Bwi*(1-Cw*(P(est)-Prw)).
5. Calculate the approximate reservoir density difference between water and oil
phases in each PVT region.
Den_diff = Denw(STC)/Bw(est)-(Denw(STC)+Rs(est)*Deng(STC))/Bo(est)
This enables the user to relate a WOC height difference in each block to a pressure
difference in each block. If Den_diff changes significantly between PVT regions
separate calculations must be done for each PVT region.
6. In BUILDER, specify that the temporary grid array CMGTemp Prop 1 for a
single layer is to be calculated from the WOC map. Copy the input specification to
all layers in the model. Calculate the array.
7. In BUILDER, define a formula to subtract the calculated WOC array from the
reference WOC for each PVT region from the calculated WOC array and then
multiply the result by Den_diff.
(Reference WOC - CMGTemp Prop1)*Den_diff
8. Use the formula above to define Water Cap Pres Shift/Offset. This will change
the WOC map into a capillary pressure offset map. If the reference DWOC or
Den_diff is significantly different in different PVT regions, the formula must
account for individual PVT regions unique reference WOC and/or Den_diff.
9. Calculate Water Cap Pres Shift/Offset. The array now contains the capillary
pressure offset required to maintain the desired tilted WOC.
In Field Units, the conversion constants in the Den_diff equation results in the following equation.
Den_diff (psi/ft)
(Denw(STC)/Bw(est)
(Deno(STC)+Rs(est)/5.615*Deng(STC))/Bo(est))/144
(Denw(STC)/Bw(est)
(Deno(STC)+Rs(est)*Deng(STC))/Bo(est))*0.009806
In SI Units,
Den_diff (kpa/m)
Procedure for Gas Water Models

1. Open the WGC map you wish to use in BUILDER.
2. Define the models PVT regions. Normally a single PVT region would be used. The
input requires the user to define a WGC depth (DWGC) which is the PVT regions
water-gas contact. This water-gas contact will be used as a reference value for the
shifted water-gas contact calculation (The WGC when PCOW_SHIFT=0.0). The
value of the reference WGC must lie between the highest and lowest WGC in the
input WGC map to minimize the offset from the reference
106 Tutorial
User's Guide IMEX
3. Use P(est) in each PVT region to determine Bg(est).

4. Use P(est) in each PVT region to determine Bw(est). Bw(est) may need to be
altered to approximately account for water compressibility if the reservoir pressure
is significantly different from Prw. Bw(est) = Bwi*(1-Cw*(P(est)-Prw)).
5. Calculate the approximate reservoir density difference between water and gas
phases in each PVT region.
Den_diff = Denw(STC)/Bw(est)-Deng(STC)/Bg(est)
This enables the user to relate a WGC height difference in each block to a pressure
difference in each block. If Den_diff changes significantly between PVT regions
separate calculations must be done for each PVT region.
6. In BUILDER, specify that the temporary grid array CMGTemp Prop 1 for a
single layer is to be calculated from the WGC map. Copy the input specification to
all layers in the model. Calculate the array.
7. In BUILDER, define a formula to subtract the calculated WGC array from the
reference WGC for each PVT region from the calculated WGC array and then
multiply the result by Den_diff.
(Reference WGC - CMGTemp Prop1)*Den_diff
8. Use the formula above to define Water Cap Pres Shift/Offset. This will change
the WGC map into a capillary pressure offset map. If the reference DWGC or
Den_diff is significantly different in different PVT regions, the formula must
account for individual PVT regions unique reference WGC and/or Den_diff.
9. Calculate Water Cap Pres Shift/Offset. The array now contains the capillary
pressure offset required to maintain the desired tilted WGC.
In Field Units, the conversion constants in the Den_diff equation results in the following
equation.
Den_diff (psi/ft)
(Denw(STC)/Bw(est) Deng(STC)/Bg(est)/5.615)/144
(Denw(STC)/Bw(est) Deng(STC)/Bg(est))*0.009806
In SI Units,
Den_diff (kpa/m)
Denw(STC)
Standard Condition water density (DENSITY WATER keyword)
Deno(STC)
Standard Condition oil density (DENSITY OIL keyword)
Deng(STC)
Standard Condition gas density (DENSITY GAS keyword)
Bw(est)
Water formation volume factor at P=P(est)
Bwi
Water formation volume factor at P=Prw (BWI keyword)
Bo(est)
Oil formation volume factor at Pb=Pb (est) in a PVT region
Bg(est)
Gas formation volume factor at P=P(est) in a PVT region
Rs(est)
Solution gas ratio at Pb=Pb (est) in a PVT region
Prw
Water reference pressure (PRW keyword)
User's Guide IMEX
Tutorial 107
P(est)
Estimated average reservoir pressure in a PVT region (estimated from

REFPRES Keyword)
Pb(est)
Estimated average bubble point pressure in a PVT region (estimated

from PBT table or PB array)
Co
Oil compressibility above the bubble point (CO keyword)
Cw
Water compressibility (CW keyword)
The simulator array PCOW_SHIFT corresponding to the BUILDER property Water Cap
Pres Shift/Offset will now be written when the data set is saved.
Finish defining the model and when complete run the simulator for one timestep. Open the
resulting SR2 file in RESULTS and examine the WOC/WGC distribution (viewing Water
Saturation) and compare this with the desired distribution. Normally differences are very small.
Some Geological packages output Capillary Pressure Entry Pressure values for each
gridblock. PCOW_SHIFT may be used to accept Entry Pressure input.
108 Tutorial
User's Guide IMEX
Horizontal Wells
Horizontal wells can be easily simulated with IMEX in the following way:
This method involves modelling the well as a line source (injector) or sink (producer). This
method neglects wellbore frictional pressure drop and liquid holdup effects. It is important to
take wellbore crossflow into account by using the '*XFLOW-MODEL *FULLY-MIXED'
crossflow model.
The same sequence applies for a producer or an injector. Initially, your data set contains:
a) *WELL
b) *PRODUCER
c) *OPERATE
Start with a constant pressure (bottomhole pressure) constraint. Use *MAX.
d) *GEOMETRY
You may set the wellbore direction with this keyword. For horizontal wells, you need to use
the *I or *J direction.
e) *PERF
It is ideal for horizontal or deviated wells. Remember to use *GEO with it.
*GEOMETRY and *PERF result in the output of well productivities. Run this data set and
observe the resulting production rates.
Remove *GEOMETRY, input your own productivities directly with *PERF and rerun. Keep
adjusting the well productivity until you get the desired production rate.
When you get the desired rate, rerun the data set using a rate constraint as the primary
operating constraint.
Multilateral horizontal wells can be modelled by using the *FLOW-TO *FLOW-FROM layer
keywords.
User's Guide IMEX
Tutorial 109
Vertical Equilibrium Calculation

IMEX currently provides seven different ways to initialize the reservoir.
1. *USER_INPUT
2. *VERTICAL *BLOCK_CENTER *WATER_OIL
3. *VERTICAL *BLOCK_CENTER *WATER_GAS
4. *VERTICAL *BLOCK_CENTER *WATER_OIL_GAS
5. *VERTICAL *DEPTH_AVE *WATER_OIL
6. *VERTICAL *DEPTH_AVE *WATER_GAS (*TRANZONE)
7. *VERTICAL *DEPTH_AVE *WATER_OIL_GAS
In the following each of these options will be described:
*USER_INPUT:
This option is used to directly specify the values of the pressure and
saturation for each grid block. The following keywords, which must be
located in the Initial Conditions section, are used with this option.
a) USER_INPUT
b) *PRES, (pressure)
c) *SW (water saturation, need not be entered if using the two phase model,
*MODEL *OILWATER)
d) *PB or *PBT (bubble point pressure)
e) *PDEW or *PDEWT (dew point pressure
*GASWATER_WITH_CONDENSATE or *VOLATILE_OIL models)
f) *PBS (solvent bubble point pressure, if using any of the pseudo-miscible
models)
g) *SO (oil saturation, need not be entered if using the two phase model,
*MODEL *GASWATER)
Example:
*INITIAL
*USER_INPUT
*PRES *CON 3400.0
*PB
*CON 2500.0
*SO
*CON
0.5
*SW
*CON
0.2
Previous to version 93.10 this option was invoked using the keywords *VERTICAL *OFF.
These keywords are still accepted by IMEX.
110 Tutorial
User's Guide IMEX
Using this option may cause a nonequilibrium initial distribution of fluids which may result
in large fluid flow during the first few timesteps. This option is typically used when a
simulation is started not from initial conditions but from another simulation run.
*VERTICAL *BLOCK_CENTER *WATER_OIL or
*VERTICAL *BLOCK_CENTER *WATER_GAS or
*VERTICAL *BLOCK_CENTER *WATER_OIL_GAS
These options are used to perform gravity-capillary equilibrium calculations
based on the values of the water-oil, oil-gas, and water-gas contacts and the
value of the pressure at a reference depth. The saturations for the entire grid
block are set equal to their values at the grid block center.
*VERTICAL *BLOCK_CENTER *WATER_OIL is used for reservoirs
containing only water and oil phases. The value of the water-oil contact
depth is specified using the keyword *DWOC.
*VERTICAL *BLOCK_CENTER *WATER_GAS is used for reservoirs
containing only water and gas phases initially. The value of the water-gas
contact depth is specified using the keyword *DWGC. *VERTICAL
*BLOCK_CENTER *WATER_OIL_GAS is used for reservoirs containing
water-oil and gas. The water-oil contact and gas-oil contact depth are
specified using the keywords *DWOC and *DGOC.
*GASWATER and *GASWATER_WITH_CONDENSATE models must be
initialized using either the *VERTICAL *BLOCK_CENTER
*WATER_GAS* or VERTICAL *DEPTH_AVE *WATER_GAS
initialization option. Use of the *DWGC keyword is required with the
*WATER_GAS initialization option.
The following keywords, which must be located in the Initial Conditions section, are used
with these options.
a) *REFDEPTH (reference depth)
b) *REFPRES (pressure at reference depth)
c) *PB or *PBT (bubble point pressure)
d) *DWOC (water-oil contact)
e) *DWGC (water-gas contact)
f) *DWOC and *DGOC (water-oil and gas-oil contacts)
g) *PDEW or *PDEWT (dew point pressure Condensate models only)
Optionally the user can define the initial water saturation along with all of the above by using
the *SWINIT option. This option attempts to adjust/alter Pcow curves to allow user defined
water saturations to be in gravity-capillary equilibrium.
Care has to be taken when this option is employed, nonzero Pcow curves must be used and
reasonable values of saturation must be entered. Highly altered Pcow can significantly effect
other aspects of the model. Please see the *SWINIT keyword for more details.
User's Guide IMEX
Tutorial 111
Example:
*INITIAL
*PB *CON 2500.0 **bubble point is constant
**throughout the reservoir.
*REFDEPTH 900.0
*REFPRES 3000.0
*DWOC
1250.0
*DGOC
788.0
For the above *BLOCK_CENTER options, the value of the saturation assigned to a given
grid block is given by the value of the saturation at the grid block centre. In this way, for
example, if a grid block has its block center slightly below the water-oil contact depth, the
water saturation assigned to the block will be equal to the value of the water saturation below
the water-oil contact.
*VERTICAL *DEPTH_AVE *WATER_OIL or
*VERTICAL *DEPTH_AVE *WATER_GAS (*TRANZONE) or
*VERTICAL *DEPTH_AVE *WATER_OIL_GAS
These options are also used to perform gravity-capillary equilibrium
calculations based on the values of the water-oil, oil-gas, and water-gas
contacts and the value of the pressure at a reference depth.
In contrast to the *VERTICAL *BLOCK_CENTER option the value of the
saturation assigned to a given grid block is the average over the grid block
volume of the saturations, and not simply the value of the saturation at the grid
block centre. In this approach, for example, if a grid block has its block center
slightly below the water-oil contact depth, the water saturation assigned to the
block is the average over the block volume of the local saturations, and not
simply the water saturation value below the water-oil contact.
The *TRANZONE option on the *WATER_GAS keyword allows the use of
Gas water capillary pressure that can be entered in the water relative
permeability table (*SWT). The *TRANZONE keyword can be omitted if
Pcgw is not entered.
*GASWATER and *GASWATER_WITH_CONDENSATE models must be
initialized using either the *VERTICAL *BLOCK_CENTER *WATER_GAS*
or VERTICAL *DEPTH_AVE *WATER_GAS initialization option. Gaswater and gas-liquid (*GASWATER_WITH_CONDENSATE condensate
option) capillary pressures are therefore only input using the *SWT table. Use
of the *DWGC keyword is required with the *WATER_GAS initialization
option.
112 Tutorial
User's Guide IMEX
The following keywords, which must be located in the Initial Conditions section, are used
with these options.
a) *REFDEPTH (reference depth)
b) *REFPRES (pressure at reference depth)
c) *PB or *PBT (bubble point pressure)
d) *DWOC (water-oil contact)
e) *DWGC (water-gas contact)
f) *DWOC and *DGOC (water-oil and gas-oil contacts)
g) *PDEW or *PDEWT (dew point pressure Condensate models only
*GASWATER_WITH_CONDENSATE or *VOLATILE_OIL)
Example:
*INITIAL
*VERTICAL *DEPTH_AVE *WATER_OIL
*PB *CON 2500.0 **bubble point is constant
**throughout the reservoir.
*REFDEPTH 900.0
*REFPRES 3000.0
*DGOC
788.0
User's Guide IMEX
Tutorial 113
Defining Multiple PVT Regions

Multiple PVT regions can easily be created in IMEX. The cards associated with their
definition are found in the Fluid Model section and Initial Conditions section.
This section will clarify the use of MPR's (Multiple PVT Regions).
Fluid Model Section
The component property section is made of three types of data, (1) data which is entered per
PVT region, (2) data which is entered once for the entire reservoir and (3) data which can either
be entered per PVT region or once for the entire reservoir.
If a single type 3 data entry is found, it is applied to the entire reservoir, if multiple entries are
found, the values are applied to each PVT region in turn.
The type 1 data (once per PVT region) are: PVT tables (*PVT, *PVTLO, *DIFLIB, *PVTAPI,
*APIGRAD *PVTG *PVTCOND), *COT, *BOT, *VOT, *VGUST, *EGUST, *ZGUST.
*BGUST, *DENSITY *OIL, *DENSITY *GAS, and *GRAVITY *GAS
The type 2 data (single entry applying to the entire reservoir) are: *REFPST, *DENSITY
*SOLVENT, *PVTS, *GORINT, *OMEGAS, *MINSS, *PADSORP, *PMIX, *PVISC,
*PREFCONC, and *PPERM
The type 3 data (either once per PVT region or a single entry applying to the entire reservoir)
are: *DENSITY *WATER, *CO, *CVO, *BWI, *CW, *REFPW, *VWI, *CVW
*TRES is a special position dependent case in that it defines the temperature used by PVT
Tables which follow its definition. *TRES may be redefined before any PVT table or ZGUST
table. Tables read in before each *TRES redefinition use the previous *TRES value, Tables
read in after each *TRES redefinition use the new value.
The type 1 and 3 data types support two types of format (excluding PVT Tables and *TRES):
Format 1:
A single entry per keyword.

For Example
*DENSITY *OIL 45.0

*DENSITY *OIL 47.0
*DENSITY *OIL 46.5
The first line refers to the density of oil in region PVT region 1, the second line to
region 2 and the third line to region 3.
Format 2:
Multiple entries per keyword (WILDCARDS NOT ALLOWED).

For Example
*DENSITY *OIL 45.0 47.0 46.5
The first value refers to the density of oil in region PVT region 1, the second value
to region 2 and the third value to region 3.
114 Tutorial
User's Guide IMEX
Type 1, 2 and 3 data are position independent (*TRES is an exception). IMEX keeps track of
the number of each type of data entered. During input processing IMEX records the maximum
number of PVT regions referred to by any type 1 or 3 keywords.
If any of the type 1 data has fewer PVT regions defined than the maximum, an error is issued.
If any of the type 3 data has only a single region defined, this value is applied to the entire
reservoir.
If any of the type 3 data has greater than one region defined, but less regions defined than the
maximum number of regions, and error is issued.
It is not possible to mix the use of the three oil compressibility options. Oil compressibility may
be entered either using *CO (a single value or one value for each PVT region), *COT (one per
PVT region), or by including an oil compressibility entry in each PVT table. Use of two or
more input options causes an error to be issued. The undersaturated oil *BOT table may be
used instead of any of any of the oil compressibility options (one per PVT region).
It is not possible to mix the use of the *CVO option and the *VOT option for determining
undersaturated oil viscosity. A single option must be used for all PVT regions.
Initial Conditions Section
1. The keywords *PBT *PDEWT *REFPRES *REFDEPTH *DWOC *DGOC

*DWGC *GOC_SW *WOC_SW *WOC_PC *GOC_PC are associated with
MPR's. The remaining keywords apply to all blocks.
2. *PBT (or *PDEWT) cards expects the user to input the PVT region number which
is associated with each table. Use of the *PB (or *PDEW) Cards define *PB (or
*PDEW) for all blocks. One PBT (or *PDEWT) table must appear for every MPR
defined in the PVT section.
3. The remaining cards (*REFPRES *REFDEPTH *DWOC *DGOC *DWGC
*GOC_SW *WOC_SW), if required, are associated with MPR's. The first card is
associated with the first MPR, the second card is associated with the second MPR.
For example:
*REFDEPTH
*REFDEPTH
*REFDEPTH
*REFPRES
*REFPRES
*REFPRES
*DWOC
*DWOC
*DWOC
'Depth of
'Depth of
'Depth of
'Pressure
'Pressure
'Pressure
'Depth of
'Depth of
'Depth of
PVT
PVT
PVT
for
for
for
WOC
WOC
WOC
region 1'
region 2'
region 3'
PVT region 1'
PVT region 1'
PVT region 1'
for PVT region 1'
for PVT region 2'
for PVT region 3'
'Depth of
'Pressure
'Depth of
'Depth of
'Pressure
'Depth of
'Depth of
'Pressure
'Depth of
PVT
for
WOC
PVT
for
WOC
PVT
for
WOC
region 1'
PVT region 1'
for PVT region 1'
region 2'
PVT region 2'
for PVT region 2'
region 3'
PVT region 3'
for PVT region 3'
is valid, as is:
*REFDEPTH
*REFPRES
*DWOC
*REFDEPTH
*REFPRES
*DWOC
*REFDEPTH
*REFPRES
*DWOC
User's Guide IMEX
Tutorial 115
is valid, as is (WILDCARDS NOT ALLOWED).:

*REFDEPTH 'PVT reg 1' 'PVT reg 2' 'PVT reg 3'
*REFPRES 'PVT reg 1' 'PVT reg 2' 'PVT reg 3'
*DWOC
'PVT reg 1' 'PVT reg 2' 'PVT reg 3'
Exact order of cards is not critical in the Initial Conditions section as all MPR's have been
defined in the Fluid Model section.
However it is important to not mix the use of the *DWOC-*DGOC keyword pair and the
*DWGC keyword to define equilibrium in multiple PVT regions. *DWGC should be used
when the initialization option is *WATER_GAS. An error will be issued if both the *DWOC
*DGOC keyword pair and the *DWGC keyword are found in the data.
Unlike the Fluid Property section where all MPR data must be defined for type 1 data and
possibly for type 3 data (if more than a single entry is found), initialization data will default
any undefined data to the value in the last PVT region refined.
For example if 6 PVT regions are defined and only three *DWOC entries are found:
*DWOC 2000.0
*DWOC 2100.0
*DWOC 2200.0
or
*DWOC 2000.0 2100.0 2200.0
The remaining three PVT regions will be assigned the value of *DWOC in *PVT region 3
(the last defined region) equivalent to:
*DWOC 2000.0
*DWOC 2100.0
*DWOC 2200.0
*DWOC 2200.0
*DWOC 2200.0
*DWOC 2200.0
or
*DWOC 2000.0 2100.0 2200.0 2200.0 2200.0 2200.0
116 Tutorial
User's Guide IMEX
Defining Wells
Wells are defined using the following keywords.
*WELL
(Required)
*PRODUCER
(Required keywords which must follow well completion keywords.)
-or*INJECTOR
-or*SHUTIN
-or*OPEN
*IWELLBORE
(Optional. Use if you want to use the wellbore model for an injector.)
*PWELLBORE
(Optional. Use if you want to use the wellbore model for a producer.)
*INCOMP
(Required if you have defined any injectors. Keyword follows
*INJECTOR.)
*OPERATE
(Optional a BHP constraint is defaulted if omitted.)
*MONITOR
(Monitoring constraints are optional.)
*GEOMETRY
(Optional. It must precede a well completion keyword which is followed by
subkeyword *GEO. Cannot be used with unweighted injectors.)
*PERF
(At least one of these three or a combination thereof, is required.)
-or*PERFV
-or*PERFRG
These keywords must all reside in the Well and Recurrent Data section of your data set.
User's Guide IMEX
Tutorial 117
It is possible to define a well, name it, and specify its group affiliation with a *WELL
become active.
118 Tutorial
User's Guide IMEX
Defining the Well Type

There are five well types. They are:
a) *PRODUCER
b) *INJECTOR
c) *SHUTIN
d) *OPEN, and
e) *AUTODRILL
Each of these keywords must appear in the Well and Recurrent Data section and
*PRODUCER or *INJECTOR must be defined before the well may be operated.
When a well is defined using the *WELL keyword, it acquires the status *SHUTIN; when the
wells type is defined with *PRODUCER or *INJECTOR, it automatically acquires *OPEN
status. *SHUTIN can be specified for a well any time after it is defined with *WELL; *OPEN
can only be specified after the well has been typed with *PRODUCER or *INJECTOR.
A well can have its completions specified with *PERF before the type is specified, but it can
operate only after it has been typed as a producer or an injector.
Example:
*WELL 1 'MED RIVER P1' *VERT 1 1
*WELL 2 'MED RIVER P2' *VERT 15 15
*WELL 3 'MED RIVER I1' *VERT 5 5
*WELL 4 'MED RIVER I2' *VERT 10 10
*PRODUCER 1:2
...
** both wells 3 and 4 are mobility weighted
** injectors.
*INJECTOR 3:4 *MOBWEIGHT
...
*PERFV 1:2
** The producer wells 1 and 2 are completed
** through K layers 1 through 3, each having
** a well index of 1.65
1:3 1.65
** The injector wells 3 and 4 are completed
** through K layers 2 and 3, each having a
** well index of 1.87.
*PERFV 3:4
2:3 1.87
User's Guide IMEX
Tutorial 119
How to Shut In a Well and Reopen It

Before shutting in a well:
1. It must be fully defined with
a) the operating constraints and any monitoring constraints, and
b) the well completion keywords.
After being fully defined a well may be shut in at any time using a *TIME or *DATE
keyword. A well may be shut in immediately after it has been defined.
You may open a shut in well any time after it has been shut in. Be sure to use its well number
when you open or shut in a well.
*WELL 1 'WATER INJECTOR'
*WELL 2 'PRODUCER'
*WELL 3 'SOLVENT INJECTOR'
** All wells have the same geometry.
**
rad
geofac wfrac skin
*GEOMETRY *K 0.25 0.34
1.0
0.0
** Initially both injectors are shut in;
** only the producer is operational. True
** for the first two years.
*INJECTOR *MOBWEIGHT 1
*INCOMP
*WATER
*OPERATE *MAX *STW
12000.0
*OPERATE *MAX *BHP
10000.0
*PERF *GEO 1
** if jf kf ff
1 1 3
1.0
** Shut in well #1.
*SHUTIN 1
*PRODUCER 2
*OPERATE *MAX *STO 12000.0
*OPERATE *MIN *BHP 1000.0
*MONITOR *GOR
10000.0 *STOP
*MONITOR *WCUT 0.8330002 *STOP
*PERF *GEO 2
** if jf kf ff
7 7 1
1.0
*INCOMP *SOLVENT
*OPERATE *MAX *STS 1.2E+7
*OPERATE *MAX *BHP 10000.0
*PERF *GEO 3
** if jf kf ff
1 1 3
1.0
** Shut in well #3.
*SHUTIN 3
120 Tutorial
User's Guide IMEX
*TIME 730.000
*OPEN 1 ** Open water injector
*TIME 1095.0
** In this case a water injector and a
** solvent injector are defined in the same
** block; shut in the open one first before
** opening the second one, which was shut in.
*SHUTIN 1 ** Shut in water injector.
*OPEN 3
** Open solvent injector.
*TIME 1460.0
*SHUTIN 3 ** Shut in solvent injector
*OPEN 1
** open water injector
*TIME 1825.0
*SHUTIN 1 ** Shut in water injector.
*OPEN 3
User's Guide IMEX
Tutorial 121
Voidage Replacement
Voidage replacement option is initiated by specifying a group (gathering centre) using the
keywords *GROUP and *WELL.
*GROUP allows the assignment of a number of wells in a voidage replacement group,
whether they be producers or injectors. With each group, its own voidage replacement ratio
is assigned using the keywords:
*GCONI
-or*GCONI
'group-name'
*VREP
*GAS
vrep_frac
'group-name'
*VREP
*WATER
vrep_frac
Voidage may be replaced by injectors using the instantaneous injection potential or using
guide rates specified by the *GUIDEI keyword.
An example data set is given below:
*RUN
*DATE 1986 4 22
*DTWELL 1.00
** Define gathering centre for voidage replacement.
*GROUP 'g1' *ATTACHTO 'field'
** Define wells attached to the group 'g1'.
*WELL 1 'GAS INJECTOR' *ATTACHTO 'g1'
*WELL 2 'PRODUCER'
*ATTACHTO 'g1'
** Injector 1 matches the voidage from producer 2
*INCOMP *GAS
** Gas is being injected.
*OPERATE *MAX *STG 1.0E+8
*OPERATE *MAX *BHP 1.0E+8
*PRODUCER 2
*MONITOR *MIN *STO 1000.0 *STOP
*MONITOR *GOR
20000.0 *STOP
*PERF 1
**
if jf kf
wi
1
1
3 1.0E+5
** Well geometry for the producer.
**
rad geofac wfrac skin
*GEOMETRY *K 0.25 0.34 1.0 0.0
*PERF *GEO 2
** if jf kf ff
10 10
1 1.0
** Specify voidage replacement. Complete voidage
** is being replaced. Instantaneous injection
** potential is used to allocate injection
*GCONI 'g1'
*VREP *GAS 1.0
*IIP
*TIME 3650.000
*STOP
122 Tutorial
User's Guide IMEX
Using the Wellbore Model

To indicate that the wellbore model is required, you may use:
a) *IWELLBORE, and
b) *PWELLBORE
They are optional keywords and must be located in the Well and Recurrent Data section.
There are two occasions where you may want to use the wellbore model:
a) When you want to use the wellbore model to calculate the wellhead pressure, or
b) When you want to use the wellhead pressure as an operating constraint. You must
use the wellbore option.
One or both keywords may be used. *IWELLBORE indicates that you want the wellbore option
for an injector. *IWELLBORE must follow *INJECTOR immediately for identification. In
turn, *PWELLBORE indicates that you want the wellbore model for a given producer.
*PWELLBORE must immediately follow the *PRODUCER keyword for identification.
Example:
** The wellbore model is desired for well 2
** which is an injector.
*INJECTOR 2
*IWELLBORE
** wd
wl
rough whtemp rtemp
wr
1100. 1225. .1365
200.3
250.0 .33
Example:
** The wellbore model is desired for well 1
** which is a producer.
*PRODUCER 12
*PWELLBORE
** depth ibhp
1100.
1
TUBING DATA FOR PRODUCERS
Tubing data is required when *PWELLBORE is used. It must appear in the Well and
Recurrent Data section after *PWELLBORE has been specified AND after the well
definitions have been completed. For single-phase gas producers the analytical pressure loss
model can be used.
With this keyword, you input on a *PTUBE keyword:
a) *OIL or *LIQUID *WATER_GAS or *CONDENSATE
b) a table number and a reference depth
c) GOR, rates, water-cuts, (etc, see *PTUBE) and wellhead pressures.
Each of these are entered on separate lines with up to a maximum of 20 values for each.
Finally, a bottomhole pressure table is entered.
User's Guide IMEX
Tutorial 123
The bottomhole pressure table:

a) References the values you give for GOR, rate, water-cut, and wellhead pressure.
b) Each value is indexed and the indices are used to reference the value in your
bottomhole pressure table.
c) The indices are then aligned with a set of up to 20 bottomhole pressure values.
The number of bottomhole pressures must be equal to the number of wellhead pressures
already entered.
Example:
*PTUBE *OIL 1
*GOR
** ratio(1) ratio(2)
500.0
1000.0
*QO
** q (1) q(2)
q(3)
q(4)
0.0
4000.0 8000.0 16000.0
**
wcut(1) wcut(2)
*WCUT
0.00
0.50
**
whp(1) whp(2) whp(3)
*WHP
200.0
900.0 1600.0
*BHPT
**igor iwcut iqo
1
1
1
2
1
1
1
2
1
2
2
1
1
1
2
2
1
2
1
2
2
2
2
2
1
1
3
2
1
3
1
2
3
2
2
3
1
1
4
2
1
4
1
2
4
2
2
4
124 Tutorial
bhp(1)
2176.1
1997.7
2646.7
2447.7
2135.5
1682.7
2618.0
2189.0
2133.6
1463.1
2630.9
2022.0
2160.1
1425.7
2696.4
2080.0
bhp(2)
2873.7
2670.9
3344.7
3124.7
2876.6
2674.6
3351.2
3132.3
2884.2
2684.5
3368.4
3152.8
2912.5
2721.3
3433.4
3231.0
bhp(3)
3573.7
3370.9
4044.7
3824.7
3576.6
3374.6
4051.2
3832.3
3584.2
3384.5
4068.4
3852.8
3612.5
3421.3
4133.4
3931.0
User's Guide IMEX
Operating and Monitoring Constraints

*OPERATE and *MONITOR indicate the constraints on a given well. If no operating
constraint is entered for a well, a bottom hole pressure constraint of one atmosphere for
producers and 10,000 atmospheres for injectors is defaulted. Monitoring constraints are
optional.
Each well introduces a new unknown variable Pbh, the bottomhole pressure, into the
simulation, and a constraint equation is required to determine this variable.
The first operating constraint in a list of operating and monitoring constraints is the primary
operating constraint. The simulator at first attempts to operate on this primary constraint and
monitors the others in the list at the same time. If one of the other operating constraints is
violated and *CONT has been used, then this violated constraint becomes the current
If more than one operating constraint is violated, then the most drastic assigned action is taken:
Most drastic:
*STOP
*SHUTIN
Least drastic:
*CONT
PRODUCERS
For a producer you should operate

a) on a rate constraint (the primary operating constraint), and
b) on a minimum bottomhole pressure, or
c) on a minimum tubing head pressure.
If your producer is an oil well, pick an oil rate constraint. If your well produces gas, pick a
gas rate constraint. A subsequent constraint to use with a producer may be a minimum
pressure constraint.
Example:
*PRODUCER 1
*OPERATE *MAX *STO 12000.0 *CONT
*OPERATE *MIN *BHP 1500.0 *CONT
This example demonstrates:

a) the use of the oil rate as the primary constraint of this oil well, and
b) the subsequent constraint of bottomhole pressure.
The action to be taken if a violation occurs is to continue and switch the primary operating
constraint to the one that has just been violated.
*CONT is the default and need not be entered.
User's Guide IMEX
Tutorial 125
INJECTORS
For an injector, you would pick:

a) a maximum injection rate constraint for the primary operating constraint, and
b) a maximum bottomhole pressure constraint, or
c) a maximum tubing head pressure constraint.
If you are injecting gas, choose a gas rate constraint. If it's a water injector, choose a water
rate, etc.
Example:
*INJECTOR 2
*OPERATE *MAX *STW
*OPERATE *MAX *BHP
10000.0 *STOP
2250.0 *STOP
This example indicates:

a) the water rate for this water injector is the primary constraint, and
b) the bottomhole pressure is a secondary operating constraint which will be
monitored at the same time.
In both cases, if either are violated, the simulation will stop.
MONITORING CONSTRAINTS
The format of the monitoring constraint includes *MONITOR; then the constraint type, a value
is then required for all but backflow. Finally, the action taken if there is a violation. Again, the
most drastic action is taken when more than one constraint is violated at the same time.
It is highly recommended that you monitor GOR and water-cut in a producer; this may
prevent some problems during the run of your simulation job.
Example:
*PRODUCER 1
*OPERATE *MAX
*OPERATE *MIN
*MONITOR
*MONITOR
126 Tutorial
*STO
1200.0
*BHP
2500.0
*GOR 15000.0
*WCUT
.98
*CONT
*CONT
*STOP
*STOP
User's Guide IMEX
Definition of Well Indices within IMEX

For a Cartesian, vari or corner point grid block, given
*GEOMETRY *K rad geofac wfrac skin
*PERF *GEO wn
i j k ff
then for the perforation of the well in this block
WI =
2 permi permj h ff wfrac

re
ln + skin
rw
where permi and permj are the permeabilities for block (i,j,k) in the x and y directions
respectively; h = grid block length in the K direction,
re = geofac
di dj
wfrac
and di and dj are the grid block lengths in the I and J directions.
For the same geometry card, but replacing the *PERF and the following line with
*PERF *KH wn
i j k kh
then for the perforation of the well in this block
WI =
2 kh wfrac
re
ln + skin
rw
If
*PERF *WI wn
i j k wi
is used, no GEOMETRY is used and
WI = wi
For producers and mobility weighted injectors WI has units of md ft for field units and md
metres for SI. For unweighted oil or water injectors, the units are bbl / psi for field,
metres**3 / kPa in SI and metres**3/(kg/cm**2) in modified SI. For unweighted gas or
solvent injectors, the units are ft**3/psi for field, metres**3/kPa in SI and
metres**3/(kg/cm**2) in modified SI.
For mobility weighted injectors, WI is multiplied by the total (block) mobility.
For producers, WI is multiplied, for each phase, by the corresponding block phase mobility.
These then can be multiplied, for a nonbackflowing well, by the pressure in the wellbore (for
the current well element) - the block pressure to obtain a well rate at reservoir conditions.
The rate at surface conditions can be obtained dividing by the appropriate block volume
formation factor for producers and wellbore volume formation factor for injectors.
User's Guide IMEX
Tutorial 127
Note that the block kr's in the above calculations, optionally may be replaced by a well
element defined rock type set number and end points as described for the *PERF keyword.
For hybrid grids or cylindrical grids, the WI calculation is the same except re is given by the
steady-state solution radial direction node location in the innermost grid block (see the
description of the *GRID keyword for further details).
Printout of the well indices, as you mentioned, can be enabled using
*OUTPRN *WELL *ALL or
*OUTPRN *WELL *LAYER or
*OUTPRN *WELL *RESERVOIR
The header for this information is: Current Well Definitions.
The well indicies for producers is indicated by wp, for injectors by wi and are given for o
(oil), w (water), g (gas) and if it exists s (solvent).
Note: wi for injectors, as you mentioned, is nonzero only for the injecting phase if it is not
backflowing. Backflowing wells have WI's which are time dependent. See the *XFLOWMODEL keyword description for details.
128 Tutorial
User's Guide IMEX
Input of Well Indices

To input well indices, these keywords are used:
a) *GEOMETRY
b) *PERF, or
c) *PERFV, or
d) *PERFRG
These keywords must reside in the Well and Recurrent Data section. The well completion
keywords are required data, while *GEOMETRY is optional. *GEOMETRY may be used
with mobility weighted injectors or producers.
*GEOMETRY requires the necessary parameters to calculate the well indices internally. The
well completion keywords, *PERF, *PERFV, and *PERFRG require the location of the well
completion and the well index which you calculate.
If *GEOMETRY is used, then a well completion keyword is required with it. *GEO is used
with the well completion keyword to indicate that the well parameters have been entered.
*GEOMETRY always proceeds *PERF, *PERFV, and *PERFRG.
*PERF is ideal for horizontal or deviated wells, but may be used with vertical wells also. It
has the format:
Example:
*WELL 1 '12-09-18-56'
*PERF 1
** if jf kf
wi
setn
1
1
2:4 1.24 1
-or*WELL 1 '12-09-18-56'
**
rad
geofac wfrac skin
*GEOMETRY *K
.375 .2488
1.0
0.0
** The well completion keyword must follow
** the geometry keyword pertains to well 1.
*PERF *GEO 1
** if jf kf
ff
setn
1
1
2:4 1.
1
If *VERT was used with *WELL, then you have specified a vertical well. Use *PERFV.
Only the K direction grid block or range of blocks need be entered, since you have already
entered the I and J location with *VERT. If you are using *GEOMETRY, use *GEO with
*PERFV.
User's Guide IMEX
Tutorial 129
Example:
*WELL 2 *VERT 2 2
*PERFV 1
** kf
wi
setn
2:4 1.56
1
If you are using refined grid blocks and wells are located within the vicinity, then *PERFRG
must be used. *GEO is again required if *GEOMETRY is used. *PERFRG requires the
location of the fundamental grid block(s) where the well is completed and the location of the
refined grid block(s) where the well is completed.
Example:
...
** Refinement will result in creating 3
** refined grids in the I direction, 3 in the
** J direction and two in the K direction in
** block (1,1,3).
*REFINE 3 3 2
*RANGE 1 1 3
...
*WELL 1
*PERFRG 1
** if jf kf ir jr kr
wi
setn
1
1
3
2
2
1:2 1.75
1
130 Tutorial
User's Guide IMEX
Stopping a Simulation Run

You must ALWAYS use the keyword *STOP to terminate the simulation run. *STOP only
appears once in your data set; it appears as the last line of the data set in the Well and
Recurrent Data section. A stop time or stop date is always given with it.
Example:
*DATE 1998 09 08
*STOP
User's Guide IMEX
Tutorial 131
Using Wildcards in Well Lists

A significant feature has been added to all keywords which allow the user to enter well lists
(e.g. *ALTER 'well1' 'well2'). Wild cards have been incorporated in the following manner.
* replaces any number of characters at the end of a well name or can be used on its own to
represent all wells (e.g. *ALTER '*' or *ALTER 'wel*').
? replaces any single character anywhere in the well name (e.g. *ALTER '?el?1').
The use of '*' followed by a character is not allowed (e.g. *ALTER 'we*l1 results in an error)
The well name or wildcarded name must appear in quotes.
For example:
If the wells in the model are 'well1' 'well2' and 'prod1', then:
'*'
would list all 3
'p*'
would list 'prod1'
'?ell?'
would list 'well1' and 'well2'
'?ell?*'
would list 'well1' and 'well2'
'?ell??'
would not match any well
132 Tutorial
User's Guide IMEX
Guidelines for Setting Up Well Data

The following guide assists you with using the Well and Recurrent Data section of this
document. When entering the well data for the first time in your data set, the following
information must be present in this order:
Only one of each of the following keywords can appear with any one *TIME or *DATE. In
other words you cannot have two sets of *AIMSET, two sets of *AIMWELL, etc. under one
*TIME or *DATE.
1. Either *TIME or *DATE is required.
2. Define a value for *DTWELL, which is the first timestep size used immediately
after the well is defined.
3. Identify all new groups using *GROUP. A hierarchy may be built using the
*ATTACHTO keywords. This step is optional, but if groups are to be used in the
run they must be defined before the wells are defined.
4. Identify all new wells using *WELL. *ATTACHTO may be used to define a
group if necessary.
5. Indicate the well locations, geometries, or the well indices using *GEOMETRY
and any of the well completion keywords (*PERF, *PERFV, or *PERFRG). This
may be entered at a time later than that at which *WELL is entered for the well.
6a. Define a new well or a well with major operating changes as *PRODUCER and
*INJECTOR. This may be done at a *TIME later than that at which the *PERF
lines were entered for the well.
6b. Indicate the need for the pressure loss in wellbore model (tubing flow curves).
You must have it if you choose *WHP as one of the operating constraints.
6c. Define the operating or monitoring constraints for that well.
Steps 1 through 6 MUST appear in any data set, although a bottom-hole pressure operating
constraint will be defaulted if none is entered. Step 3 is optional. Step 5 may follow step 6,
but if so the PERF lines must be entered at the same *TIME as the well type information.
7. A well is defaulted as shut when *WELL is entered for it and may be explicitly
shut in after *WELL has defined the well.
8. Use *OPEN to reopen a previously shut in well. *OPEN can only be entered for a
well after the type has been defined with *PRODUCER or *INJECTOR.
9. Be aware that different keywords are required depending on what options you are
using.
Subsequent well changes at different times or dates are done with steps 10 through 16:
10. Define new wells. Use steps 1,4,5, and 6 before adjusting the parameters of
existing wells.
11. You may alter the primary operating constraint of any well with *ALTER, once
the wells type has been defined with *PRODUCER or *INJECTOR. Use with
*TIME or *DATE.
User's Guide IMEX
Tutorial 133
12. You may alter the polymer concentration if using the polymer concentration with
*ALTERCP, once the well has been defined as a water injector. Use with *TIME
or *DATE.
Only one of each of the following keywords can appear with any one *TIME or *DATE. In
other words you cannot have two sets of *AIMSET, two sets of *AIMWELL, etc. under one
*TIME or *DATE.
13. You may adjust the implicitness/explicitness as required (*AIMSET, *AIMWELL).
14. You may adjust the input/output controls and the transmissibility multipliers as
required.
15. The keywords *DTWELL, *DTMAX, *DTMIN may also appear in subsequent
well changes.
16. You MUST always terminate the simulation run with *STOP.
Assume you have a 10 x 10 x 12 Cartesian system. A section of the well data may look like this:
** Well data
*RUN
*DTWELL
0.025
*WELL 1 'OIL PRODUCER'
*PRODUCER 1
** Operating and monitoring constraints for the
** producer only.
*MONITOR *BACKFLOW *STOP
**
rad
geofac wfrac skin
1.0
0.0
*PERF *GEO 1
**if jf kf ff
1
1
3
1
*TIME 10.0
** Change the primary constraint value at 10.0
** days.
*ALTER 1
100.0
*TIME 50.0
*ALTER 1
1000.0
** Define an injector. Also modify the primary
** operating constraint for the producer again.
*TIME 720.0
*INJECTOR 2 *MOBWEIGHT
*INCOMP *WATER
*OPERATE *MAX *STW 12000.0
**
rad
geofac wfrac skin
1.0
0.0
*PERF *GEO 2
**if jf kf ff
7 7 3 1
*ALTER 2
100.0
** Stop at 10 years
*TIME 3650.0
*STOP
134 Tutorial
User's Guide IMEX
Recurrent Data from Other Sections

There are a few keywords from the other sections whose parameters may be changed during
the course of a simulation run. These keywords must be positioned after a *DATE or *TIME
keyword in the Well and Recurrent Data section. At a given time or date the older or existent
parameters will be overridden by the new ones.
From the Input/Output Control section, these keywords may be reassigned:
a) *WRST
b) *WPRN
c) *OUTPRN
d) *WSRF, and
e) *OUTSRF
From the Reservoir Description section, the refined grid definition keywords and the
transmissibility modifiers may appear in the Well and Recurrent Data section:
a) *RANGE
b) *REFINE
c) *TRANSI (*TRANLI)
d) *TRANSJ (*TRANLJ)
e) *TRANSK (*TRANLK)
From the Rock-Fluid Data section, only the rock type array and the end point arrays may
appear in the Well and Recurrent Data section:
a) *RTYPE
b) *SWCON
c) *SWCRIT
d) *SGCON
e) *SGCRIT
f) *SOIRW
g) *SORW
h) *SLCON
i)
*SORG
j)
*KRWIRO
k) *KROCW
l)
*KRGCL
m) *KROGCG
n) *PCWMAX
o) PCGMAX
User's Guide IMEX
Tutorial 135
p) JFWMAX
q) JFGMAX
From the Numerical Methods Control section, these keywords may be changed:
a) *DTMAX, and
b) *DTMIN
136 Tutorial
User's Guide IMEX
Subsidence Output
In IMEX, keyword *SBDZ (for subsidence) has been added to the grid output list to display
the vertical deformation of grid blocks. The formula used in IMEX is similar to the one that
was implemented in STARS. It is available as a *GRID option for both output files and SR2
files as well as being available as a *SPECIAL variable
It is important to note that the subsidence calculated does not alter reservoir volume or flow
characteristics and is used only for display purposes.
The process of development of the formula for subsidence output is shown as follows:
V =
(1)
dVb
Vb0
Where:
V is volumetric strain.
Vb0 is initial bulk volume of a block.
The equation (1) can be written as:
V =
Vbn Vb0
dVb
=
Vb0
Vb0
(2)
Assuming that only pore volume of a block is changing with pressure and the matrix portion
of the block is constant. The above equation (2) can be rewritten as:
V
(V
=
n
p
) (
Vmatrix Vp0 + Vmatrix

Vb0
) = (V
n
p
Vp0
Vb0
(3)
Where:
Vpn
Vp0
Vmatrix
=
=
=
Pore volume of a block at k or n iteration

Initial pore volume of a block
Matrix volume of a block
However,
Vpn = Vb0 n
(4)
Vp0 = Vb0 n
(5)
Substitute (3) and (4) into (2) and simplify to give:
V = ( n 0 )
(6)
Where:
n = Porosity at k or n iteration
0 = Initial porosity.
User's Guide IMEX
Tutorial 137
Furthermore, assuming that deformation does not occur in the X and Y direction. The crosssectional area parallel to the XY plane is, thus, constant. Based upon this assumption,
equation (2) can be written as:
V =
(V
n
b
Vb0
Vb0
) = (Ah
Ah 0
Ah
) = h
h0
(7 )
Where:
A is cross section area that is parallels to the XY plane.
hn is thickness of the block at k or n iteration.
h0 is initial thickness of the block.
Comparing equation (6) and equation (7), this will lead to:
h = h 0 ( n 0 )
(8)
Equation 8 is used to calculate subsidence for each block.

The total vertical subsidence (i.e. the sum of the subsidence of each vertical layer) is also
calculated when *OUTSRF *GRID *SBDZ is requested and can be displayed using
Results3D.
138 Tutorial
User's Guide IMEX
Parallel IMEX
If you are running on a pc using launcher, then to run in parallel you may modify the
simulator icon and check the box Run simulator in multi-processor mode, and for a two
processor machine, enter 2 in the box following Number of processors to use. This is
equivalent to using the command line options -doms parasol 2.
The command line option -doms is equivalent to using *DPLANES in the numerical
methods section.
The command line option parasol n behaviour has changed to being equivalent to using:
*SOLVER *PARASOL
*PPATTERN *AUTOPSLAB n
*PNTHRDS n
*PPATTERN *AUTOPSLAB n has been tested for n up to 64 corresponding to 64
processors. If an n is specified which is greater than the logical number of CPUs on the
machine, then *PPATTERN *AUTOPSLAB n will be used along with PNTHRDS m
where m is the minimum of n and the number of logical CPUs on the machine.
Thus, if you wish to use keywords for the parallel options instead of the command line
switches, the base set of keywords for running the simulator in parallel is:
*DPLANES
*SOLVER *PARASOL
*PNTHRDS n
Where n is the desired number of logical CPUs to be used.
There are a number of techniques for improving solver performance in PARASOL runs.
Setting PPATTERN 2 or PPATTERN 3 for 2 or 4 logical CPUs may work as well (in
some cases better) than the corresponding *PPATTERN *AUTOPSLAB n.
Setting *CHECKRB to *CHECKRB *ON will avoid the use of the red-black system
reduction when there are too few red cells to make the reduction worthwhile.
If there is a significant number of solver failures in the run, as reported at the end of the .log
file, then increasing *NORTH and *ITERMAX often helps, such as
*ITERMAX
*NORTH
80
80
Another technique to improve solver convergence is to try a higher linear solver degree such
as SEDGREE 2
If you are specifying *SDEGREE greater than 1 (the default) then we recommend that you
also use, in the Numerical Methods section:
*REORDER-CBDY *ON
This tends to reduce interclass induced fill.
User's Guide IMEX
Tutorial 139
The above are the techniques we have found most frequently to improve PARASOLs
performance over an initial run with defaulted settings. While the full set of PARASOL
controlling keywords allows more detailed control than contained in the above, these
constitute a good starting point.
In order obtain timing diagnostics, you may put in the Input/Output Control section:
*DEBUG *CPUTIME *SOLVER
This will provide a summary at the end of the log and out file giving a breakdown of elapsed
(wall clock ) and CPU times for different portions of the run. Equivalently, this option may
be enabled by using the command-line argument -cputime
140 Tutorial
User's Guide IMEX
Using the Seawater Injection Option

In IMEX, seawater can be injected into the reservoir. Injected seawater is treated as an active
tracer in the water phase. Combined with the well scale buildup option, seawater tracking
allows the user to model perforation damage due to the scale deposition caused by seawater
mixing with formation brine. When the seawater injection option is in use, the term water
throughout the simulation denotes the total water which includes the formation water and the
seawater.
There are two reservoir models which use the seawater injection option. Both the
*BLACKOIL and *OILWATER can model seawater injection by specifying *MODEL
*BLACKOIL_SEAWATER or *MODEL *OILWATER_SEAWATER respectively. These
two keywords must be declared in Component Properties section. Additional keywords can
optionally be used as below.
1. Input/Output Control section
Use the sub-keyword *SEAWF in the grid_list of *OUTPRN/*OUTSRF *GRID
to output the distribution of seawater volume fraction during the run.
2. Component Properties section
Use *SVISC to define the seawater viscosity if applicable. By default, the formation
water viscosity of PVT Region 1 will be used.
3. Initial Conditions section
*SEAWATFRC can be used to initialize the reservoir seawater volume fraction as a
user defined distribution. The default initial seawater fraction is zero. It is normally
not necessary to define this as initially only formation water exists in the reservoir.
4. Numerical Method Control section
Use *CONVERGE *MAXRES *SEAWATER to change the default convergence
criteria if necessary. This keyword is normally not required.
5. Well and Recurrent Data section
Use *INCOMP *SEAWATER when defining an injector to indicate the injection
of seawater. By default the seawater is 100% of the injection stream. However, a
fraction following this keyword, if given, can be used to define the volume fraction
of seawater injected. The following example indicates that of the 5000 bbls/day of
injection water, 80% is seawater and remainder is formation water.
*INCOMP *SEAWATER 0.8
... *OPERATE *MAX *STW 5000.0
The volume fraction of injected seawater can be altered by using *ALTERFS in a

manner analogous to *ALTERCP.
User's Guide IMEX
Tutorial 141
Using the Scale Buildup Option

Scale deposition is a common problem in hydrocarbon reservoirs where injected seawater mixes
with formation brine. The scale buildup option in IMEX allows the user to model well damage (a
reduction of productivity) due to the scale deposited on individual perforations of a well.
The well damage can either be modeled as a function of both water production and the fraction of
seawater produced in the water or as a function of water production alone and a fixed scale
deposition rate per volume of water produced (if the seawater injection option is not used).
In IMEX, the mass of scale deposited within a time step is calculated as the volume of water
which has flowed through the perforation during the time step multiplied by a corresponding
precipitation. The precipitation is a function of the seawater fraction, which is given through
a user defined deposition table. The accumulated scale causes damage to the well PI. The
relationship between the PI damage factor and the existing scale is also given through a user
defined damage table.
Modeling of scale buildup as a function of both seawater fraction and water production
requires four steps.
1. Use one of the two seawater injection models to enable seawater tracking in the
reservoir. Please see Using the Seawater Injection Options.
2. In the Rock-Fluid Data section, define the scale deposition tables and well damage
tables by using *SCLDPS and *SCLDMG respectively. Multiple tables can be
specified.
3. In the Well and Recurrent Data section, assign both the deposition and damage
tables to all perforations (layers) in a well or specific perforations (layers) of a
well. This is done by using the keyword *SCLTBL-WELL or *SCLTBL-LAYER.
The *SCLTBL-WELL keyword is applied to all perforations of the well(s) and the
*SCLTBL-LAYER keyword is applied to specified perforations of a well.
4. In the Well and Recurrent Data section define seawater injection wells rather than
formation water injection wells using: *INCOMP *SEAWATER
Perforations (layers) linked to deposition and damage tables will keep track of scale
deposition at a perforation over a time step, keep track of the total scale deposited on a
perforation, and calculate damage based on total scale deposited/length of perforation.
Within the recurrent data, connections can be reassigned to different tables. Deposited scale
accumulates until it is totally or partly be removed. Scale removal is performed by using the
keyword *SCLRMV-WELL or *SCLRMV-LAYER. Again the *SCLRMV-WELL keyword
is applied to all perforations of the well(s) while the *SCLRMV-LAYER keyword is applied
to the specified perforations of a well.
If the scale buildup option is to be used without the seawater injection option, scale precipitation
is no longer considered as a function of seawater fraction. The scale buildup option can be used
without seawater tracking but damage becomes a function only of the amount of water produced.
Scale buildup/well damage can be modeled using the following steps:
142 Tutorial
User's Guide IMEX
1. Use any PVT model.

2. In Rock-Fluid Data section, define the deposition tables (*SCLDPS) with a zero
seawater fraction row and constant scale precipitation. For example,
*SCLDPS
**Seawater fraction
0.00
1.00
Scale precipitation (lbm/ft3)

1.10e-05
1.10e-05
Since seawater flow in not modeled when the seawater injection option is not used,
the seawater fraction in the reservoir is always zero. Therefore, in the above table,
only the first entry (at zero volume percent seawater) is ever used, and hence a
constant precipitation of 1.10e-05 pounds of scale deposited per cubic foot of
produced water is employed. The damage tables (*SCLDMG) can be defined
normally.
3. In the Well and Recurrent Data section, scale table assignment and scale removal
actions can be defined normally.
User's Guide IMEX
Tutorial 143
Simulating a Volatile Oil Reservoir

To simulate a volatile oil reservoir:
1. Use *MODEL *VOLATILE_OIL.
2. Enter saturated gas and oil phase PVT properties either using:
The *PVTCOND table, optionally gas properties for undersaturated gas

may be entered using the *EGUST (*BGUST or *ZGUST) and
*VGUST tables.
The *PVTVO table, *EGUST (*BGUST or *ZGUST) and *VGUST

cannot be entered, as both wet and dry (undersaturated) gas properties
are entered in the *PVTVO table format.
3. Enter all oil phase properties which are required for pressures above the bubble
point, such as *CO, *BOT, *COT, *CVO, *VOT.
These keywords must be located in the Fluid Component Data section of the data set.
4. Establish the gas-oil and oil-water contacts as if initializing a black oil model
(*MODEL *BLACKOIL).
5
Define the initial Bubble Point using the *PB or *PBT keywords.
6. Define the initial Dew Point using the *PDEW or *PDEWT keywords.
7. Define all other initial conditions as if this were a black oil model.
These keywords must be located in the Initial Conditions Data section of the data set.
8. Enable output options which aid in model analysis with the following steps:
a) *OUTPRN/*OUTSRF *GRID *RV *DPP outputs Oil Content and Dew
Point Pressure to the output and SR2.
c) *PSPLIT *ON adds well production split information to the well output
(Output file and SR2 file). The surface oil production is made up of oil
from condensate and black oil. The surface gas production is made up of
free gas and solution gas.
These keywords must be located in the Input Output Control section of the data set.
9. Enable numerical control options if necessary:
a) *NORM *PDW allows you to control timestep size based on dew point
pressure changes. *NORM PBUB and *NORM PDW both set a single
timestep control parameter which control the average timestep size as a
function of bubble point pressure and/or dew point pressure change.
b) *MAXCHANGE *PDW allows you to control maximum timestep size
allowed based on dew point pressure changes. *MAXCHANGE PBUB
and * MAXCHANGE PDW both set a single timestep control parameter
which control the maximum timestep size as a function of bubble point
pressure and/or dew point pressure change.
These keywords must be located in the Numerical Methods Control section
of the data set.
144 Tutorial
User's Guide IMEX
Using the Subdomain-Dual Permeability-Reinfiltration Model

The SD-DK-R model is controlled by the following keywords:
*SUBDOMAIN idiv (turns on subdomain dual porosity model, each subdomain
consists of idiv vertical blocks) Reservoir Description section.
*FRACVOL (controls the thickness of individual subdomain divisions)
Reservoir Description section
*TRANSD (controls SD-DK-R inter-subdomain connection) Reservoir
Description section
*SD_REINF (determines if SD-DK-R inter-subdomain connection is used to
represent reinfiltration) Rock-Fluid Data section
The SD-DK-R is capable of modeling a wide variety of dual porosity scenarios
based on TRANSD and SD_REINF input values.
In general the number of SUBDOMAIN divisions should be set to 5. It is acceptable to use
more divisions if CPU time and memory are not constraining factors.
In addition, FRACVOL should be used to reduce the thickness of both the top and bottom
subdomain divisions. In the case of *SUBDOMAIN 5, a FRACVOL setting of:
*FRACVOL 0.02 0.32 0.32 0.32 0.02 minimizes the error in capillary holdup.
Modeling Capability of SD-DK Model
1. Standard Subdomain Model

If the *TRANSD keyword is not included in the model *SD_REINF is not
required and the model reverts to the standard *SUBDOMAIN model. It is still an
excellent idea to use *FRACVOL to minimize the error in the capillary holdup.
2. Gravity Drainage between Subdomains
Using the *TRANSD keyword to set non-zero transmissibility multipliers between
domains allows the fluid in the bottom of a subdomain to flow into the top of the
subdomain directly below it without having to flow into the fracture first. In
essence this is a hybrid method which combines the Subdomain model and the
Dual Permeability model and gives similar results. As there is no capillary holdup
in the matrix, *FRACVOL use does not significantly improve the models results.
3. Reinfiltration between Subdomains
As in 2) the non-zero *TRANSD keyword enables inter-subdomain flow. The use
of *FRACVOL and *SD_REINF together allows the user to use the bottom-most
subdomain block as a proxy for a horizontal fracture. The use of *FRACVOL to
reduce the thickness of the bottom division of the subdomain reduces the amount
of fluid stored in the fracture proxy block, and minimizes the error in capillary
holdup. *SD_REINF = 1.0 sets the bottom subdomain divisions capillary pressure
to 0.0 (the fractures value) and introduces a real capillary discontinuity in the flow
between the two subdomains.
User's Guide IMEX
Tutorial 145
4. Gravity Drainage and Reinfiltration between Subdomains

Actual drainage from Subdomain blocks is a combination of gravity drainage and
reinfiltration. This may be approximately accounted for by scaling back the bottom
subdomain divisions capillary pressure rather than setting it explicitly to zero.
This can be accomplished by setting *SD_REINF to a value between 0.0 and 1.0.
The fracture proxy blocks capillary pressure is scaled by 1-*SD_REINF.
146 Tutorial
User's Guide IMEX
Modelling Non Darcy Flow in Hydraulic Fractures Accurately

Using a Grid Based Approach
It is possible to use the *NONDARCY *GENERAL model to accurately model the non
Darcy response of a fracture within a simulation using a pseudo fracture. This requires the
user to generate a reasonably fine grid representation of the fracture. However we have
developed techniques which allow the 0.10 inch wide fracture to be represented by grid
blocks as large as 1 foot wide.
The keys to the proper modelling of non Darcy flow in the 1 foot wide fracture are
1. Adequate representation of flow around the fracture. By this we mean a fine grid
surrounding the fracture.
2. The use of an effective fracture permeability (Kfeff) to ensure that the flow in the
fracture is properly accounted for.
3. The use of a properly calculated non Darcy correction term (*NDARCYCOR)
which corrects the Forchheimer number in the fracture for the effective fracture
permeability used in (2).
This discussion will assume we are modelling a single 500 ft ( length) vertical fracture in a
0.10 mD gas water reservoir. The fracture permeability (Kf) has been calculated as 10,000
mD and the fracture width is 0.10 inch. The fracture is assumed to extend from the top to the
bottom of the pay.
As mentioned above, the fracture is going to be modeled using a number of 1 foot wide
blocks whose total length is 1000 feet. When setting up this grid it is crucial that the well
block pressure approximately equals the grid block pressure. In order to do this we represent
the intersection of the well and the vertical fracture by a 1 foot by 1 foot grid block. The well
radius is not critical, except that it must be close enough to the grid blocks effective
(Peaceman) radius to allow the well pressure to be nearly the same as the grid block pressure.
For a 1 foot by 1 foot block, a well radius of 1/8th of a foot will accomplish this.
Proper representation of pressure in the reservoir is important, so the blocks surrounding the
fracture should vary in a logarithmic fashion away from the fracture both in the direction
parallel and perpendicular to the fracture. There should be a detailed discretization of the
area next to the fracture which is able to capture the pressure field surrounding the fracture in
a reasonable fashion.
Fracture Pseudoization
As the fracture is being modeled as a number of 1 foot wide blocks, the flow down the
fracture must be pseudoized in order produce results similar to the flow down a 0.10 inch
wide fracture.
The fracture Kf x A should be identical in the simulation model using a 1 foot fracture wide
to the simulation model using a 0.1 inch wide fracture.
Therefore the Kfeff used in this model is reduced by the ratio of the models fracture area to
the actual fracture area, which in this case is equal to the ratio of model fracture width to the
actual fracture width (1/0.00833). This gives us a fracture permeability of 83.33 mD.
User's Guide IMEX
Tutorial 147
The use of the effective fracture width corrects Darcy flow down the fracture, but does not
properly correct the Forchheimer number in the fracture to account for the reduced fracture
permeability. In order to accomplish this, the NDARCYCOR correction term must be
defined in the blocks which represent the fracture (only). NDARCYCOR outside of the
actual fracture may be set to 1.0 to model non Darcy flow outside the fracture blocks or to 0.0
to ignore non Darcy flow outside the fracture blocks. Outside the fracture, non Darcy effects
are normally smaller in magnitude.
In the fracture blocks:
The value of NDARCYCOR = (Kf/Kfeff) (2-N1g)
N1g is the exponent of the (Krg x K) term in the correlation for factor. The factor is used
in the Forchheimer equation (see keyword *NONDARCY).
Assuming we are using Geertsmas model for , which uses N1g = 0.5, we obtain:
NDARCYCOR = 10000/83.33(2-N1g) = 10000/83.331.5 = 1314.6.
If another correlation were to be used, for example, where N1g = 1.5, NDARCYCOR would
be 10000/83.330.5 (= 10.954). The value of the correction is not important, as long as the
value of N1g used in the calculation of NDARCYCOR is the same N1 used in the
*NONDARCY *GENERAL model.
This example assumes the fracture width and/or fracture permeability does not vary in the
fracture. If this were to be modeled, each block would have its own Kfeff and NDARCYCOR
to account for varying Kf and fracture width.
In addition there should be a correction factor for each phase (if N1 for different phases differed),
but normally identical factor correlations are used for each phase, so this is not an issue.
The combination of the effective fracture permeability (to preserve Kf x A in the fracture) and
the NDARCYCOR factor as calculated above (to correct the Forchheimer number for Kfeff)
will produce a pressure drop in the coarse fracture model which is nearly identical to the one
produced using a 0.10 inch fracture and that will run significantly faster.
LGR Fracture Representation in a Variable Thickness/Depth Grid
In a variable thickness/depth grid it is very convenient to represent the fracture by LGR

(Local Grid Refinement). The discussion of proper representation of pressure (above) still
applies, but the use of LGR is a flexible way to model one or more hydraulic fractures
without increasing overall grid dimensions significantly.
Case 1 - LGR to Model a Fracture whose Length is equal to the Parent Block Length
Assume for this example that the fracture is parallel to the I direction.
The fracture is represented by a single row of refined blocks in the I direction in

the center of the parent block.
Assume that the refined block in the fracture which connects the fracture to a sink
(well) is a 1 ft. x 1 ft. block in the center of the refined grid and hence in the center
of the fracture.
The number of grid blocks used in the refined grid should be able to adequately
represent the pressure drop away from the fracture and should be odd in both the I
and J directions in order to center the fracture in the cell.
148 Tutorial
User's Guide IMEX
In this example we will use a 21 x 17 block refined grid to model the 200 ft long fracture
within a 200 x 150 ft parent block.
In the I direction, where Di (parent block) = 200.0, Di (iwell) = 1.0
Determine the center refinement:
If NLGRi = 21, the central refinement = Int (NLGRi/2) + 1 = 11
This sets the dimension in the I direction of the well block and its location in the I direction
within the LGR
Di (iwell) = Di (11)
The fracture is symmetric about block 11 in the I direction so
Di (iwell+1) = Di (iwell-1), Di (iwell+2) = Di (iwell-2),
Di (iwell+3) = Di (iwell-3), etc
or
Di (12) = Di (10), Di(13) = Di (9), Di(14) = Di (8), etc.
The LGR block sizes are all based on the size of the well block (Di (iwell)) so that
Di (iwell+I) = Di (iwell) x (RI R (I-1)) where I = 1, Int (NLGRi/2)
and
Di (iwell-I) = Di (iwell) x (RI - R (I-1)) where I = 1, Int (NLGRi/2)
Where R = (Di (Parent Block) / Di (iwell)) (1/ (Int (NLGRi/2)))
For the J direction a similar calculation is done:
In the J direction, where Dj (parent block) = 150.0, Dj (jwell) = 1.0
Determine the center refinement:
If NLGRj = 17, the central refinement = Int (NLGRj/2) + 1 = 9
This sets the dimension in the j direction of the well block and its location in the j direction
within the LGR
Dj (jwell) = Dj (9)
The fracture is symmetric about block 9 in the j direction so
Dj (jwell+1) = Dj (jwell-1), Dj (jwell+2) = Dj (jwell-2),
Dj (jwell+3) = Dj (jwell-3), etc
or:
Dj (10) = Dj (8), Dj (11) = Dj (7), Dj (12) = Dj (6), etc.
The LGR block sizes are all based on the size of the well block (Dj (jwell)) so that
Dj (jwell+J) = Dj (jwell) x (RJ R (J-1)) where J = 1, Int (NLGRj/2)
and
Dj (jwell-J) = Dj (jwell) x (RJ - R (J-1)) where J = 1, Int (NLGRj/2)
Where R = (Dj (Parent Block) / Dj (jwell)) (1/ (Int (NLGRj/2)))
User's Guide IMEX
Tutorial 149
An example follows:
Case 2 - LGR to Model a Fracture whose Length is longer than a Parent Blocks Length
In this case we will assume that the fracture length is exactly equal to 3 parent blocks length
in the I direction. The J direction LGR discretization is calculated using the same procedure
for all 3 parent blocks. The central blocks I direction discretization remains unchanged.
The blocks on either side of the central parent block in the I direction (as mentioned) above
have the same J direction discretization as the central parent block (NLGRj = 17). However
it is only necessary to ensure that the I direction discretization produces LGR blocks of equal
size in the I direction whose I direction sizes are approximately equal to the largest LGR Di
calculated for the central parent block. This will vary with problem size but normally should
not require NLGRi to be larger than 5 for the two blocks representing the ends of the fracture.
An example follows.
Example - No LGR Grid - Partial Dataset (Fracture is Parallel to the I Direction)
** Reservoir Description section
** Fracture begins at the first 70 foot DI block and ends at the last
70 foot DI block
DI IVAR 20*200 200 150 100 6*70.0 39.5 20.0 10.0 5.0 2.25 1.50 1.25
1.0 1.25 1.50 2.25 5.0 10.0 20.0 39.5 6*70.0 100 150 200 20*200
DJ JVAR 20*200 150.0 100.0 70.0 40.0 20.0 10.0 5.0 2.25 1.50 1.25 1.0
1.25 1.50 2.25 5.0 10.0 20.0 40.0 70.0 100.0 150.0 20*200
** Permeability in the fracture is set to Kfeff
PERMI CON 0.10
MOD 24:50 31 1 = 83.33
PERMJ EQUALSI
PERMK EQUALSI
** Rock Fluid Property section
** Use the Geertsma Beta factor
NONDARCY GENERAL 0.50
48511.34 0.5 5.5
10000.0
48511.34 0.5 5.5
10000.0
48511.34 0.5 5.5
10000.0
** Set the pseudoized fracture non Darcy flow correction factor (in
the fracture blocks)
NDARCYCOR CON 1 ** (or 0.0)
MOD 24:50 31 1 = 1314.6
150 Tutorial
User's Guide IMEX
Example - LGR Grid from Case 1 - Partial Dataset (200 ft Fracture in 1 block)
REFINE 26 26 1 INTO 21 17 1
** In the I direction, R = (200.0/1.0)** 1/ (Int (NLGRi/2))
**
(200.0/1.0)**0.10 = 1.69864665
**DI(12) = DI(10) = 0.5*(1.69864665-1.0) = 0.34932
**DI(13) = DI(9) = 0.5*(1.69864665**2 - 1.69864665)
=
**DI(14) = DI(8) = 0.5*(1.69864665**3 - 1.69864665**2) =
**DI(15) = DI(7) = 0.5*(1.69864665**4 - 1.69864665**3) =
**DI(16) = DI(6) = 0.5*(1.69864665**5 - 1.69864665**4) =
**DI(17) = DI(5) = 0.5*(1.69864665**6 - 1.69864665**5) =
**DI(18) = DI(4) = 0.5*(1.69864665**7 - 1.69864665**6) =
**DI(19) = DI(3) = 0.5*(1.69864665**8 - 1.69864665**7) =
**DI(20) = DI(2) = 0.5*(1.69864665**9 - 1.69864665**8) =
**DI(21) = DI(1) = 0.5*(1.69864665**10 - 1.69864665**9) =
0.59337
1.00793
1.71213
2.90830
4.94017
8.39161
14.2544
24.2132
41.1296
DI IVAR 51*200
DI RG 26 26 1 IVAR 41.1296 24.2132 14.2544 8.39161 4.94017 2.90830
1.71213 1.00793 0.59337 0.34932 1.0 0.34932 0.59337
1.00793 1.71213 2.90830 4.94017 8.39161 14.2544
24.2132 41.1296
** In the J direction, R = (150.0/1.0)** 1/ (Int (NLGRj/2))
**
(150.0/1.0)**0.125 = 1.870731277
**DJ(10) = DJ(8) = 0.5*(1.870731277-1.0)
=
**DJ(11) = DJ(7) = 0.5*(1.870731277**2 - 1.870731277)
=
**DJ(12) = DJ(6) = 0.5*(1.870731277**3 - 1.870731277**2) =
**DJ(13) = DJ(5) = 0.5*(1.870731277**4 - 1.870731277**3) =
**DJ(14) = DJ(4) = 0.5*(1.870731277**5 - 1.870731277**4) =
**DJ(15) = DJ(3) = 0.5*(1.870731277**6 - 1.870731277**5) =
**DJ(16) = DJ(2) = 0.5*(1.870731277**7 - 1.870731277**6) =
**DJ(17) = DJ(1) = 0.5*(1.870731277**8 - 1.870731277**7) =
DJ JVAR 51*150
DJ RG 26 26 1 JVAR 34.9087 18.6605
0.81445 0.43536
5.33212 9.97496
** Permeability in the fracture is
PERMI CON 0.10
PERMI RG 26 26 1 JVAR 8*0.10 83.33
PERMJ EQUALSI
PERMK EQUALSI
48511.34 0.5 5.5
10000.0
48511.34 0.5 5.5
10000.0
48511.34 0.5 5.5
10000.0
0.43536
0.81445
1.52362
2.85029
5.33212
9.97496
18.6605
34.9087
9.97496 5.33212 2.85029 1.52362

1.00.43536 0.814451.52362 2.85029
18.6605 34.9087
set to Kfeff
8*0.10
** Set the pseudoized fracture non Darcy flow correction factor (in the fracture
blocks)
NDARCYCOR CON
0
NDARCYCOR RG 26 26 1 JVAR 8*0 1314.6 8*0
User's Guide IMEX
Tutorial 151
Example - LGR Grid from Case 2 - Partial Dataset (600 ft Fracture in 3 Blocks)
REFINE 26 26 1 INTO 21 17 1
REFINE 27 26 1 INTO 5 17 1
REFINE 25 26 1 INTO 5 17 1
DI IVAR 51*200
DI RG 26 26 1 IVAR 41.1296 24.2132 14.2544 8.39161 4.94017 2.90830
1.71213 1.00793 0.59337 0.34932 1.0 0.34932 0.59337
1.00793 1.71213 2.90830 4.94017 8.39161 14.2544
24.2132 41.1296
DI RG 27 26 1 IVAR 40.0 40.0 40.0 40.0 40.0
DI RG 25 26 1 IVAR 40.0 40.0 40.0 40.0 40.0
DJ JVAR 51*150
DJ RG 26 26 1 JVAR 34.9087 18.6605 9.97496 5.33212 2.85029 1.52362
0.81445 0.43536 1.00.43536 0.814451.52362 2.85029
5.33212 9.97496 18.6605 34.9087
DJ RG 27 26 1 JVAR 34.9087 18.6605 9.97496 5.33212 2.85029 1.52362
0.81445 0.43536 1.00.43536 0.814451.52362 2.85029
5.33212 9.97496 18.6605 34.9087
DJ RG 25 26 1 JVAR 34.9087 18.6605 9.97496 5.33212 2.85029 1.52362
0.81445 0.43536 1.00.43536 0.814451.52362 2.85029
5.33212 9.97496 18.6605 34.9087
** Permeability in the fracture is set to Kfeff
PERMI CON 0.10
PERMI RG 26 26 1 JVAR 8*0.10 83.33 8*0.10
PERMI RG 27 26 1 JVAR 8*0.10 83.33 8*0.10
PERMI RG 25 26 1 JVAR 8*0.10 83.33 8*0.10
PERMJ EQUALSI
PERMK EQUALSI
48511.34 0.5 5.5
10000.0
48511.34 0.5 5.5
10000.0
48511.34 0.5 5.5
10000.0
** Set the pseudoized fracture non Darcy flow correction factor (in the fracture
blocks)
NDARCYCOR CON
0
152 Tutorial
User's Guide IMEX
Using the Fracture Clean-up Model

The fracture clean-up model is made up a number of options specifically developed to allow
IMEX to model flow back of fracturing fluids around a newly fractured well.
The option was developed to be used with the polymer option in IMEX as the fracturing fluid
is modeled as a water-polymer mixture (Only the velocity and polymer concentration
dependent water viscosity and fracture width correction features actually require the polymer
model to be used).
The option is made up of the following features.
Features
Velocity and polymer concentration dependent water-polymer mixture viscosity can be used
to model shear thinning/thickening fluids.
Fracture width correction for reservoir condition velocity to be used in the above waterpolymer mixture viscosity calculation. A zero value for the correction eliminates the velocity
dependency in regions away from the fracture. A correction value of 1.0 uses the block
average reservoir condition velocity directly.
Block and direction dependent pressure gradient thresholds to flow in specified regions.
Once a threshold has been exceeded the flow connection will always allow flow regardless of
the pressure gradient. This can be used to approximate yield stress
Non equilibrium initial saturation overrides can be entered in user defined regions (e.g. near a
fracture) in a model which is initialized everywhere else using a gravity equilibrium
initialization technique. This can be used to set up fracture fluid distributions in the fracture
without the need to model fracture fluid injection.
The ability to initiate the pressure gradient threshold option at a restart is added. This allows
the model to simulate fracture fluid injection followed by clean-up (after the restart).
All of the above features can be used with or without the non Darcy flow option and can be
used independently of each other.
Velocity Dependent Polymer Mixture Viscosity (Component Properties Section)
A new option in the *PMIX keyword has been added.

*PMIX *VELTABLE
This signals IMEX to expect velocity dependent polymer-water mixture viscosity data.
IMEX expects to read the keyword VWT followed by an initial velocity value on the same
line. Following this at least two relative polymer concentration - relative water-polymer
mixture viscosity rows are expected to be read in.
After the entire table at the initial reservoir condition velocity value has been entered the next
occurrence of VWT followed by a larger reservoir condition velocity will trigger the reading
of the next relative polymer concentration - relative water-polymer mixture viscosity table (at
least two rows).
If only a single reservoir condition velocity is entered (only one value of VWT), the table will
become velocity independent.
User's Guide IMEX
Tutorial 153
The relative polymer concentration is the actual polymer concentration divided by the
reference polymer concentration *PREFCONC. The relative water-polymer mixture
viscosity is the mixture viscosity divided by pure water viscosity.
An example of the *PMIX *VELTABLE keyword is given below for a shear thinning fluid at
velocities 0.10, 10.0 and 40.0 ft/day. The velocities used for the table lookup are intrinsic
reservoir condition velocities and so are calculated using the average velocity in a gridblock
divided by the block porosity. In addition, the velocities are multiplied by a fracture width
correction term which will be discussed shortly (Velcor = FRWIDTHCOR * Reservoir
Condition Velocity/).
*PMIX *VELTABLE
*VWT 0.10
0.0
1.00
0.2
1.50
0.6
2.00
1.0
5.00
*VWT 10.0
0.0
1.00
0.2
1.20
0.6
1.75
1.0
3.00
*VWT 40.0
0.0
1.00
0.2
1.10
0.6
1.40
1.0
2.00
For velocities below 0.10 ft/day, the table at 0.10 ft/day is used. For velocities above 40.0
ft/day the table at 40.0 ft/day is used.
Fracture Width Correction (Rock-Fluid Data Section)
The model includes a fracture width correction array *FRWIDTHCOR which is used to
enable/disable velocity dependent water-polymer mixture viscosity calculations in portions of
the reservoir and to increase the effective velocity in the fractured region due to numerical
fracture representation.
*FRWIDTHCOR = 0.0 turns off the velocity dependent viscosity calculation for the block it
is defined in. The model will always use the lowest velocity (*VWT) in the *PMIX
*VELTABLE table in this case.
A non zero *FRWIDTHCOR is used to multiply the intrinsic water-polymer mixture velocity
for use in the *PMIX *VELTABLE table. For example if a 0.1 millimeter fracture is
represented using a 0.2 meter gridblock a *FRWIDTHCOR of 2000.0 might be used to
approximate the much higher velocity along the actual fracture.
When *PMIX *VELTABLE is encountered, a two dimensional interpolation is used. Waterpolymer mixture velocity is calculated for each gridblock. The velocity is divided by
gridblock porosity and multiplied by a correction factor (*FRWIDTHCOR), or:
Velcor = FRWIDTHCOR * Reservoir Condition Velocity/
154 Tutorial
User's Guide IMEX
Where Velcor is the corrected velocity used in the VELTABLE table, FRWIDTHCOR is the
fracture width correction term entered on a block by block basis, Velocity is the average
velocity in a block and is the blocks porosity
Block and Direction Dependent Pressure Gradient Thresholds (Rock-Fluid Data
Section)
The fracture clean-up model approximately accounts for yield stress effects by including
pressure gradient thresholds in the three directions. The arrays *PTHRESHI, *PTHRESHJ,
and *PTHRESHK define these pressure gradients between every gridblock and its neighbor.
The array value *PTHRESHI *IJK I, J, K defines the pressure gradient (pressure drop / unit
of length) required to initiate flow between block I, J, K and I+1, J, K.
The array value *PTHRESHJ *IJK I, J, K defines the pressure gradient (pressure drop / unit
of length) required to initiate flow between block I, J, K and I, J+1, K.
The array value *PTHRESHK *IJK I, J, K defines the pressure gradient (pressure drop / unit
of length) required to initiate flow between block I, J, K and I, J, K+1.
The pressure gradient threshold is a one time switch. Once it is exceeded between two blocks
in a specific direction, the interblock connection remains open to flow thereafter.
A zero value of the thresholds indicates the connection is always open to flow.
Regions in and around the fracture may have non zero values of *PTHRESHI, *PTHRESHJ,
and *PTHRESHK, while regions in the reservoir might have zero values for thresholds.
Non Equilibrium Initial Saturation Override (Initial Conditions Section)
In cases where the initial injection of fracturing fluid into the fracture is not modeled, it
would be convenient to be able to initialize the entire reservoir to a gravity-capillary pressure
equilibrium state and just override this state in the region in and around the fracture which
has just had fracturing fluid pumped into it.
This can be done using the keywords *SONEQ and *SWNEQ. These keywords override the
gravity-capillary pressure equilibrium saturations determined when using the *VERTICAL
*BLOCK_CENTER or *VERTICAL *DEPTH_AVE equilibrium options in specified
regions.
*SONEQ and *SWNEQ should only be defined (using the *IJK array reading option) in the
region around the fracture the user wishes to override saturations in. Normally around the
fracture we expect a high saturation of the fracturing fluid, this is modeled as water with a
high polymer concentration. Using high initial *POLYCONC (polymer concentration) and
high *SWNEQ (low *SONEQ) in this region is a convenient way to do this.
Initiating the Pressure Gradient Threshold Option at a Restart
On occasion it may be necessary to also model the fracturing fluid injection period of the
fracture clean-up process. It is possible to accomplish this by restarting from the end of the
injection period (or shutin period) before the flow back period and only introducing pressure
gradient threshold calculations in the restarted run.
The fracture cleanup model can determine if the *PTHRESHI, *PTHRESHJ, *PTHRESHK
keywords were in the data set of the model to be restarted from. If they were not, and the
pressure gradient threshold keywords are encountered in the current data set, then the
User's Guide IMEX
Tutorial 155
pressure gradient threshold option is initiated on the restart. On the restart, the pressure field
read from the restart record is set as the initial pressure field. A pressure gradient that will
initiate flow in the I direction is then defined as:
DP = [(P(I+1,J,K) Prestart(I+1,J,K)) - (P(I,J,K) Prestart(I,J,K))]/[Distance_I]
Distance_I is the distance between block centers in the I direction
Prestart is the initial pressure field read from the restart.
P is the current pressure field
DP is compared with *PTHRESHI, when DP is greater than *PTHRESHI (defined in block
I,J,K) flow is initiated between blocks I,J,K and I+1,J,K.
On restarts before or after the introduction of the pressure gradient threshold keywords, the
model restarts normally.
As the introduction of the pressure gradient threshold keywords on a restart can only be done
once per run, this model cannot be used when wells are fractured at different times.
It is also possible and desirable to change the water-polymer mixture viscosity during the
restart which separates the injection period from the flow back period. In the injection
portion of the run, the fracturing fluid is a very high viscosity fluid. During flow back the
water-polymer mixture fluid has a very much lower viscosity. Similarly, it is also possible to
alter *FRWIDTHCOR on the restart.
When the mixture viscosity is altered on a restart, it is critical that the *PMIX *VELTABLE
table be used both for injection and production periods. The values in the table, the number
of points in each table, and the number of velocity tables may be altered as required, but the
*VELTABLE option must be used.
Output of Pressure Thresholds and Fracture Width Correction (Input/Output Control
Section)
The *RES list of input arrays has been expanded to include FRWIDTHCOR and the three
pressure threshold arrays (PTHRESHI, PTHRESHJ, PTHRESHK).
156 Tutorial
User's Guide IMEX
Using the Oil Wet Option

Notes:
1. The original oil wet option initialization in IMEX is still available with the
BLOCK_CENTER reservoir initialization method. This version of the oil wet
option initialization assumes the reservoir began its existence as an oil wet
reservoir.
The new oil wet option initialization option assumes that the reservoir is initially
water wet and has undergone a wettability alteration to become an oil wet
reservoir.
These assumptions have a significant impact on how fluids are placed during
reservoir initialization. The user should understand completely the difference in
order to obtain the initial conditions which most reflect his reservoir model.
2. Within a PVT region, all rock types should either be oil wet or water wet. If both
oil wet and water wet rock types are found, unphysical results may occur. The
DEPTH_AVE initialization option will warn the user if this occurs and stop the
simulation.
Oil Wet Initialization (BLOCK_CENTER initialization):
The initialization of a BLOCK_CENTER oil wet reservoir assumes that the reservoir is
initially oil wet. We assume that initially we have a 100% oil filled reservoir into which
water migrates. As water is heavier than oil it fills the region of the reservoir below the
present water oil contact.
As the migrating water cannot replace the connate oil:
The water zone below the water oil contact is made up of water and connate oil.
The oil zone does not contain any water at all.
The gas zone is made up of gas and connate oil.
The Pcwo (non-wetting minus wetting pressure) controls the transition zone of oil into the
water zone.
It is possible to adjust the oil in the oil zone to be less than 100% by using the *WOC_SW
keyword. When using water wet initialization this allows the user to adjust the amount of
water in the water zone. When using the oil wet option with BLOCK_CENTER
initialization, this now refers to the oil saturation (wetting phase saturation) in the oil zone.
Hence it is possible to alter the oil phase saturation for each PVT region. Setting a
WOC_SW equal to 0.80 limits the oil saturation in the oil zone to a maximum of 80%.
It is also possible to adjust the amount of oil above the gas oil contact by using the
*GOC_SW keyword. When using the oil wet option with BLOCK_CENTER initialization,
*GOC_SW now controls the wetting phase (oil) saturation in the gas zone. Setting GOC_SW
equal to 0.05 can change the maximum oil saturation in the gas cap to 5%.
This initialization option although entirely consistent does not represent the type of oil wet
reservoirs normally encountered. More commonly oil wet reservoirs initially begin life as
water wet and change wettability at a later time. This type or reservoir should be initialized
using the DEPTH_AVE initialization option.
User's Guide IMEX
Tutorial 157
Oil Wet Initialization (DEPTH_AVE initialization):
This type of initialization (when used with the oil wet option) assumes that the reservoir is
initially water wet and has had its wettability altered (to oil wet). The capillary pressure
between non wetting and wetting phases still govern the transition zones, but the gas zone
contains gas and water, the oil zone contains oil and water (both mobile and immobile) and
the water zone contains water (not water and oil).
Unless specified explicitly, the WOC_PC (the capillary pressure at the water oil contact) is
internally set equal to the maximum Pcwo (see Pcwo definition below), which occurs at the
lowest oil (wetting phase) saturation. Therefore the transition zone occurs within the oil zone
and not below the water oil contact.
If a WOC_PC equal to the minimum Pcwo, which occurs at the largest oil (wetting phase)
saturation, is input in an oil wet model, the WOC would represent the free water level in the
reservoir. The oil water transition zone would occur below the free water level.
A different WOC_PC can be entered for each PVT region.
This type of initialization produces results which are very similar to a water wet reservoir
initialization which at a later date undergoes a wettability change to oil wet, and is then
allowed to re-equilibrate.
Flow in an Oil Wet Model
In an oil wet model (regardless of initialization option), three phase flow is modelled
assuming water is the intermediate phase. Therefore Kro is a function of oil saturation, Krg is
a function of gas saturation, while Krw is a function of both oil and gas saturation (through
Krwo and Krwg).
The normal three phase flow models (e.g. Stone 1, Stone 2, segregated, etc) are used to
determine water relative permeability not oil permeability.
( )
K rg = f S g
K ro = f (S o )
K rw = f S o , S g
When modelling 3 phase flow for example using Stones 2nd model we would obtain Krw
from:
K rwg
K rw = K rwco * rwo + K ro *
+ K rg K ro K rg
K rwco
K rwco
Where Krwco is the water relative permeability at connate oil, Krwo is the water relative
permeability obtained from the water-oil table (So+Sw = 1) and Krwg is the water relative
permeability obtained from the liquid-gas table (Sl+Sg = 1, So = connate oil). This approach
for oil wet rock is a direct extension of Stones model which originally developed for water
wet systems.
Relative Permeability/Capillary Pressure Curves in an Oil Wet Model
When the oil wet option is used the meanings of the columns in the relative permeability
tables are altered. Normally water is the wetting phase and oil is the nonwetting liquid phase.
158 Tutorial
User's Guide IMEX
SWT Table
When the oil wet option is active the column which normally contains water saturation (the
first) should now contain the saturation of the wetting phase (oil). The Krw column (2nd
column) should contain the wetting phase relative permeability (oil relative permeability).
The Krow column (3rd column) contains the nonwetting phase relative permeability (water
relative permeability).
The fourth and fifth columns which normally (for water wet models) contain the Pcow and
Pcowi now contain Pcwo and Pcwoi. These represent the positive capillary pressure between the
non wetting water phase and the wetting oil phase. Since the tables still tabulate non wetting
phase pressure minus wetting phase pressure versus wetting phase saturation (So), the shape
of the Pcwo curves resembles that of the Pcow curves versus water saturation in a water wet
system.
p cwo = f (S o ) = p w p o and p cow = f (S w ) = p o p w

SLT Table
The meanings of the columns of the gas-liquid tables are not altered. However the liquid
saturation in the table is made up of water and connate oil. In the DEPTH_AVE initialization
option the Pc curve entered in the liquid gas table must be the Pcog curve (even though oil is
not the intermediate phase. When using the BLOCK_CENTER initialization option the
capillary pressure curve in the liquid gas table must be the Pcwg curve.
If the capillary pressure received is from a water-gas system, pcgw = pg pw, to accommodate
IMEX table input for DEPTH_AVE initialization, the user needs to convert pcgw to pcgo using
pcgo = pcgw + pcwo.
Endpoint Arrays in an Oil Wet Model
The use of the oil wet option also changes the meaning of the user specified grid block
specific end point arrays. Unless these changes in definition are accounted for, the unscaled
relative permeability curves will be incorrect.
The normal definitions of the affected arrays are.
Swcon
Swcrit
Soirw
Sorw
Sorg
Slcon
User's Guide IMEX
=
=
=
=
=
=
Connate water saturation

Critical water saturation
Irreducible oil saturation
Residual oil saturation
Residual oil saturation (gas liquid table)
Connate liquid saturation which is equal to connate water
(Swcon) plus irreducible oil saturation (Soirg) if the *NOSWC option
is not used, and equal to irreducible oil saturation (Soirg) alone, if
the *NOSWC option is used
Tutorial 159
When using the oil wet option they are modified to be.
Swcon
Swcrit
Soirw
Sorw
Sorg
Slcon
=
=
=
=
=
=
Connate wetting phase (oil) saturation

Critical wetting phase (oil) saturation
Irreducible nonwetting phase (water) saturation
Residual nonwetting phase (water) saturation
Residual nonwetting phase (water) saturation (gas liquid table)
Connate liquid saturation which is equal to connate oil
(Swcon) plus irreducible water saturation (Soirg) if the *NOSWC
option is not used, and equal to irreducible water saturation (Soirg)
alone, if the *NOSWC option is used
The scaling ARRAYS *KRWIRO, *KROCW, *KROGCG, *PCWMAX, *PCGMAX,

*JFWMAX and *JFGMAX must also be used with altered definitions if the oil wet option are
used.
The normal definitions of these arrays are:
Krwiro
Krocw
Krogcg
Pcwmax
Pcgmax
Jfwmax
Jfgmax
=
=
=
=
=
=
=
Water rel. perm. at irreducible oil (oil-water table)

Oil rel. perm at connate water (oil-water table)
Oil rel. perm at connate gas (liq.-gas table)
Oil water capillary pressure at connate water
Gas oil capillary pressure at connate liquid
Oil water J function at connate water
Gas oil J function at connate liquid
When using the oil wet option they are modified to mean.
Krwiro
Krocw
Krogcg
Pcwmax
Pcgmax
=
=
=
=
=
Jfwmax
Jfgmax
=
=
160 Tutorial
Wetting Phase (Oil) rel. perm. at irreducible water (oil-water table)

Non Wetting Phase (water) rel. perm at connate oil (oil-water table)
Non wetting phase (water) rel. perm at connate gas (liq.-gas table)
Water-oil capillary pressure at connate oil
Gas oil capillary pressure at connate liquid (DEPTH_AVE)
Gas water capillary pressure at connate liquid (BLOCK_CENTER)
Water oil J function at connate oil
Gas oil J function at connate liquid (DEPTH_AVE)
Gas water J function at connate liquid (BLOCK_CENTER)
User's Guide IMEX
Example Oil Wet Rock Type Curves (Relative Permeability)

Figure 1
Water and Oil Wet Curves vs Water Saturation
1
0.9
0.8
0.7
Krw (WW)
Kr
0.6
Krw (OW)
0.5
Kro (WW)
0.4
Kro (OW)
0.3
0.2
0.1
0
0
0.2
0.4
0.6
0.8
Sw
Figure 2
Oil Wet Curves vs. Oil Saturation
1
0.9
0.8
0.7
Kr
0.6
Kro (OW)
0.5
Krw (OW)
0.4
0.3
0.2
0.1
0
0
0.2
0.4
0.6
0.8
So
Oil wet curves are extracted from figure 1 and plotted against Oil saturation which for an oil
wet case is the wetting saturation (figure 2). It is the curves in figure 2 which must be entered
as relative permeabilities in an oil wet rock type. In column 1, oil (wetting phase) saturation
is entered instead of water saturation. In column 2 relative permeability to oil (wetting phase)
is entered instead of relative permeability to water and in column 3, relative permeability to
water (non wetting phase) is entered instead of relative permeability to oil.
User's Guide IMEX
Tutorial 161
It is interesting to note that the relative permeability curves for the water wet and oil wet
curves in Figure 1 are in good agreement if they are plotted vs. wetting phase (water for water
wet and oil for oil wet). The figure below presents this.
Figure 3
Water Wet Curves vs. Water Saturation
1
0.9
0.8
0.7
Kr
0.6
Krw (WW)
0.5
Kro (WW)
0.4
0.3
0.2
0.1
0
0
0.2
0.4
0.6
0.8
Sw
Oil Wet Curves vs. Oil Saturation

1
0.9
0.8
0.7
Kr
0.6
Kro (OW)
0.5
Krw (OW)
0.4
0.3
0.2
0.1
0
0
0.2
0.4
0.6
0.8
So
Example Oil Wet Rock Type Curves (Capillary Pressure)
If the capillary pressure received are in terms of Pcow (water phase pressure minus oil phase
pressure) in an oil wet system versus water saturation (the non wetting phase) the expected
capillary pressure curve would resemble Figure 4 below.
162 Tutorial
User's Guide IMEX
Figure 4
Pcow (OW) vs. Water Saturation
0
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
-10
-20
Pcow (psi)
-30
-40
Pcow (OW)
-50
-60
-70
-80
-90
Sw
To use this data in IMEX, it is necessary to translate the Pcow capillary pressure into the
Pcwo curve (non wetting phase pressure minus wetting phase pressure) that the oil wet model
requires as a function of wetting (oil) phase saturation.
For example p cwo = p cow and S o = 1 S w
The resulting curve is shown below. This curve is similar in shape to pcow versus water
saturation in a water wet system.
Pcwo (OW) vs. Oil Saturation
90
80
70
Pcwo (psi)
60
50
Pcwo (OW)
40
30
20
10
0
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
Oil Saturation
User's Guide IMEX
Tutorial 163
The relative permeability and capillary pressure examples are from Anderson, W.G. (JPT
Oct. and Nov. 1987).
Oil Wet Option and Hysteresis
Due to the fact that the oil wet option rock type tables are tabulated against wetting phase
saturation, all of the hysteresis options in IMEX function. Pcwo hysteresis and nonwetting
phase relative permeability hysteresis are modelled. It must be remembered that in the oilwet
model the nonwetting liquid saturation referred to is water so HYSKRO somax refers to a
water residual saturation and not an oil residual saturation.
164 Tutorial
User's Guide IMEX
Special Keywords for Advanced Users (SHIFT)

Note:
1. These keywords greatly affect the performance of the simulation model and should
only be used when the specific issues raised in the section below are seen. These
keywords are not listed in the normal keyword section of the manual and in
BUILDER as this would make their accidental modification possible.
The Shift Keywords:
In the numerical section the following keywords may be entered:

SHIFT PRESS p_shift
SHIFT SATUR s_shift
SHIFT WHP whp_shift
Normally the values of s_shift, p_shift, and whp_shift are defaulted by the simulator and do
not cause problems. The shifts are used in the determination of numerical derivatives.
However under the following conditions it is possible to alter them and obtain far better
numerical performance.
1. Reservoir block volumes are relatively small, but at least one flow area is very
large. This might happen then an areally large block has a very small block
thickness. For example a 500 x 500 x 0.01 ft block. In this example the block
gross volume is only 2500.00 ft3, but the term which is used in the geometric
portion of the vertical transmissibility (DX * DY)/DZ is 500x500/0.01 or 2.5E6.
The normal pressure and saturation shifts in the simulator can cause a very large
change in flow and hence very slow convergence. Use of smaller pressure and
saturation shifts overcome this. In this case p_shift of 5.0e-6 psi and s_shift of
1.0e-6 can be tried.
2. The well head pressure being calculated is close to an atmosphere or the results are
very sensitive to the well head pressure. The default whp_shift can be reduced
from its default value of 0.0145 psi to a value on the order of 1.0e-5 psi.
User's Guide IMEX
Tutorial 165
Modelling Non Darcy Flow in Fractured Shale Gas Reservoirs

This work extends the work done with IMEX to model non Darcy flow in hydraulically
fractured reservoirs using LGR grids. In that work, it was found that the fracture could be
represented by a one foot thick pseudo fracture if Darcy flow and the Forchheimer number
were pseudoized to predict reasonable pressure drops in the fracture. In addition, the region
around the fracture needed to be modelled using a finely gridded LGR in order to reproduce
the pressure drop around the fracture accurately.
In the modelling of non Darcy flow in fractured shale gas reservoirs we extend the previous
work by using LGR grids to model flow into networks of fractures connected to either
vertical or horizontal wells. We investigate the effect that the low matrix permeability (100
nano-Darcies) has on the models and recommend the use of a simpler Dual Permeability
LGR grid. We also investigate simple approaches to add shale-gas desorption to IMEX and
compare results with and without desorption affects.
LGR modelling of Fractured Shale Networks
Fracture Width Pseudoization
The basic LGR grid used to model vertical hydraulic fractures was slightly modified to create
a template which could be used for a rectangular network of connected fractures in a very low
permeability reservoir. In this example we assume that the fracture length is 200 feet (1/2
length = 100 feet), the fracture permeability is 4000 mD, the reservoir permeability is 0.0001
mD, and the fracture thickness is 0.001 feet.
We first refine the rectangular region of the reservoir we believe contains a rectangular
network of fractures as shown below.
REFINE 47:55 47:55 1 INTO 21 21 1
We then define a fine scale LGR grid within the blocks to be refined. The fine grid is
normally required to obtain accurate pressure drop modelling in the regions adjacent to the
fractures. We will see that for low permeability reservoirs this is not the case.
DI IVAR
101*200.0
DI RG 47:55 47:55 1 IVAR
33.0 25.0 16.0 8.0 6.0 4.0 3.0 2.0 1.5 1.0 1.0 1.0 1.5 2.0 3.0 4.0
6.0 8.0 16.0 25.0 33.0
DJ JVAR
101*200
DJ RG 47:55 47:55 1 JVAR
33.0 25.0 16.0 8.0 6.0 4.0 3.0 2.0 1.5 1.0 1.0 1.0 1.5 2.0 3.0 4.0
6.0 8.0 16.0 25.0 33.0
We then define the matrix permeability in the shale and the fracture permeability for fractures
which extend in both areal directions. As with the simpler single plane hydraulic fracture
models, the fracture permeability and fracture Forchheimer number are pseudoized in order to
be able to use a computationally efficient 1 foot fracture width.
166 Tutorial
User's Guide IMEX
PERMI CON
1e-4
PERMI RG 47:55 47:55 1 ALL
10*1e-4 4.000 10*1e-4
10*1e-4
10*1e-4 4.000 10*1e-4
10*1e-4
10*1e-4 4.000 10*1e-4
10*1e-4
10*1e-4 4.000 10*1e-4
21*4.000
10*1e-4 4.000 10*1e-4
10*1e-4
10*1e-4 4.000 10*1e-4
10*1e-4
10*1e-4 4.000 10*1e-4
10*1e-4
10*1e-4 4.000 10*1e-4
4.000 10*1e-4
4.000 10*1e-4
4.000 10*1e-4
10*1e-4 4.000 10*1e-4

10*1e-4 4.000 10*1e-4
10*1e-4 4.000 10*1e-4
4.000 10*1e-4
4.000 10*1e-4
4.000 10*1e-4
10*1e-4 4.000 10*1e-4

10*1e-4 4.000 10*1e-4
10*1e-4 4.000 10*1e-4
The fracture set up in each of these LGRs is pictured below
The fracture is indicated by the red colored region. The remainder of the LGR, the blue
region is matrix.
When a rectangular region of these grids is defined as in a fractured shale model the
connected LGR regions take on the appearance shown below. At the resolution presented
below it is impossible to see the red fracture blocks as can be seen in the figure above.
However, the clustering of grid blocks near the fracture makes the fracture locations obvious.
User's Guide IMEX
Tutorial 167
The first effort in this work was determining if the 1 foot wide fractures could be used to
adequately model the flow in the extremely narrow fractures. In this model the fracture
conductivity and fracture widths were assumed to be quite small. The fracture with was set at
0.001 foot (0.012 inch) and the fracture permeability was varied from 4 mD-ft to 12 mD-ft.
Fracture conductivity as low as 0.5 mD-ft were used as the permeability of fractured shale
and width of fracture are relatively small due to the small amounts of fracturing media used.
Models of the type above were set up to model Darcy and non Darcy flow from LGR grids
using 0.001 ft. fracture thicknesses and 1 ft. thicknesses. In this pseudoization test model, we
only used a vertical well configuration. An extra 8 grid blocks were used in the fine grid LGR
regions so better resolution could be used capturing the flow near the 0.001 ft. wide fractures.
So in the 0.001 ft. wide fracture models, the LGR consisted of a 29 x 29 block grid. In the 1
ft. wide fracture model the LGR used was the 21 x 21 block grid defined in the data above.
In the 0.001 ft. wide fracture model, neither the fracture permeability nor the Forchheimer
number was pseudoized. There was no need as the actual fracture dimensions were used.
The results of the first 300 days of production were compared. The running time of the 0.001
ft. wide fracture model made it impractical to produce for the full 10 years of the simulation;
however the first 300 days of production are enough to determine the validity of the
pseudoization technique.
168 Tutorial
User's Guide IMEX
The important finding in this figure is that the non Darcy correction could be used to model a
pseudo fracture 1000 times wider than the actual fracture size.
This simulation was used as validation of the 1 ft. wide fracture width pseudoization results.
In all other simulations models, one foot fracture width pseudoizations will be the reference
solutions when comparing simpler model fracture network models
Low Permeability Shale and Dual Permeability LGR Fracture Network Models
When investigating possible simplifications to the LGR grids used in these models it became
evident that the very low permeabilities in the matrix (100 nano-Darcies) allowed the use of
far simpler grids. The most interesting of the grids used is a Dual Permeability representation
within an LGR grid (DK-LGR). Instead of using a 1 foot wide grid cell to model a fracture, a
dual permeability fracture fairway was used. This grid modelled the fracture flow path to
the well as a relatively low permeability 40 ft. wide region (fairway). The non Darcy flow
Forchheimer correction factor was used in this model to correct a 0.001 ft wide fracture to a
represent the non Darcy flow in a 40 ft wide fairway. The Forchheimer correction value was
over 13,000.0 to achieve this. An example of the grid is shown below.
User's Guide IMEX
Tutorial 169
Only the fractured part of the dual permeability grid is shown for clarity. A dual permeability
model was used with equally sized 5 x 5 LGR refinements. The fracture permeability and
porosity were chosen to represent the effective permeability and porosity of a 0.001 ft wide
fracture in a 200 foot block. Note that in this grid the fracture porosity is zero in the
unfractured portion of the grid and in the region not designated as a fracture fairway. In
theses regions only matrix porosity is defined. Additionally in the fracture fairway, no matrix
porosity is defined. The matrix fracture transfer terms were calculated assuming a horizontal
fracture spacing of 200 ft and assuming no horizontal fractures. However since there are no
co-located fracture and matrix blocks, the fracture spacing (and matrix-fracture transfer term)
is not used. The CMG grid module makes the appropriate connections between matrix-only
and fracture-only regions.
For the same 4 mD-ft fracture network as was previously used, we compare the DK-LGR
model with the standard LGR model for both Darcy and non Darcy flow. The results are
presented below.
170 Tutorial
User's Guide IMEX
An excellent result is observed using dual permeability 5 x 5 LGR grids which produces
similar results to far more refined 21 x 21 single porosity LGR grids.
We will show later in this work that these excellent results cannot be duplicated with more
permeable matrices, when the matrices are more permeable good pressure drop resolution
becomes critical, but we have been able to produce good matches with matrix permeabilities
as high as 0.01 mD.
We have run the DK-LGR model on several variants of the basic fractured shale reservoir.
The fracture conductivity is increased to 12 mD-ft
The fracture conductivity is decreased to 0.5 mD-ft
The fracture spacing is decreased from 200 ft to 25 ft
Horizontal well in the 4 mD-ft fracture conductivity model
Horizontal well in the 12 mD-ft fracture conductivity model
Horizontal well in the 25 ft fracture spacing model
Three dimensional 4 mD-ft model (10 layers only perfed in layer 2)
12 mD-ft model in a 0.01 mD matrix permeability shale
In all of these models the DK LGR model produces results virtually identical to the standard
LGR based models for both Darcy and non Darcy flow.
User's Guide IMEX
Tutorial 171
Dual Permeability LGR GRID Permeability and Porosity Pseudoization
The distance between vertical fractures in both the X and Y directions is 200 ft. The 200 ft x
200 ft blocks are broken in to 25, 40 ft x 40 ft blocks in an LGR. The LGR blocks in the row
j=3 and column i=3 are set to have fracture porosity and so represent the fracture fairway.
The LGR blocks in the row j=3 and column i=3 also have zero matrix porosity. In each of
these cells, a single 0.001 ft wide fracture is assumed to exist which runs for the entire length
of the 40 ft long LGR block, hence the average fracture porosity (assuming the porosity in the
0.001 ft wide fracture is 100%) is (0.001 x 40 / (40 x 40).
This is 0.000025.
The permeability of the fracture is adjusted so that Kfrac x Areafrac = Kfraceff x Arealgrblock.
In the case with a 4 mD-ft fracture this gives us an effective fracture permeability of
(0.001/40) x 4000 mD which equals 0.10 mD
Non Darcy Correction Factor Setup (Single Porosity LGR Grid)
For the single porosity reference grids (21 x 21 block LGR). The non Darcy correction setup is
virtually identical to the setup for a hydraulically fractured well. The only difference is the two
dimensional nature of the fracture. This is reflected in how the NDARCYCOR parameter is
defined.
The value used in the NDARCYCOR factor is (Kfrac/Kfraceff) (2-1.1045) or 485.847 where kfrac
equals 4000 mD and Kfraceff equals 4 mD. Kfraceff is 4 mD due to the use of a 1 foot wide
fracture to model a 0.001 ft wide fracture.
The value 1.1045 in the NDARCYCOR equation is the exponent which multiples
permeability in the Beta factor term of the Forchheimer number. See the tutorial section
concerning non Darcy flow in fractures for a discussion of the non Darcy correction factor.
In this example, the Beta A-term = 27.3e9 and the B-term = 1.1045 were taken from an
example in the textbook by Dake (Fundamentals of Reservoir Engineering).
**Beta A Beta B
27.3e9
1.1045 0
27.3e9
1.1045 0
27.3e9
1.1045 0
NDARCYCOR CON
0
NDARCYCOR RG 47:55 47:55 1 ALL
10*0 485.847 10*0
10*0 485.847 10*0
485.847 10*0
10*0 485.847 10*0
10*0 485.847 10*0
485.847 10*0
10*0 485.847 10*0
10*0 485.847 10*0
21*485.847
10*0 485.847 10*0
10*0 485.847 10*0
485.847 10*0
10*0 485.847 10*0
10*0 485.847 10*0
485.847 10*0
10*0 485.847 10*0
10*0 485.847 10*0
172 Tutorial
10*0 485.847 10*0
10*0
10*0 485.847 10*0
10*0
10*0 485.847 10*0
10*0
10*0 485.847 10*0
10*0
User's Guide IMEX
Fracture Porosity and Non Darcy Correction Factor Setup (Dual Perm. LGR Grid)
In the Dual Permeability grid, we set up the non Darcy correction factor in all fracture blocks
which are defined by the keyword POR FRACTURE.
How the values of fracture permeability and porosity were determined were discussed in the
previous section.
Initially define matrix (shale) porosity everywhere at 3%
POR MATRIX CON 0.03
Initially define no fractures in the Dual Permeability model

POR FRACTURE CON 0.0
Define fractures in every block in the hydraulically fractured region

MOD 47:55 47:55 1 =
0.000025
The fractures in the LGR region are limited to fracture fairways

POR FRACTURE RG 47:55 47:55 1 ALL
0
0
0.000025
0
0
0
0.000025
0
0.000025 0.000025
0.000025 0.000025 0.000025
0
0
0.000025
0
0
0
0.000025
0
0
0
0
0
The matrix porosity is limited to the non fractured region in the refined area.
POR MATRIX
0.03
0.03
0.00
0.03
0.03
RG 47:55 47:55 ALL

0.03
0.0000
0.03
0.03
0.0000
0.03
0.00
0.0000
0.00
0.03
0.0000
0.03
0.03
0.0000
0.03
0.03
0.03
0.00
0.03
0.03
The volume lost by nulling out the matrix which is co-located with the fracture fairway is
made up by increasing the volume of the remaining matrix blocks in the refined region:
VOLMOD MATRIX RG 47:55 47:55 1 CON 1.5626 ** = 1/(4/5x4/5)
In the statement above, the 16 remaining matrix blocks are given the volume of the original 25.
The non Darcy correction factor is set to zero in the matrix and to a non-zero value in the
fracture as shown below.
**Beta A Beta B
27.3e9
1.1045 0
27.3e9
1.1045 0
27.3e9
1.1045 0
NDARCYCOR MATRIX
CON 0
NDARCYCOR FRACTURE CON 13217.36
The value of the non Darcy correction is based on the ratio of the actual fracture permeability
(4000 mD) to the effective Dual Permeability fracture permeability (0.10 mD) and equals
(4000/0.10) (2-1.1045) or 13,217.36 How the Dual Permeability model fracture permeability of
0.1 mD was arrived at has been previously discussed.
The value 1.1045 in the NDARCYCOR equation is the exponent which multiples
permeability in the Beta factor term of the Forchheimer number. See the tutorial section
concerning non Darcy flow in fractures for a discussion of the non Darcy correction factor.
User's Guide IMEX
Tutorial 173
Use of the Dual Permeability Grids in models with Higher Shale Permeabilities
The DK-LGR model is not significantly different from any other coarsely defined grid we
could have used to model this process. What makes the use of coarse grids possible in these
shale models is the very low permeability and hence low pressure drop in the matrix. In
order to determine the usability of this option we investigated how the match changed when
the matrix permeability was increased from 0.0001 mD to 0.01 mD.
For this study we investigated both the 4 mD-ft and 12 mD-ft conductivity fracture cases.
Note that the 4 mD-ft base case models were not run to completion but were run far enough
to indicate how well the DK-LGR models match the more finely gridded single porosity LGR
models
Vertical Well, 4 mD-ft Fracture Conductivity, 200 foot fracture spacing, Kshale=0.01 mD
21x21 LGR 1 foot wide fracture pseudoization compared to dual perm 5x5 LGR
Vertical Well, 12 mD-ft Fracture Conductivity, 200 foot fracture spacing. Kshale=0.01 mD
21x21 LGR 1 foot wide fracture pseudoization compared to dual perm 5x5 LGR
174 Tutorial
User's Guide IMEX
The above figures indicate that this simple Dual Permeability representation is valid for shale
permeabilities as high as 0.01 mD.
When the 4 mD-ft model was run again with a shale permeability of 0.10 mD, the results of
the simplified DK-LGR model and the detailed LGR model differed significantly.
User's Guide IMEX
Tutorial 175
The figure above models a vertical well in 0.10 mD shale with a 4 mD-ft conductivity
fracture network. The 5x5 block LGR resolution does not properly model gas flow in the
matrix.
Both the Darcy flow and non Darcy flow models exhibit the same behavior which is due to
the poor representation of the pressure distribution in the more permeable matrix.
This result just emphasizes the importance of proper discretization around the fractures to be
able to properly model the rapidly changing pressure in and near the fractures.
From these results we conclude that a simplified 5x5 grid is usable for shale permeabilities of
0.01 mD and less which should be sufficiently high for most shale gas reservoirs.
Gas Adsorption Effects
All of the above studies assume that the primary source of gas for a fractured shale reservoir
is from evolved gas flowing in the matrix. They do not incorporate desorbed gas from
organics in the shale to the overall production.
However, it is relatively easy to model gas desorption from a shale gas reservoir in a black oil
model using a technique developed by Seidle and Arri (Pet. Soc. CIM/SPE 90-118). This
procedure uses a black oil models solution gas ratio to replicate a Langmuir Isotherm and
was used in IMEX to include gas desorption effects in the base model.
The 4 mD-ft vertical well base model was rerun using the single porosity LGR grid with
desorption included for the Darcy flow case and compared to the case without desorption.
The results are presented below.
176 Tutorial
User's Guide IMEX
The Isotherm parameters used for the simulation above are derived for 10% total organic
Carbon New Albany Shale (from Zuber et al, SPE 77469):
Langmuir Volume constant = 104.2 SCF/ton
Langmuir Pressure = 142.5 psia
Shale Density = 2.4 gm/cm3
Using the Langmuir parameters listed above the run with desorption produces about 10%
more gas over the 10 years simulated.
Setting up the Black Oil Model to Simulate Desorption
The use of a black oil model to simulate gas desorption assumes diffusion is fast compared to
desorption and so does not limit desorption.
We need to use a black oil model rather than a gas water model to simulate this process. As it
is necessary to define a small amount of oil in the model (Som); this immobile pseudo oil
represents the organic material which is the source of desorbed gas.
We recommend Som = 0.0001, a value small enough so that it is unnecessary to adjust initial
saturation or porosity conditions.
In each PVT region, we must assume constant matrix porosity as we need to calculate a
Langmuir equation which uses pseudo oil properties and constant matrix porosity.
In the PVT table, pseudo oil properties need to be set up to model an immobile oil phase
which produces gas in response to pressure changes.
Oil Viscosity
=
1.0E+12 cP (makes oil immobile)
Oil FVF
=
1.0E-03 Bbl/STB
Any Reasonable Oil Density can be used (say 40 lb/ft3)
Co
=
Cvo = 0.0
Water and gas properties can be entered normally
The high oil viscosity ensures that the oil does not flow.
The oil FVF of 1.0E-03 need not be realistic; the use of a small value allows Som to be made
small enough to have no affect on initial conditions.
The Langmuir isotherm is represented by a solution gas ratio (Rs) using an equation of the
form
Rs= 5.615 * Lvol * Bo / (Som * Porm) * P/Plang/ (1+P/Plang)
Rs in Field units (SCF/Bbl)
Plang
Lvol
=
=
Langmuir Pressure (psi)

Langmuir Volume (SCF gas/ CF Coal)
Som
Porm
=
=
1e-4
Porosity of matrix
It is necessary to ensure the PVT table has enough resolution to capture the curvature of the
Langmuir Equation.
Use DEPTH_AVE initialization and note that Bubble Point pressure is the desorption
pressure.
User's Guide IMEX
Tutorial 177
The Gas Liquid relative permeability table endpoint (Slcon) needs to be increased to Swc
+0.0001 to account for the pseudo oil. Pcog = 0.0.
The oil relative permeability in the black oil tables should always remain small. A straight
line curve varying from 0.0 to 1.0E-04 is sufficient
Assuming Porm
Lvol
=
=
0.03;
104.2 SCF/ton;
Density of shale
matrix
Langmuir Pressure
C
2.4 gm/cm3;
=
=
142.5 psia;
0.031185 converts the units of Lvol * Shale Density
((SCF/ton)*(gm/cm3)) to SCF/CF.
The pseudo Rs (SCF/Bbl)

= [5.615 * C * 104.2 * 2.4 * 1.00E-03 / (1.00E-04*0.03)] * (P/Plang) / (1+P/Plang)
or
Rs = 14596.6 * (P/Plang) / (1+P/Plang)
The resulting Rs vs. Pressure table is presented below.
P
Rs
14.7
1364.94
200
8523.56
400
10762.47
800
12389.68
1200
13047.24
2000
13625.76
4000
14094.48
178 Tutorial
User's Guide IMEX
The WATER_FIX Keyword

In the numerical section the following keyword may be entered:
WATER_FIX 2
Normally the IMEX formulation and variable switching algorithms are robust enough to
handle even the most difficult problems.
However under certain circumstances, when oil attempts to enter a block which is completely
empty of any hydrocarbon, numerical issues may occur. These can be identified as periods of
a simulation where timestep sizes fails to build up significantly and where timestep repeats
are common.
Using *WPRN *ITER *MATRIX to turn on more timestep convergence information, these
periods are fairly obvious as times when oil and gas residuals fail to reduce in blocks at the
border/interface of hydrocarbon and completely water filled blocks. The maximum residuals
of oil and gas fail to converge at these border blocks and appear to be stuck at values above
the convergence tolerances. Occasionally the message Newton Method is Stuck will
appear.
We have developed an optional algorithm to improve our formulation under these
circumstances. The algorithm may be started by specifying *WATER_FIX 2 in the
numerical data section or by adding the water_fix 2 command line option.
User's Guide IMEX
Tutorial 179
Introducing a Hydraulic Fracture in Recurrent Data

It is possible to model the fracturing of a well at a time other than the simulations initial start
time by introducing an LGR refinement in recurrent data, followed by the redefinition of the
permeability arrays in the refinement (*PERMI, *PERMJ ,and *PERMK) and if required
redefinition of the non-Darcy flow correction term *NDARCYCOR in the refinement.
It is not necessary to define the LGR refinement in recurrent data. The refinement may be
defined at the start of the run and allowed to inherit the parent block properties. The
definition of fracture permeability and non-Darcy correction factors in the LGR representing
the fracture at a later time is sufficient.
If the LGR keywords are introduced at a later time, followed by permeability arrays which
redefine the properties of the newly defined LGR, it is necessary to separate the *REFINE
keywords from the *PERMI, *PERMJ, *PERMK keywords by a keyword from a section
other than from the Reservoir Description Section. Any recurrent section keyword will
suffice.
This is required since the LGR grid is not completely defined until the grid module is exited
from (which is triggered by reading a non grid module keyword). Upon reentry to the grid
module to read the permeability arrays, the LGR grids will be completely set up.
The user can separate the *REFINE keywords from the *PERM keywords by including the
*NDARCYCOR keyword between the two groups.
The user may wish to separate the definition of the LGR and the redefinition of permeability
and non-Darcy correction factor by a *DATE keyword and essentially refine the grid a short
period of time (say one day) before the *PERMI, *PERMJ, *PERMK and *NDARCYCOR
arrays are redefined.
The redefinition of permeability in recurrent data will operate properly with compaction
tables. However the compaction table itself cannot be read in or redefined in recurrent data.
The permeability multipliers in the compaction tables will operate on whatever permeability
is being used at the time the modified interblock transmissibility is being calculated.
When permeability is read in recurrent data and the SR2 option to output permeability is
active (*OUTSRF *GRID *PERM), permeability will be sent to the simulator results file.
When non-Darcy correction factor is read in recurrent data and the non Darcy option is
active, the new non-Darcy correction factors are written to the simulation results file at the
time they were defined (assuming *OUTSRF *RES *ALL is used). This allows the display
of the currently active correction factors at the proper dates.
180 Tutorial
User's Guide IMEX
The Trapped Oil Hysteresis Option for Pcow and Krow (Oil-Water
System)
An optional oil-water hysteresis approach has been implemented in IMEX based on the
consideration of trapped oil in rock. In this approach, the imbibition capillary pressure curve
is given with a new end point, Sormax (as shown in Fig-1a). Sormax is the residual oil saturation
defined by the boundary imbibition process of capillary pressure and oil relative permeability.
It is the maximum residual oil that can be obtained (which is due to trapped oil). The
boundary imbibition process starts from connate water (Swcon), and stops at Sormax. Because of
the trapped oil, the Sormax is greater than Sorw (Fig-1b).
In the *SWT table, if the imbibition Pcow is given like that in curve (b) of Fig-1a where the
values of imbibition Pcow at saturations equal to or greater than the 1-Sormax saturation are
constant, the trapped oil hysteresis option is activated for all grid blocks using that table.
Tables not using the Trapped Oil option have their bounding imbibition Pcow curve spanning
Swcon to 1-Soirw, the same range as the primary drainage Pcow curve
If activated, both Pcow and Krow will be treated with trapped oil hysteresis using the same
Sormax value. Keyword *HYSKRO is no longer necessary and is ignored. Krw is not included
in the hysteresis because of its smaller dependency on trapped oil.
With trapped oil hysteresis, the input drainage Pcow is always used as the reservoir
initialization capillary pressure, the secondary drainage curve and further scanning curves are
formed by Killoughs method (1976) internally using the given primary drainage and
boundary/primary imbibition curves.
User's Guide IMEX
Tutorial 181
Pc
Fig1aHysteresis Pcow
BasicCurves
60
InputCurve
20
(a)PrimaryDrainage
Sw
0
Swcon
(c)2ndDrainage
(b)BoundaryImbibition
20
1Sormax
60
Krow
1.0
Fig1bHysteresis Krow
BasicCurves
InputCurve
(a)PrimaryDrainage
Sw
0.0
0
Swcon
1Sormax
1Sorw
Figure 1: Basic Curves
Pcow Hysteresis
The hysteresis processes are illustrated by Fig-2a. Initially, all reservoir blocks are on the
drainage curve. For any reservoir block, if water saturation begins to increase, the imbibition
process starts and its oil saturation Soh = 1-Sw at that moment is kept as the historically
attained maximum oil saturation.
182 Tutorial
User's Guide IMEX
P
80 c
Fig2aHysteresisPcow
ArbitraryCycle
60
(b)Boundary
Imbibition
40
1Soh
(a)PrimaryDrainage
20
(f)
Sw
(e)Im>DrScanning
(d)Dr>Im
Scanning
20
Sw,rev
40
1Sormax
1Sorh
Fig2bHysteresis Krow
ArbitraryCycle
Krow
1.0
1Soh
(b)Boundary
Imbibition
(a)PrimaryDrainage
(d)Dr>ImScanning
Sw
0.0
0
Swcon
1Sormax
1Sorh
1Sorw
Figure 2: Arbitrary Cycle
The imbibition process scans the curve (d) towards its end point 1-Sorh. The saturation Sorh is
the residual oil saturation of this imbibition scanning curve and is calculated as a function of
the Soh by Lands correlation. The difference, Sorh Sorw is the trapped oil saturation.
The imbibition process moves along the curve (d) as long as water saturation increases but
may reverse itself at any saturation (i.e. if water saturation decreases) and start the drainage
(imbibition to drainage) process (for example curve (e) or curve (f) in Fig-2a). Both curves
(e) and (f) will drain back to the water saturation of 1-Soh, where the original imbibition
process started.
User's Guide IMEX
Tutorial 183
The drainage process on curve (e) or (f) may continue and pass through the point 1-Soh to
lower water saturation. In that case, the maximum historically attained oil saturation, Soh, of
the block increases. On the other hand, if the process on (e) or (f) reverses before the 1-Soh
point is reached (i.e. imbibition begins before reaching 1-Soh), the same curve (e) or (f) is
scanned back as shown by arrows in Fig-2a.
In above descriptions, the imbibition process of curve (d) only starts when water saturation,
Sw, is between Swcon and 1-Sorw. Blocks with Sw greater then 1-Sorw cannot start imbibition
from first drainage simply because the existing oil is immobile.
Following are the descriptions of how Killoughs method is used to calculate scanning curves
(d), (e) and (f) using the input primary drainage Pcow curve (a) and the input
boundary/primary imbibition Pcow curve (b), as shown in Fig-2a, to an arbitrary historically
attained oil saturation, Soh.
Drainage to imbibition scanning curve (d), and imbibition to drainage scanning curve (e), is
calculated by
Pcow(d) = fi Pcow(b) + (1 fi) Pcow(a) ,

Pcow(e) = fd Pcow(a) + (1 fd) Pcow(b) ,
where fi and fd are given by
Sorh Soh + epspc

fi =
Sorh Soh
Sw (1 Sorh)
,
Sw (1 Sorh) + epspc
(1 Sorh) Sw
Soh Sorh + epspc
fd =

.
Soh Sorh
(1 Sorh) Sw + epspc
In the above equations, Sw is the block water saturation. The curvature parameter epspc is
defined via keyword *EPSPC or is set by default with a value of 0.1. Sorh is the residual oil
saturation of the imbibition process scanning curve, which is associated with Soh by Lands
correlation (below)
Sorh = Sorw +
1
1/(Sormax Sorw) 1/(Somax Sorw) + 1/(Soh Sorw)
Where Somax = 1 Swcon. Please note here the whole range of Soh varies from 1-Swcon to Sorw,
therefore the corresponding Sorh ranges from Sormax to Sorw.
If the imbibition process, on curve (d), is reversed at Sw,rev, another imbibition to drainage
curve (f) will be scanned towards 1-Soh. Pcow along scanning curve (f) is calculated using
curve (e) and (d)
Pcow(f) = fd Pcow(e) + (1 fd) Pcow(d) ,

with
Sw, rev (1 Soh) + epspc

fd =
Sw, rev (1 Soh)
184 Tutorial
Sw, rev Sw
.
(Sw, rev Sw + epspc
User's Guide IMEX
Krow Hysteresis
Using the same end points, Krow hysteresis in this approach is consistently coupled with Pcow
hysteresis. Sormax for Krow hysteresis is read from the Pcow boundary/primary imbibition curve.
The input Krow is taken as the primary drainage curve and is used to calculate the drainage to
imbibition scanning curves by Carlsons method (1968) as shown in Fig-2b
Krow(d, Sw) = Krow(a, Sw(shifted)) ,

where
Sw(shifted) = 1 Sorw (1 - Sw Sorh)
(Soh Sorw)
.
(Soh Sorh)
If the imbibition continues, Krow (d) is zero at water saturation = 1-Sorh and higher. If the
imbibition process is reversed, the curve (d) is scanned back to water saturation of 1-Soh.
Damped Pcow Scanning Curves

By default, the Pcow value at 1-Sorh of the scanning curves is treated as the constant which is
the Pcow value at 1-Sormax of the boundary imbibition curve (see Fig-3, default cycle). In fact,
when historical maximum attained oil, Soh, of a block is close enough to Sorw, the created Pcow
scanning curves will tend towards a vertical line between Pcow(a) and Pcow(b) at 1-Sorw. This
discontinuity could be a source of numerical instability, depending on the difference between
Pcow(a) and Pcow(b).
80
Hysteresis Pcow
DampedCycle
60
DefaultCycle
DampedCycle
40
(a)PrimaryDrainage
20
1Sorw
1Soh
Sw
0
0
20
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
Pcow(1Sorh)
40
60
1Sormax
1Sorh
Figure 3: Damped Pcow Curves
User's Guide IMEX
Tutorial 185
The optional keyword *DAMP_PCOW_TROIL in Rock-Fluid Data section can be used to

reduce the vertical extent of the Pcow cycle. As shown by Fig-3, the damped Pcow is formed
by shrinking the default Pcow with a constant ratio and meanwhile placing the Pcow(1-Sorh) onto
the straight line between Pcow(b,1-Sormax) and Pcow(a,1-Sorw).
Sormax for Grid Blocks and Well Layers

The maximum residual oil can be assigned to each grid block by using the keyword
*SORMAX. *SORMAX can be entered in the Rock-Fluid Data section. For example:
*SORMAX *ALL
2000 * 0.35
2500 * 0.40
Please note the assignment of *SORMAX to a block is effective only when the rock type uses
Sormax in a calculation. That includes two cases: (1) the block has trapped oil hysteresis option
enabled or (2) the block does not use trapped oil hysteresis however it has the Krow hysteresis
activated by keyword *HYSKRO.
*SORMAX is only used to adjust the already existing Sormax (from the rock table). It cannot
be used to assign a Sormax to a table which did not already have a Sormax defined by data.
Keyword *SORMAX can also be applied in recurrent data. However, it should be used with
caution because *SORMAX invokes recalculation of Sorh, which can cause a discontinuity of
the trapped oil saturation. Similar difficulties can occur if the block Swcon or Sorw is changed in
recurrent data when hysteresis is used.
For well layers, Sormax can also be applied as an end point if the rock type of the well has
trapped oil hysteresis activated. For a well layer, the subkeyword *SORMAX can be used in
the *KRPERF keyword, for example:
*KRPERF
10
10
0.35
1:20 *SORMAX
Please note the *SORMAX defined in *KRPERF only effects Krow hysterisis. Capillary
pressure effects are not included in the well equations pressure drop.
Remarks and Restrictions

In this documentation, descriptions are based on a water wet reservoir. The approach can be
applied to oil wet reservoirs if the oil and water are read as non-wetting phase and
wetting phase respectively. Therefore the trapped oil is replaced by trapped non-wetting
phase and Sormax refers to the maximum residual non-wetting phase saturation.
If gas exists in the block, the gas is counted as the non-wetting phase no matter if the
reservoir is water wet or oil wet. The Krow is looked up with respect to wetting phase
saturation. Hence it will be inconsistent if the 3-phase Kro is looked up using Stones method
with the *SO option enabled (see*KROIL). For the same reason, the *KROIL
*SEGREGATED option does not work with the trapped oil hysterisis option.
186 Tutorial
User's Guide IMEX
The trapped oil hysterisis option assumes the input primary drainage curve is used to
calculate the initial equilibrium Pcow. The option that specifies the use of the
boundary/primary imbibition Pcow as the initial equilibrium capillary pressure is inconsistent
with the approach, so the use of the keyword *RPT set number *IMBIBITION will be
flagged as an error.
The rock type of a grid block should not be altered in recurrent data or altered in the Rock
Fluid section of a restart run when the trapped oil option is enabled. The historically
maximum attained oil saturation of a block may not be consistent if rock types are altered
during a run.
Output of Sormax and Sorh Distributions

Block Sormax and Sorh can be output to simulation report as well as to the Results SR2 file, if
any part of the reservoir is modeled using trapped oil hysteresis. The subkeywords
*SORMAX and *SORH can follow after either *OUTPRN *GRID or *OUTSRF *GRID
keyword in the IO data section.
User's Guide IMEX
Tutorial 187
Keyword Data Entry System
Introduction to Keyword System

INTRODUCTION
In a keyword input system, each data item or group is preceded by a keyword indicating what
that data item or group is. For example,
*MAXERROR 10
indicates that a maximum of 10 data entry errors are allowed before the simulator stops. Many
data items have defaults, which are used if the keyword is not found in the input data file.
CHARACTER SET
There is a set of allowed characters that may be used in referring to keywords. Any character
not in this set will be interpreted as a blank. Characters in quotes or comments are not
checked, but will be passed along unchanged to the output.
The purpose of the character set is to detect invisible non-blank characters, such as tab, which
some editors may insert in your data file.
The CMG keywords are composed of the upper and lower case alphabet, numerals 0-9,
keyword indicator (*), and arithmetic operators (=, +, -, /). Extra characters are included in
the set to accommodate the *TRANSLATE facility (see below).
You may increase the character set at installation time by expanding the data definition of the
array CHRSET in subroutine RDLINE in the simulator source code. The only restriction is
that the characters must be supported by the computer operating system.
KEYWORD INDICATOR
The optional keyword indicator * (asterisk) may appear immediately before the keyword with
no blanks between.
An example of a keyword is the porosity keyword:
por or POR or *POR
In this manual, keywords are shown in capitals with '*' so that they stand out in the text.
However, mixed case, and without '*', are allowed.
Two keyword indicators or asterisks, in a row, indicate a comment line, as in:
** This is a comment line. The comment line may
** appear almost anywhere in the data set. It is
** very useful for documenting your data set.
User's Guide IMEX
Keyword Data Entry System 189
The comment indicator may be changed by using the *COMMENT keyword described later
in this section.
ORDER OF KEYWORDS
All keywords used in the keyword input system are grouped into keyword groups.
Keyword groups must appear in the data file in the same order as they appear in this
document. Keywords within a keyword group may appear in any order, unless specifically
stated otherwise.
There are a few keywords which may appear at any point in the data file. These keywords
are *LIST, *NOLIST, *INCLUDE, *COMMENT and *TRANSLATE.
Some keywords may appear both within their keyword group, and in recurrent data.
The description of each keyword notes whether the keyword is optional or required. Some
keywords are optional or required with the use of certain other keywords. Optional keywords
have default values which are used if the keyword is not found in the data file.
STRINGING KEYWORDS
Keywords that are not subkeywords of a keyword or keywords that are in fact separate
entities under one keyword heading cannot be strung along on one line.
The rule is one keyword, with its valid subkeyword or subkeywords, per line, unless
otherwise stated or illustrated.
For example, the aquifer option requires 3 keywords. *AQUIFER has a list of subkeywords
and requires the location of the aquifer. The option also needs aquifer properties
(*AQPROP) and the aquifer influence function (*AQFUNC). The three keywords are
separate entities and may not be strung along on the same line.
An incorrect presentation of the first two keywords is:
*AQUIFER *BOUNDARY *AQPROP
*AQFUNC
0.01 0.112
...
...
1000
.4
.002
230
1.0
The correct presentation of these 3 keywords is:

** The keyword *BOUNDARY is a designated subkeyword
** of *AQUIFER and so can be strung on the same
** line as *AQUIFER.
*AQUIFER *BOUNDARY
*AQPROP
** thickness
porosity
permeability
radius
angle
1000.0
.4
.002
230.0
1.0
*AQFUNC
** dimensionless
influence
**
time
function
0.01
0.112
...
...
190 Keyword Data Entry System
User's Guide IMEX
CASE
Keywords and alphanumerical strings may be in upper case, lower case, or any combination.
Filenames must conform to the requirements of the operating system being used, for example,
upper case for IBM mainframe systems.
LINE LENGTH
Any line in a data file has a maximum length of 512 characters. Any characters after number
512 are ignored. This maximum input length may be changed by changing the parameter
MDLINE in the source code file KWCOM1.INC, and recompiling the simulator.
DELIMITERS
Keywords, numbers, and character strings must be separated from each other by blanks,
commas, or new line characters. Consecutive commas, with nothing except blanks between
them should not occur in the data file.
CHARACTER STRINGS
Character strings ALWAYS must be enclosed in a pair of single quotes (e.g. '5-35-48-W5').
Embedded single quotes are not allowed; however, double quotes (") may be used inside
single quotes, e.g., 'This is the "right" way'.
TRANSLATION
You can use your own keyword for any main keyword if you define the translation rule using
*TRANSLATE.
NUMBERS
Numbers are input in free format. Real numbers do not require decimal points.
Exponentiation is indicated by 'E', 'e', 'D' or 'd'. Numbers must not contain embedded blanks.
If an integer is expected, and a number with a decimal fraction is read in, an error message
will be issued, and the program will stop.
The following are examples of valid real numbers:
25.040
-3
1.23E+02
0.02D-4
34.e02
+2.3
+.3
-.3
The following are NOT valid real numbers:

34. E 02 <-- blanks in number
- 34.E02 <-- blank in number
34.E.2
<-- decimal in exponent
Sequences of numbers may be separated either by commas or by blank spaces.
User's Guide IMEX
REPEAT COUNT
There is a simple way to input multiple sequential occurrences of a number. Suppose you
have five numbers in order:
.23
.23
.23
.41
.27
There are two ways to input these numbers. One is to write them as they appear directly above.
However a shortcut measure is to write them using the multiple occurrence indicator ("*").
Since the first three numbers in sequence are the same you can write the numbers this way:
3*.23
.41
.27
Note that there MUST NOT be a space either before or after the "*".
INTEGER RANGE
In any instance where a sequence of INTEGER values is required, a colon must be used to
indicate a range of values from one integer to another integer. Blanks cannot be present
between either integer and the colon. For example:
1
2
3
4
6
and
1:4
6
10:12
10
11
12
are two equivalent ways of giving the same sequence of INTEGERS. Note that this method
of input will not work if real numbers are expected.
TABLES
The keyword documentation sometimes indicates that a table of data must be entered. All the
required data items are listed in a particular order. Always enter the data in the order shown
in the documentation.
A value is also expected for each data item. Data items cannot be omitted except where it is
explicitly shown in the documentation that certain items, located at the ends of rows, are optional.
Optional items are shown enclosed within round brackets in the particular table syntax.
An example of such an event includes the water-oil relative permeability tables (*SWT
keyword). Pcow is optional, and need not be entered, but in this case the user has capillary
pressure data.
*SWT
**Sw
0.2
0.2899
0.3778
0.4667
0.5556
0.6782
0.7561
0.8325
0.9222
1.0000
krw
krow
(Pcow)
0.0
0.022
0.018
0.061
0.143
0.289
0.450
0.780
1.000
1.000
1.0
0.6769
0.4153
0.2178
0.0835
0.0123
0.0
0.0
0.0
0.0
45.0
19.03
10.07
4.09
1.80
.50
.10
.0
.0
.0
User's Guide IMEX
If the capillary pressure is not used (Pcow = 0), then the table would be entered as
*SW
**Sw
0.2
0.2899
0.3778
0.4667
0.5556
0.6782
0.7561
0.8325
0.9222
1.0000
krw
krow
0.0
0.022
0.018
0.061
0.143
0.289
0.450
0.780
1.000
1.000
1.0
0.6769
0.4153
0.2178
0.0835
0.0123
0.0
0.0
0.0
0.0
The following table is unacceptable, as items anywhere within the row cannot be omitted:
*SWT
**Sw
0.2
0.2899
0.3778
0.4667
0.5556
0.6782
0.7561
0.8325
0.9222
1.0000
krw
krow
(Pcow)
0.0
0.022
1.0
0.6769
0.4153
0.2178
0.0835
45.0
19.03
10.07
4.09
1.80
.50
.10
.0
0.061
0.143
0.289
0.780
1.000
1.000
0.0
0.0
0.0
0.0
However, use of the *INT keyword (See *INT) to interpolate omitted entries allows tables
such as the one above to be used. For example:
*SWT
**Sw
0.2
0.2899
0.3778
0.4667
0.5556
0.6782
0.7561
0.8325
0.9222
1.0000
User's Guide IMEX
krw
krow
(Pcow)
0.0
0.022
*INT
0.061
0.143
0.289
*INT
0.780
1.000
1.000
1.0
0.6769
0.4153
0.2178
0.0835
*INT
0.0
0.0
0.0
0.0
45.0
19.03
10.07
4.09
1.80
.50
.10
.0
*INT
*INT
ERROR AND WARNING MESSAGES

During data input, the lines in the data file are echoed to the print output file. If an error is
detected, an error message or a warning is issued. Depending on the type of error, the
message may refer to the line printed above or below the error or warning message.
If *NOLIST has been used, the data line on which the error or warning has occurred will not
be printed. It is therefore recommended the *NOLIST only be used for production runs, after
the data has been thoroughly debugged.
User's Guide IMEX
Comments (Optional)
PURPOSE:
** (two keyword indicators) may be used to add comments explaining where data came from,
why options are being used, etc.
FORMAT:
** comment_text
DEFAULTS:
Optional. No defaults.
CONDITIONS:
A comment may appear at any point in the data file.
EXPLANATION:
Two consecutive keyword indicators ('**') indicate the start of comment text. The portion of
the input line after the two keyword indicators is ignored. Comment lines may be used to add
comments explaining where data came from, why options are being used, etc.
Comments are copied to the output print file with the rest of the data file (subject to
*NOLIST and *LIST keywords). Otherwise, comment lines are ignored.
An example of a comment is:
*MAXERROR 14 ** Change maximum number of errors.
User's Guide IMEX
Blank Lines (Optional)

PURPOSE:
Blank lines may be used to separate sections of a data file, and generally make the data file
more readable.
CONDITIONS:
Blank lines may appear at any point in the data file.
EXPLANATION:
Blank lines are copied to the output print file with the rest of the data file (subject to
*NOLIST and *LIST keywords). Otherwise, blank lines are ignored.
User's Guide IMEX
Scan Mode for Checking Errors
*CHECKONLY
PURPOSE:
*CHECKONLY indicates that the well data is to be checked for format errors, but the
simulation will not be run.
FORMAT:
*CHECKONLY
DEFAULTS:
If this keyword is absent, timestep calculations will be done.
CONDITIONS:
This keyword must appear in the Input/Output Control section.
It is also possible to specify a checkonly run from the command line by entering the "checkonly" flag. This use of the checkonly mode has the additional benefit of not requiring
an IMEX run license to function.
Example:
C:\CMG\IMEX\98.00\exe\dyn\mx9800.exe -checkonly
User's Guide IMEX
Include Files (Optional)

PURPOSE:
The *INCLUDE keyword indicates that reading of the primary input data set is suspended.
Instead, a secondary file will be read.
FORMAT:
*INCLUDE 'filename'
DEFAULTS:
Optional. No defaults.
CONDITIONS:
The *INCLUDE keyword must appear on a line by itself. Only one secondary file may be
open at a time. Nesting of *INCLUDE keywords is not allowed.
EXPLANATION:
Enclose the include filename in single quotes.
When a *INCLUDE keyword is encountered, the named secondary input file is opened and
data is read from the file. When the end of the secondary file is reached, the secondary file is
closed and data reading continues in the primary (or original) input file.
User's Guide IMEX
Controlling Data File Listing (Optional)
*LIST, *NOLIST
PURPOSE:
*LIST specifies listing the input data file, from this point forward, to the output print file. By
default only the first 20 lines of input data is listed unless the keyword *NOLISTLIM is
found in the Input/Output Control section (See *NOLISTLIM below)
*NOLIST specifies not listing the input data file to the output print file, starting immediately
after the current line.
FORMAT:
*LIST
*NOLIST
DEFAULTS:
Optional keywords. Default: *LIST
CONDITIONS:
*LIST or *NOLIST may appear at any point in the data file, but must be on a line by itself.
EXPLANATION:
By default, the entire data file is listed to the output print file prior to the start of the
simulation run. When large arrays are listed under *LIST, only the first 20 lines of data are
echoed, in order to echo all lines of data (this can result in large files) the *NOLISTLIM
keyword in the Input/Output Control section is required. If a *NOLIST keyword is inserted
in the data file, the data file is not listed from the point of the *NOLIST keyword until a
*LIST keyword or the end of data file is reached.
User's Guide IMEX
Controlling Data File Listing Using NOLISTLIM (Optional)

*NOLISTLIM
PURPOSE:
*NOLISTLIM specifies that the default 20 lines of keyword echoing under the *LIST
keyword be extended so that all values of the keyword are echoed regardless of the number of
lines required to do so. This can create large files if used, but allows the user to view all of
the data.
FORMAT:
*NOLISTLIM
DEFAULTS:
Optional keywords. Defaults to a list limit of 20 lines per keyword
CONDITIONS:
*NOLISTLIM must appear on a line by itself in the Input/Output Control Section.
EXPLANATION:
simulation run. When keywords containing large amounts of data are listed under *LIST,
only the first 20 lines of data are echoed. In order to echo all lines of data (this can result in
large files) the *NOLISTLIM keyword in the Input/Output Control section is required.
User's Guide IMEX
Changing the Comment Indicator (Optional)
*COMMENT
PURPOSE:
*COMMENT changes the two character sequence that denotes the beginning of a comment.
FORMAT:
*COMMENT 'ab'
DEFINITION:
ab
A two-character string denoting the start of a comment. The string 'ab' must
be enclosed in quotes.
DEFAULTS:
Optional keyword. Default: *COMMENT '**'
CONDITIONS:
*COMMENT may appear at any point in the data file, but must be on a line by itself. All
subsequent comments following the appearance of the *COMMENT keyword must be
preceded by the two-character sequence 'ab'.
EXPLANATION:
By default, comments in the data file are denoted by the character string '**'. This may be
changed by using the *COMMENT keyword.
Example:
*COMMENT '--'
*TRANSLATE 'KX' 'PERMI' -- This is a translate
-- rule
From this point on in the data file all comments should begin with '--'. In the above example
the two lines beginning with '--' are comments.
User's Guide IMEX
Changing the Keywords by Using Translate Rules (Optional)

*TRANSLATE
PURPOSE:
*TRANSLATE changes or translates your own favorite keyword into a CMG simulator
recognizable keyword.
FORMAT:
*TRANSLATE 'your_keyword' 'CMG_keyword'
DEFINITION:
your_keyword
A single-word keyword that you want the simulator to recognize. The
allowed characters are those in the character set specifier in subroutine
RDLINE in the simulator source code; no blanks, commas or asterisks are
allowed. You may add any character that your computer operating system
supports to this character set. Enclose the string in single quotes.
CMG_keyword
The CMG simulator keyword (WITHOUT asterisk) that you want to replace.
This must be a valid keyword recognized by the simulator. This must be
enclosed in single quotes.
DEFAULTS:
Optional keyword. Default: Use the internal simulator keywords.
CONDITIONS:
*TRANSLATE may appear at any point in the data file, but must be on a line by itself.
Subsequently, a simulator keyword may be referred to by using either 'your_keyword'
(defined by a *TRANSLATE keyword) definition or the internal simulator keyword
'CMG_keyword'.
EXPLANATION:
If you need to redefine a keyword because you want to make the keyword more meaningful to
yourself, or simply for convenience, the *TRANSLATE keyword will accomplish this task.
Example:
*TRANSLATE
'KX'
'PERMI'
This translate rule translates the *KX or KX keyword such that the simulator recognizes this
to mean *PERMI. Subsequent to this keyword *KX, KX, *PERMI, or PERMI may be used
to refer to the *PERMI keyword.
A keyword may have more than one translate rule,
Example:
*TRANSLATE 'KX' 'PERMI'
*TRANSLATE 'x_permeability' 'PERMI'
*TRANSLATE 'permx'
'PERMI'
User's Guide IMEX
Input of Grid Property Arrays

ARRAY
Grid properties which are input are, in fact, arrays of data with one array element for each
grid block. Grid properties are indicated by 'ARRAY:' in the left column immediately
following the title on the manual page which describes them.
ARRAY READING OPTIONS
An array assignment consists of five parts, two of which are optional. The syntax is:
grid_array (array_qualifier) read_option data (array_modifier)
DEFINITIONS:
grid_array
The property being assigned, such as *POR. In the manual this is denoted as
ARRAY: *POR
array_qualifier
This is used to assign data to different elements of the grid block (e.g., matrix
and fracture). The array_qualifier is optional. Choices are
*MATRIX
*FRACTURE
*RG block_address
*EQUALSI
If no array_qualifier is present then *MATRIX is assumed. These
array_qualifier keywords are described separately.
Each of the above array reading qualifiers will access only the element
indicated. The user must ensure that all elements of each grid block have
been assigned required data.
read_option
The read options are
*CON
*IVAR
*JVAR
*KVAR
*ALL
*IJK
These read_option keywords are described separately.
User's Guide IMEX
data
These are the actual values for the grid_array. The amount of data depends
on the read_option.
array_modifier
Once an array has been input, it can be modified immediately using *MOD.
This allows modification of blocks or regions after the read_option is done.
The *MOD keyword is described separately.
User's Guide IMEX
Entering Matrix Grid Properties
*MATRIX
PURPOSE:
*MATRIX is used immediately after a grid property keyword to indicate that a matrix
property is being input.
KEYWORD:
*MATRIX
EXPLANATION:
Note that *MATRIX is the default for entering all grid properties if no array input qualifier is
present. Therefore, you need to use *MATRIX only when the natural fracture option has
been enabled.
Any of the array reading options can be used with *MATRIX. The read_option keyword
must follow the *MATRIX keyword.
Example: To input the matrix porosity in a dual porosity system:
*POR *MATRIX *ALL
.12 5*.16 .18 .22 .21 8*.20
.19 10*.18 3*.21 .19 .16
User's Guide IMEX
Entering Fracture Grid Properties
*FRACTURE
PURPOSE:
*FRACTURE is used immediately after a grid property keyword in a dual porosity system to
indicate that a fracture property is being input.
KEYWORD:
*FRACTURE
EXPLANATION:
Any of the array reading options can be used with *FRACTURE. The array reading option
keyword must follow the *FRACTURE keyword.
Example: Suppose the planes of grid blocks with J = 2 and J = 3 are fractured. You want to
input the fracture porosities of these blocks.
*POR
1:10
*FRACTURE *IJK
2:3 1:3 .08
User's Guide IMEX
Entering Refined Grid Properties
*RG
PURPOSE:
*RG is used to assign values of an array to refined grid blocks.
KEYWORD:
*RG i j k
DEFINITIONS:
i
I direction index of the fundamental grid block containing the refined grid.
j
J direction index of the fundamental grid block containing the refined grid.
k
K direction index of the fundamental grid block containing the refined grid.
EXPLANATION:
Refined grids are initially defined using the *REFINE keyword in the Reservoir Description
section.
By default, all refined grid blocks are assigned the values assigned to the fundamental grid
block. The *RG keyword allows input of different values for each refined grid block.
Any of the array reading options may be used with *RG. The array of properties input is that
of the refined grid, and corresponds to the number of blocks in the refined grid, not the
fundamental grid. The array reading option keyword must follow the *RG keyword.
Example: Suppose block (1,1,1) contains a refined grid. It is divided up into ni=3, nj=2 and
nk=1. You want to input the porosity of each of the refined grid blocks.
*POR *RG 1 1 1 *ALL
.08 .079 .078 .081
User's Guide IMEX
.08
.076
J and K Direction Data from I Direction
*EQUALSI
PURPOSE:
*EQUALSI indicates that values in the J and K directions are the same as those in the I
direction, or that the values given for the I direction may be modified by division,
multiplication, etc.
KEYWORD:
*EQUALSI
EXPLANATION:
*EQUALSI is used with direction-dependent keywords, such as the transmissibility,
permeability and dispersion coefficients. The I direction keyword must have all of its
elements (matrix, fracture, wellbore, etc.) assigned before *EQUALSI can be used with the J
and K direction keywords.
Example: Permeabilities in a single-porosity system. J-direction values are equal to the I
direction, but the K-direction values are twice the I-direction values
*PERMI *CON 100.0
*PERMJ *EQUALSI
*PERMK *EQUALSI * 2.
Example: The same as above, only with a natural fracture option in effect.
*PERMI
*PERMI
*PERMJ
*PERMK
*MATRIX *CON 100.0

*FRACTURE *CON 10000
*EQUALSI
*EQUALSI * 2.
User's Guide IMEX
Constant Value Arrays
*CON
PURPOSE:
*CON indicates that a constant value is entered for all array elements. The value may be
entered on the same line or the next line.
KEYWORD:
*CON value
EXPLANATION:
Example: Assume you have a reservoir with a constant value of porosity of 0.16, and a
constant permeability in the I direction of 100 md.
*POR *CON
0.16
*PERMI *CON
User's Guide IMEX
100.
Array Input in IJK Notation
*IJK
PURPOSE:
*IJK assigns a constant value of a grid property within the region defined by the minimum
and maximum block number in each of the three directions.
KEYWORD:
*IJK
i1:i2 j1:j2 k1:k2 value
DEFINITIONS:
i1
Initial I direction grid block index.
i2
Final I direction grid block index.
j1
Initial J direction grid block index.
j2
Final J direction grid block index.
k1
Initial K direction grid block index.
k2
Final K direction grid block index.
value
Constant value of the array for the defined region.
EXPLANATION:
The *IJK array reading option assigns a constant value of a grid property within the region
defined by the minimum and maximum block number in each of the three directions. Later
lines will overwrite previous lines if they refer to the same grid blocks.
You must input enough lines so that every grid block has been assigned a value.
If both the initial and final grid block indices in any direction are the same, the colon and
final grid block index may be omitted. For example, a porosity of .2 exists in a grid region
given by i1=1, i2=3, j1=3, j2=4, and k1=k2=2. Input may be either:
1:3
or
1:3
3:4
2:2
.2
3:4
.2
User's Guide IMEX
Example: Suppose you have a reservoir with a porosity of 0.16 everywhere except for an
interior region where the porosity is 0.22. You are using a grid where ni=20, nj=25, and nk=6
and the region of higher porosity is between grid block indices I=5 and I=9, J=14 and J=23
and for K=4. First, assign the entire grid with a porosity of 0.16. Then overwrite the interior
region with its higher porosity.
*POR *IJK
1:20 1:25
5:9 14:23
User's Guide IMEX
1:6
4
0.16
0.22
Array Input of Values that Vary in the I Direction
*IVAR
PURPOSE:
*IVAR is used to indicate values that vary in the I direction, but which are constant in the
other two directions.
KEYWORD:
*IVAR value(1) ... value(ni)
DEFINITIONS:
value(1)
Value assigned to all grid blocks with an I direction index of 1.
ni
Number of grid blocks in the I direction.
EXPLANATION:
Enter nj values separated by spaces or commas.
Example: I direction block sizes where ni = 10:
*DI *IVAR
2*1000 1100
1050
3*800
860
1010
1100
Note that the structure '2*1000' indicates the value '1000' occurs twice.
Example: I direction block sizes where ni = 3:
*DI
*IVAR
3000.0
4000.0
5000.0
User's Guide IMEX
Array Input of Values that Vary in the J Direction
*JVAR
PURPOSE:
*JVAR is used to indicate values that vary in the J direction, but which are constant in the
KEYWORD:
*JVAR value(1) ... value(nj)
DEFINITIONS:
value(1)
Value assigned to all grid blocks with a J direction index of 1.
nj
Number of grid blocks in the J direction.
EXPLANATION:
Enter nj values separated by spaces or commas.
Example: The J direction increments for a problem where nj=10 are: 755, 755, 755, 825, 825,
1000, 1000,1100,800,800.
*DJ
*JVAR
3*755
2*825
2*1000
1100
2*800
Example: The J direction has just 3 blocks:

*DJ
*JVAR
User's Guide IMEX
3000.0
4000.0
3000.0
Array Input of Values that Vary in the K Direction
*KVAR
PURPOSE:
*KVAR is used to indicate values that vary in the K direction, but which are constant in the
KEYWORD:
*KVAR value(1) ... value(nk)
DEFINITIONS:
value(1)
Value assigned to all grid blocks with a K direction index of 1.
nk
Number of grid blocks in the K direction.
EXPLANATION:
Enter nk values separated by spaces or commas. This is convenient for entering properties
vary only by layer.
Example: Porosity varies for each of the layers of a system where nk=5, but is constant
within each layer. The layer porosities are: .0810, .210, .180, .157, and .200.
*POR *KVAR .081 .21 .18 .157 .2
Example:
**
**
**
**
**
Each of the I, J, and K permeabilities

are constant within each layer of the
reservoir but vary from layer to layer.
Hence use *KVAR to input them layer
by layer.
*PERMI *KVAR 200.0 50.0 500.0

*PERMJ *KVAR 200.0 50.0 500.0
*PERMK *KVAR 20.0 40.0 60.0
User's Guide IMEX
Values that Vary for Most or All Grid Blocks
*ALL
PURPOSE:
*ALL is used to indicate that values vary in most or all the grid blocks. The number of values
expected is the number of grid blocks in the grid, including all null or zero-porosity blocks.
KEYWORD:
*ALL value(1) ... value(ni*nj*nk)
EXPLANATION:
Values are entered starting with block (1,1,1) and in increasing block order where the I
direction block index increases fastest and then the J direction block index second fastest and
the K direction block index the slowest.
See Figure 1 in Appendix D for the ordering of grid blocks.
Example: Porosities for each grid block in a three-dimensional system vary in almost every
grid block: ni=10, nj=3, nk=2
*POR
.08
.15
.074
.095
.11
.08
*ALL
.08
.134
.12
.13
.12
.09
User's Guide IMEX
.081
.08
.12
.12
.134
.144
.09
.087
.154
.157
.157
.143
.12
.157
.167
.17
.157
.123
.15
.145
.187
.18
.18
.16
.09
.12
.121
.184
.18
.165
.097
.135
.122
.122
.098
.102
.087
.18
.08
.084
.09
.10
.011
.092
.08
.09
.09
.10
Modifying Array Data (Conditional)
*MOD
PURPOSE:
*MOD indicates the modification of an input grid property.
FORMAT:
*MOD
i1:i2
-or*MOD
j1:j2
k1:k2
(*)
(-)
(+)
(/)
(=)
value
(+)
(-)
(*)
(/)
(=)
value
DEFINITIONS:
i1
Initial I direction grid block index for the region to modify.
i2
Final I direction grid block index for the region to modify.
j1
Initial J direction grid block index for the region to modify.
j2
Final J direction grid block index for the region to modify.
k1
Initial K direction grid block index for the region to modify.
k2
Final K direction grid block index for the region to modify.
+
Indicates that value is added to the existing grid property value in the region
defined.
User's Guide IMEX
Indicates that value is subtracted from the existing grid property value in the
region defined.
*
Indicates that the existing grid property is multiplied by value in the region
defined.
/
Indicates that the existing grid property is divided by value in the region
defined.
=
Indicates that the existing grid property is overwritten by value in the region
defined.
CONDITIONS:
The *MOD keyword must appear immediately after the array property has been input.
Only one mod keyword can follow a property, but that mod keyword may refer to multiple
regions to modify if the format which enters i, j, k ranges is used (as in the example
immediately below).
EXPLANATION:
The *MOD option is used to modify the last grid property data array input by adding,
subtracting, multiplying, dividing or replacing array elements by a number.
If the operator follows immediately after the *MOD keyword, then the modification is
applied to all grid blocks.
If the initial and final grid block indices in one direction are the same, the colon and final
index may be omitted.
Example: Suppose for a 10 x 6 x 1 grid you want to modify the porosities in the region with I
indices 1 through 3, J indices 1 through 4 and with K index of 1 by adding 0.01. You further
wish to assign the value of .13 to the block with I=5, J=2, and K=1. Enter *MOD after the
array values. The data looks like this:
*POR *ALL
.08 .08
.15 .134
.074 .12
.095 .13
.11 .12
.08 .09
*MOD
1:3 1:4 1
5
2
1
User's Guide IMEX
.081
.08
.12
.12
.134
.144
.09
.087
.154
.157
.157
.143
.12
.157
.167
.17
.157
.123
.15
.145
.187
.18
.18
.16
.09
.12
.121
.184
.18
.165
.097
.135
.122
.122
.098
.102
.087
.18
.08
.084
.09
.10
.011
.092
.08
.09
.09
.10
+ .01
= .13
Example: To modify the entire grid to reduce the porosity of each grid block to 95% of the
original value:
*POR
.08
.15
.074
.095
.11
.08
*MOD
*ALL
.08 .081
.134 .08
.12 .12
.13 .12
.12 .134
.09 .144
* .95
.09
.087
.154
.157
.157
.143
.12
.157
.167
.17
.157
.123
.15
.145
.187
.18
.18
.16
.09
.12
.121
.184
.18
.165
.097
.135
.122
.122
.098
.102
.087
.18
.08
.084
.09
.10
.011
.092
.08
.09
.09
.10
You are not permitted to repeat a required array keyword after it has been entered once.
For example, the following data entry is incorrect. The user wants to change some of the
porosities to 0.22 after initially assigning 0.30 to all grid blocks:
*POR *CON 0.3
*POR *IJK
5:8 14:23 4
0.22
The correct procedure is to use the *MOD keyword on the line immediately following *POR:
*POR *CON 0.3
*MOD 5:8 14:23
4 = 0.22
User's Guide IMEX
Interpolating Table Data (Optional)
*INT
PURPOSE:
*INT indicates that the corresponding table entry should be filled by interpolation.
EXPLANATION:
The *INT keyword may be used in table input. This keyword enables the calculation of the
table entry by interpolation. Essentially the table entry corresponding to *INT is replaced by
a linearly interpolated value. This option is useful when not all table entries are known. This
feature is explained in further detail with the help of an example.
Suppose that it is required to enter a water-oil relative permeability table into the simulator.
Also assume that the water and oil relative-permeabilities are known at different saturations
*SWT
**Sw
0.2
0.3
0.4
0.5
0.6
0.7
0.8
1.0
Krw
Krow
0.0
0.05
*INT
0.40
*INT
0.8
1.0
1.0
1.0
*INT
0.7
*INT
0.5
*INT
0.0
0.0
In the above table values denoted by *INT will be calculated by linear interpolation by the
simulator.
NOTE:
Interpolation is done with respect to the first column. Thus the *INT keyword cannot appear
in the first column of the table.
The *INT option can be used for extrapolation, however the use of *INT may violate
saturation table end point checks (due to the extrapolation). The use of *INT for
extrapolation on saturation tables is not advised.
The input of only one non *INT entry is handled as a special case to allow the user to set an
entire column to the same value.
User's Guide IMEX
Command Line Input/Output Control

IMEX has several useful command line options.
-f input_data
Defines the input-data-file.
-r input_restart
Defines the restart irf file.
-checkonly
Runs the model in check only mode. An IMEX license is not

required for a check only run specified from the command line.
-dimsum
Prints detailed information on model dimensioning at run time.
-onestep
Runs the model for one timestep only. See the *MAXSTEPS
description in the Numerical Methods section for more details.
-maxsteps
Sets the maximum timestep number to n. See the *MAXSTEPS

description in the Numerical Methods section for more details.
-wait
Wait for License

If all available licenses are being used, this argument keeps the
process in a sleep mode until a license is available (up to 72 hrs.)
-dd
Output files will be written to the directory that contains the data
file. This option is intended to be used when an absolute pathname
has been supplied via the -f argument.
-wd path
Output files will be written to the directory given by path. This

option is useful in an environment where the current directory
may not be defined.
User's Guide IMEX
Input/Output Control 221
-log (log_path)
Specifies that consol diary output will be redirected to a file. If

log_path is present, this file has pathname log_path and extension
.log will be added if it is not already present. If log_path is
absent, this files name has the same base as the output files but
extension .log. This file will not contain error or status messages
from the operating system. Screen input is read from a file which
has the root name of the data set and the extension ".in". If a file
name has not been defined using the f command line option and
the log option is used, it can be entered in the ".in" file. In
addition, the user *INTERRUPT *INTERACTIVE choice is also
entered in the ".in" file when the log option is used. A carriage
return in the ".in" file where the simulator is expecting to read the
interrupt option defaults to *RESTART-STOP.
-doms
Used to enable Jacobian building parallelization from the

command-line, its use requires that you have a license with
Parallel IMEX enabled (see *DPLANES in the Numerical
Methods section for more details). "-jacpar" is now recommended.
-parasol n
Used to enable Solver parallelization from the command-line for a

target of n logical CPU's, its use requires that you have a license
with Parallel IMEX enabled (see *SOLVER and *PPATTERN in
the Numerical Methods section for more details).
-aimsol
Overrides numerical keyword specifying that PARASOL is to be

used as the solver.
-jacdoms n
Used to enable *JACDOMS Jacobian building parallelization from

the command-line for a target of n logical CPU's, its use requires
that you have a license with Parallel IMEX enabled (see
*JACDOMS in the Numerical Methods section for more details).
"-jacpar" is now recommended.
-jacpar
Used to enable *JACPAR *ON Jacobian building parallelizatin

from the command line. This option uses Jacobian domains based
on the Parasol classes. It uses the same number of threads as
specified by *PARASOL or *PNTHRDS options.
Its use requires that you have a license for Parallel IMEX.
-pnthrds n
Used to enable *PNTHRDS n, where n is the desired number of

threads. This option may be used to override the default number of
threads specified using other options.
-cputime
This command line option enables non-intrusive clock and CPU

time profiling of IMEX equivalent to the keywords *DEBUG
*CPUTIME *SOLVER.
Please note that LAUNCHER enables the parallel option by using command line options
when submitting batch jobs. The command line options will override selected parallel
keywords in the data set.
222 Input/Output Control
User's Guide IMEX
If you want to ensure that the parallel control keywords in the dataset are always honored, it
is necessary to run the Simulator from LAUNCHER as if in serial mode (specify that you
want to use only one processor).
If this is done, all parallel keywords in the NUMERICAL section of the data set will be
honored. Obviously when doing this ensure the NUMERICAL section completely defines
the parallel options to be used.
User's Guide IMEX
Values Stored in Binary Form
*BINARY_DATA
PURPOSE:
Builder uses this keyword to indicate that grid-array data is stored in binary form.
KEYWORD:
*BINARY_DATA
EXPLANATION:
In Builder
Normally Builder writes data in text-format files. However, Builder is able to write some
grid definition and property data in binary form to a separate binary-format file. This option
is invoked in Builder via menu "File/Save As.../Array Saving Method/Binary File Format
(*.cmgbin)". The binary file is saved in the same folder as the main dataset file and given the
same root name but extension .cmgbin. Unlike the *INCLUDE facility which can involve
multiple include files, there is at most one binary format file associated with a main data file.
The grid definition and property data that can be written in binary form are (1) corner-point
definition keywords *XCORN, *YCORN, *ZCORN, *COORD and *CORNERS, and (2) all
grid-array keywords using read option *ALL in the initialization (non-recurrent) data
sections. In any given data set, all such data is written in the same form (text or binary)
according to the selected Array Saving Method. You may switch between the binary and textonly writing formats whenever you wish.
For grid property keywords, only that data associated directly with subkeyword *ALL is
written in binary form. Specifically, *MOD data lines are preserved as text after the
*BINARY_DATA subkeyword. Therefore, you may add or modify *MOD data lines after
the *BINARY_DATA subkeyword, just as you would after *ALL and its data.
Non-uniform grid property data from sources like maps or existing simulator results from
SR2 file sets usually are written in *ALL format in the text file. Such data can be written
directly to the binary file, avoiding *ALL text writing altogether.
Writing non-uniform data to a binary file has some distinct advantages. First, the reading of
binary data is much faster than text and so for large models can speed up significantly the
transfer of data to and from Builder. Second, binary format occupies less space than a
comparative text representation (e.g., 8 bytes versus 20 to 30 bytes). Third, the original
precision of data obtained from existing SR2 result files can be preserved by not passing the
data through a text-writing step.
In the Simulator
When the simulator detects subkeyword *BINARY_DATA in its initial data scan, it opens
the associated binary file which is assumed to have the same root name as the main data file
but extension .cmgbin. Each time the simulator encounters *BINARY_DATA during the
data loading pass, it locates that property in the binary file and reads one value for each block
in the grid, similar to the *ALL option. A mismatch between the text and binary parts of the
data set will result in an error.
User's Guide IMEX
The reading of binary data is much faster than text and so for large models can speed up
significantly data reading in the simulator. Text formatted data can differ slightly in value
from its associated binary data, so text and binary versions of the same data may give slightly
different simulation results.
Examples
These are examples of data fragments you might see in the text data file written by Builder in
binary file format.
** Null block distribution from map
*NULL *BINARY_DATA
** Permeability from map, with matching adjustments
*PERMI *BINARY_DATA
*MOD 1:5 1:10 1:5 * 0.9
1:3 1:4 1:2 * 1.2
*PERMJ *EQUALSI
*PERMK *EQUALSI / 5.
** Natural fracture: matrix porosity from map
*POR *MATRIX *BINARY_DATA
*POR *FRACTURE *CON 0.008
** Initial saturation, natural fracture system
*SW *MATRIX *BINARY_DATA
*SW *FRACTURE *BINARY_DATA
** Initial solution gas from primary production
*MFRAC_OIL 'SOLN GAS' *BINARY_DATA
*MOD 4 5 1:9 = 0.3 ** Enriched zone around well
User's Guide IMEX
ASCII Characters
IMEX Data File and Directories
It is crucial that IMEX data files and data/output directories do not include non-ASCII
characters (such as accents, Greek symbols, etc). The IMEX data line interpreter does not
handle such characters properly.
Please ensure that IMEX data set names and the complete data and output directory paths
(from the Drive Letter to the data/output directories) only include standard ASCII text.
User's Guide IMEX
Input/Output File Names (Optional)
*FILENAMES
PURPOSE:
Specify names for input and output files.
FORMAT:
*FILENAME(S) file_types name_options
where file_types are one or more of:
*OUTPUT
*INDEX-OUT
*MAIN-RESULTS-OUT
*REWIND-OUT
*INDEX-IN
*MAIN-RESULTS-IN
*REWIND-IN
For file_types *INDEX-OUT, *MAIN-RESULTS-OUT, *REWIND-OUT, *MAINRESULTS-IN and *REWIND-IN one of these name_options may be specified.
''
'filename'
*PROMPT
For file_type *OUTPUT one of these name_options may be specified.
''
'filename'
*PROMPT
*SCREEN
For file_type *INDEX-IN one of these name_options may be specified.
'filename'
*PROMPT
DEFINITIONS:
*FILENAME(S)
File name keyword. The trailing 'S' is optional.
*OUTPUT
Indicates the output file to which formatted simulation results will be written.
*INDEX-OUT
Indicates the index-results-file (irf) to which the simulation results ASCII
data is written.
User's Guide IMEX
*MAIN-RESULTS-OUT
Indicates the main-results-file (mrf) to which the simulation results binary
data is written.
*REWIND-OUT
Indicates the rewindable-results-file (rrf) to which the restart data is written
when the *REWIND option is used.
*INDEX-IN
Indicates the index-results-file from which simulation results and restart
records are read. This file is only necessary for restart runs.
*MAIN-RESULTS-IN
Indicates the main-results-file from which the simulation results and restart
records (binary) are read. This file is only necessary for restart runs.
*REWIND-IN
Indicates the rewindable-results-file from which the rewound restart records
(binary) are read. This file is only necessary for restart runs.
''
Empty string, denoting that the default file name will be used. This is the
default if the *FILENAMES keyword is missing from the input-data-file.
'filename'
A character string which is the file name, limited to 80 characters.
Acceptable file names depend on the operating system being used.
*PROMPT
Indicates that the user will be prompted for this file name via the
screen/keyboard, if the file is required.
*SCREEN
Indicates that this file will go to the screen (standard output device).
DEFAULTS:
If any required input/output file name has not been specified via *FILENAME, then
*PROMPT is assumed. For example, *FILENAME implies *FILENAME *OUTPUT
*PROMPT.
If *FILENAME file_type is used to assign a file name but the name_options list is absent
then '' is used, triggering the default file names, below. For example, *FILENAME
*OUTPUT implies *FILENAME *OUTPUT ''.
User's Guide IMEX
Default File Names

A default file name is used only if the name_option '' is specified. Defaults may be specified
for any of the files independently. However, the default name of some files depends on the
name chosen for other files, e.g., the input-data-file.
A default file name has the form 'xxxxxxxx.yyy', with root 'xxxxxxxx' and suffix 'yyy'. The
suffix depends on the option associated with the file, e.g., 'out' for main output, 'irf' for SR2
index and 'mrf' for SR2 main-results-file. The root is derived from another file name as
follows:
a) discard suffix with '.' delimiter
b) discard everything except the right- most group of alphanumeric and '-' characters
c) discard everything after column 8.
For example, the UNIX pathname '/CMG/DATA/TEST-3.DAT' has the root 'TEST-3'.
The following table summarizes the root and suffix of the default file names.
File
Root Based On
Suffix
*OUTPUT
*INDEX-OUT
*MAIN-RESULTS-OUT
*REWIND-OUT
*INDEX-IN
*MAIN-RESULTS-IN
*REWIND-IN
data file name

*OUTPUT
*OUTPUT
*OUTPUT
(no default)
*INDEX-IN
*INDEX-IN
.out
.irf
.mrf
.rrf
.irf
.mrf
.rrf
With this defaulting system, the user is able to perform a series of 'bootstrapped' restart runs
by changing only the data file name and *INDEX-IN file names for each run.
Example #1: Data file is 'cycle.dat', so
*OUTPUT default is 'cycle.out'.
If the *OUTPUT default is used, then
*INDEX-OUT default is 'cycle.irf' and
*MAIN-RESULTS-OUT default is
'cycle.mrf'.
Example #2: Data file is 'cycle.dat'. The first run segment has
*FILENAME *OUTPUT 'trial1'.
*INDEX-OUT default is 'trial1.irf'
'trial1.srf'
For the second run segment, modify 'cycle.dat' to add the *RESTART keyword, and use
*FILENAME *OUTPUT 'trial2.out', and
*FILENAME *INDEX-IN 'trial1.irf'.
User's Guide IMEX
Do not rename or delete files.

*INDEX-OUT default is 'trial2.irf'
'trial2.mrf'
*MAIN-RESULTS-IN default is
'trial1.mrf'
CONDITIONS:
*FILENAMES must be the first keyword(s) in the input-data-file. If not, the user will be
prompted for the file names.
File names must be the first keywords in the Input/Output Control section of an input-datafile.
All output files are opened with 'UNKNOWN' status and therefore are unprotected from
overwriting.
All input files are opened with 'OLD' status and so must be present for the simulation to
proceed.
EXPLANATION:
CMG's new Simulation Results File System (Generation 2) is also referred to as SR2. The
SR2 file system consists of three files that work together. These are the index-results-file
(irf), the main-results-file (mrf) and the rewindable-results-file (rrf). It is necessary for the
graphics post-processor "RESULTS" that two of these files exist. These are the irf and the
mrf files. These files are also required for restart runs. If the *REWIND option was used to
write restart records, then the rewindable-results-file (rrf) is also required for restart runs.
PLEASE NOTE:
The SR2 file system is incompatible with the previous SRF file system. In order to create an
SRF file from the SR2 file system please use the SR2 to SRF conversion program provided
with this release.
Example:
**
**
**
**
use default file name for index-out-file

'root.irf'.
Prompt for input mrf file *MAIN-RESULTS-IN by
leaving it out of the input-data-file.
*FILENAMES
*OUTPUT
*INDEX-OUT
*MAIN-RESULTS-OUT
*INDEX-IN
'root.out'
'name.mrf'
'input.irf'
User's Guide IMEX
Important Notes (Run Time Dimensioning)

DEFAULTS:
IMEX does not assume any defaults for problem dimensions, however; there are minimum
values which override the automatic allocation procedure. These minimums in turn can be
overridden if the user explicitly specifies small dimensions using the *DIM card.
IMEX is always run time dimensioned. The absence of Dynamic Dimensioning section in
the data file does NOT mean static dimensioning is employed.
In order to maintain backward compatibility, this entire section has been developed as
optional. Users need not enter any of the keywords for the Run Time Dimensioning section,
most existing data sets should work without any difficulty.
The *DIM section will normally be omitted.
User's Guide IMEX
Problem Type Where IMEX Fails to Allocate Adequate /

Appropriate Storage
One type of model causes IMEX to allocate too much memory. This is data which has
pinched out blocks.
Data sets which have blocks pinched out by the grid module cannot have the number of
active blocks accurately determined. IMEX will print a warning and set the number of active
blocks equal to the number of total blocks and continue on with initialization.
After Grid Module initialization IMEX prints out the number of active blocks used (in the
pinched out case --- total blocks) and the minimum value of active blocks in the correct card
format to be able to be pasted directly into the data set directly.
Scan the output file for
*DIM *ACTIVE_BLOCKS
and place this card in your data file if the dimensions used are significantly different from the
minimum dimensions calculated.
A timestep need not be completed for the minimum dimensions to be determined. It is only
necessary to complete the Grid Initialization section of IMEX.
One type of model may cause IMEX to allocate too little temporary storage. These are
problems with a small (less than 20%) percentage of active blocks compared to total blocks
(i.e. NX*NY*NZ).
When IMEX warns that insufficient temporary storage is available, increase the number of
active blocks dimensioned for, using the *DIM *ACTIVE_BLOCKS keyword.
If you need to enter run time dimensioning control data, place this data before the
Input/Output Control section. If there are any *FILENAME keywords that are being used,
the *FILENAME keywords should be placed above the Run Time Dimensioning section.
User's Guide IMEX
Run Time Dimensioning (Optional)
*DIM
PURPOSE:
Specify a dimensioning variable with a value.
FORMAT:
*DIM dim_variable dim_value
where dim_variable are one or more of:
*ACTIVE_BLOCKS
*NULL_BLOCKS
*MAX_FAULTS
*MAX_AQUIFERS
*MAX_SECTORS
*MAX_LEASELINES
*MAXBLKS_IN_LEASELINES
*MAX_PVT_REGIONS
*MAX_ROCK_TYPES
*MAXENTRIES_IN_KR_TABLE
*MAXENTRIES_IN_PVT_TABLE
*MAX_SPECIALVARS_IN_SRFOUTPUT
*MAXPERCENT_OF_FULLYIMPLICITBLOCKS
*SOLVER_DIMENSIONING
The *SOLVER_DIMENSIONING card must be read before the following can be read
*MAX_ORTHOGONALIZATIONS
*MAX_INTERBLOCK_CONNECTIONS
*MAX_OFFDIAGONAL_ENTRIES
*MAX_BLOCKENTRIES_IN_LU_FACTORS
*MAXSIZE_SOLVER_VECTOR
*MAX_SOLVER_DIAG_ENTRIES
*MAX_NONZERO_L_OR_U_ENTRIES
DEFINITIONS:
*DIM
DIM keyword. Required to dimension variables.
*ACTIVE_BLOCKS
The number of active blocks in a given run. Active blocks are all the active
blocks including dual porosity, hybrid or refined grids, excluding null (zero
porosity) blocks.
Default is to obtain this information directly by reading the data file
accounting for any NULL blocks.
Example: *DIM ACTIVE_BLOCKS 2000
User's Guide IMEX
*NULL_BLOCKS
The number of null blocks in a given run. The default would be no NULL
blocks. All blocks would be considered ACTIVE.
Example: *DIM NULL_BLOCKS 500
*MAX_FAULTS
Maximum number of faults
Example: *DIM MAX_FAULTS 15
Default value of MAX_FAULTS is set to 15.
*MAX_AQUIFERS
Maximum number of aquifers
Example: *DIM MAX_AQUIFERS 15
Default number of MAX_AQUIFERS is 15
*MAX_SECTORS
Maximum number of sectors
Example: *DIM MAX_SECTORS 50
Default number of MAX_SECTORS is 50
*MAX_LEASELINES
Maximum number of lease-lines
Example: *DIM MAX_LEASELINES 12
Default number of MAX_LEASELINES is 15
*MAXBLKS_IN_LEASELINES
Maximum number of blocks in a lease-line
Example: *DIM MAXBLKS_IN_LEASELINES 5000
Default number of MAXBLKS_IN_LEASELINES is 150
*MAX_PVT_REGIONS
Maximum number of PVT regions allowed.
Example: *DIM MAX_PVT_REGIONS 50
Default number of MAX_PVT_REGIONS is 15.
*MAX_ROCK_TYPES
Maximum number of rock types allowed.
Example: *DIM MAX_ROCK_TYPES 100
Default number of MAX_ROCK_TYPES is 30.
User's Guide IMEX
*MAXENTRIES_IN_KR_TABLE
Maximum number of entries in a rock-fluid table
Example: *DIM MAXENTRIES_IN_KR_TABLE 500
Default number of MAXENTRIES_IN_KR_TABLE is 101.
*MAXENTRIES_IN_PVT_TABLE
Maximum number of entries per PVT table
Example: *DIM MAXENTRIES_IN_PVT_TABLE 51.
Default number of MAXENTRIES_IN_PVT_TABLE is 51.
*MAX_SPECIALVARS_IN_SRFOUTPUT
Maximum number of *SPECIAL variables in SRF output allowed.
Example: *DIM MAX_SPECIALVARS_IN_SRFOUTPUT 500
Default number of MAX_SPECIALVARS_IN_SRFOUTPUT is 100.
*MAXPERCENT_OF_FULLYIMPLICITBLOCKS
Maximum percent of fully implicit blocks. Chosen between 10 and 100. If
fully implicit, choose 100.
Example: *DIM MAXPERCENT_OF_FULLYIMPLICITBLOCKS 100.
Default number of MAXPERCENT_OF_FULLYIMPLICITBLOCKS is 100.
*SOLVER_DIMENSIONING
Should solver arrays be custom dimensioned. Requires a 'ON' or 'OFF' (in
quotes) as a response.
Example: *DIM SOLVER_DIMENSIONING 'ON' to turn on custom
dimensioning
Example: *DIM SOLVER_DIMENSIONING 'OFF' to turn off custom
dimensioning.
By default, SOLVER_DIMENSIONING is set to 'OFF'.
*MAX_ORTHOGONALIZATIONS
Number of orthogonalizations in the linear solver. Reducing the number of
orthogonalizations will save storage, but will degrade performance.
Conditional input valid only when SOLVER_DIMENSIONING is set to 'ON'.
Example: *DIM MAX_ORTHOGONALIZATIONS 5
The default value of MAX_ORTHOGONALIZATIONS is set to 30.
User's Guide IMEX
*MAX_INTERBLOCK_CONNECTIONS
Maximum number of interblock connections. Interblock connections include
connections between grid blocks and those due to wells.
Example: *DIM MAX_INTERBLOCK_CONNECTIONS 5
The default value of MAX_INTERBLOCK_CONNECTIONS is estimated
based on the number of fundamental grid blocks, NI, NJ and NK, the number
of null blocks and the problem types to be run.
*MAX_OFFDIAGONAL_ENTRIES
Maximum number of entries in off-diagonal Jacobian matrix.
Example: *DIM MAX_OFFDIAGONAL_ENTRIES 5
The default value of MAX_OFFDIAGONAL_ENTRIES is estimated based
on the number of fundamental grid blocks, NI, NJ and NK, the number of
null blocks and the problem types to be run.
*MAX_BLOCKENTRIES_IN_LU_FACTORS
Maximum number of block entries in LU Factors.
Example: *DIM MAX_BLOCKENTRIES_IN_LU_FACTORS 5
The default value of MAX_BLOCKENTRIES_IN_LU_FACTORS is
estimated based on the number of fundamental grid blocks, NI, NJ and NK,
the number of null blocks and the problem types to be run.
*MAXSIZE_SOLVER_VECTOR
Maximum size of solver solution vector
Example: *DIM MAX_SOLVER_VECTOR 5
The default value of MAXSIZE_SOLVER_VECTOR is estimated based on
the number of fundamental grid blocks, NI, NJ and NK, the number of null
blocks and the problem types to be run.
*MAX_SOLVER_DIAG_ENTRIES
Maximum number of solver diagonal entries
Example: *DIM MAX_SOLVER_DIAG_ENTRIES 5
The default value of MAX_SOLVER_DIAG_ENTRIES is estimated based
on the number of fundamental grid blocks, NI, NJ and NK, the number of
null blocks and the problem types to be run.
User's Guide IMEX
*MAX_NONZERO_L_OR_U_ENTRIES
Maximum number of nonzero L or U entries
Example: *DIM MAX_NONZERO_L_OR_U_ENTRIES 5
The default value of MAX_NONZERO_L_OR_U_ENTRIES is estimated
based on the number of fundamental grid blocks, NI, NJ and NK, the number
of null blocks and the problem types to be run.
User's Guide IMEX
Background (Run Time Dimensioning)

During every run of the simulator, certain amount of computer system's memory resources
are required to store the variables used by the simulator. Depending on how a variable is
declared in the program, this storage could be allocated to it when the program is being
compiled or when the program is being run. For example, the following dimension statement
in the program code would cause the array A to be allocated memory at compile time
DIMENSION A(5)
Note that A has been dimensioned to 5 units in the above code. The memory to store array A
has been allocated at compile time and this way of dimensioning A could be called "compile
time dimensioning (CtD)". The advantage of CtD is its simplicity and merely declaring A
and its size is sufficient to allocate the required memory. The main disadvantage of CtD is
that we may not know in advance the maximum sizes of many dimensioned variables. For
example, what if in a particular application we need 6 values of A to be represented. With
CtD we have to change A(5) in the above DIMENSION statement to A(6), recompile the
entire program and run it.
IMEX 96.00 and older versions of IMEX have followed CtD. They have consistent sets of
maximum dimensioning values for all its dimensioned variables that result in different sized
executables when compiled.
The current version of IMEX uses a different approach to dimensioning variables. It uses a
pointer based approach to dimension its variables at run time of the simulation and not at
compile time. To illustrate run time dimensioning, let us reconsider the same DIMENSION
statement mentioned above. This statement is now written as,
pointer(p_A, A)
p_A = malloc((size of A in bytes)*5)
In the above lines, p_A is the pointer variable and is associated with A through the pointer
statement. In the following statement, the exact memory required to store A(5) is computed
in bytes and allocated using a special function by name malloc. The pointer variable, p_A,
holds the address of this allocated space and makes it available to store the values of A in the
program. This is the scheme of memory allocation followed by IMEX.
User's Guide IMEX
Advantages of Run Time Dimensioned IMEX

Run time dimensioning confers several advantages to IMEX. The principal advantages are,
1. IMEX comes in only one executable. There is no need for any custom sized
executables. This results in saving some disk space.
2. IMEX uses only the memory it requires for a given problem. With the same
hardware, this allows the solution of bigger (sized) problems than before.
3. Only hardware limitations would dictate the maximum size of the problems that
could be solved using IMEX. There would be no such limitations on the software
side.
User's Guide IMEX
Project Main Title (Optional)
*TITLE1
PURPOSE:
*TITLE1 is used for project identification.
FORMAT:
*TITLE1
string
DEFINITIONS:
string
Any alphanumeric character string enclosed in single quotes (40 characters
maximum)
DEFAULTS:
Optional keyword. Defaults to blanks.
CONDITIONS:
This keyword must appear in the Input/Output Control keyword group, at the start of the
input-data-file.
EXPLANATION:
This keyword identifies an alphanumeric character string used for project identification. It
will appear both in the output file and in the index-results-file.
Any character string with embedded blanks or commas must be enclosed with single quotes.
Examples:
*TITLE1
'DUAL POROSITY/DUAL PERMEABILITY RUN NO.1'
User's Guide IMEX
Project Second Title (Optional)
*TITLE2
PURPOSE:
*TITLE2 is used for project identification. It is used in addition to *TITLE1 to provide a
second line for project identification.
FORMAT:
*TITLE2
string
DEFINITIONS:
string
maximum)
DEFAULTS:
CONDITIONS:
input-data-file.
EXPLANATION:
Examples:
*TITLE1
*TITLE2
'Run by F.R. Sangiovanni, December 16, 1988. IBM '
User's Guide IMEX
Project Third Title (Optional)
*TITLE3
PURPOSE:
*TITLE3 is used for project identification. It is used in addition to *TITLE1 and *TITLE2 to
provide a third line for project identification.
FORMAT:
*TITLE3
string
DEFINITIONS:
string
maximum)
DEFAULTS:
CONDITIONS:
input-data-file.
EXPLANATION:
Examples:
*TITLE1
*TITLE2
'Run by F.R. Sangiovanni, December 16, 1988. IBM '
*TITLE3
'IMEX 4.0; 4200 grid blocks; variable thickness'
User's Guide IMEX
Case Identification (Optional)
*CASEID
PURPOSE:
*CASEID is used to identify specific case runs.
FORMAT:
*CASEID string
DEFINITIONS:
string
maximum)
DEFAULTS:
CONDITIONS:
input-data-file.
EXPLANATION:
This keyword identifies an alphanumeric character string used to name specific case runs. It
Examples:
*CASEID 'Case 21c'
User's Guide IMEX
Check Only (Optional)
*CHECKONLY
PURPOSE:
*CHECKONLY indicates that the well data is to be checked for format errors, but the
simulation will not be run. Also, please note that on the command line "-checkonly" enables
this option without requiring the *CHECKONLY Card. In addition using the command line
version of checkonly disables License Management for the "CHECKONLY" run. This
allows the user to run a model and check data simultaneously using a single IMEX License.
FORMAT:
*CHECKONLY
DEFAULTS:
Optional keyword. No check on the well data is performed unless this keyword is present.
CONDITIONS:
This keyword must appear in the Input/Output Control keyword group.
User's Guide IMEX
Trap UNIX Signal 2 Interrupt (Optional)
*INTERRUPT
PURPOSE:
*INTERRUPT indicates that when the user interrupts the simulation run by simultaneously
pressing the "CTRL-C" keys or by typing "kill -2" followed by the process identification
number (PID), one of the following user specified actions can take place:
1. Continue the simulation run.
2. Finish the current timestep, write a restart record and stop the simulation.
3. Flush buffers, close files and stop the simulation run immediately.
4. Write an impromptu restart and continue.
5. For PC only: Stop cold (immediately and unconditionally). Interactively, this
action has to be requested with a Control-Break.
Use of this keyword (1-3) prevents corruption of the irf and mrf files when aborting a
simulation run and optionally writes a restart record before stopping. It (4) also allows
writing of an impromptu restart record.
FORMAT:
*INTERRUPT
(*RESTART-STOP)
(*WRST-CONT)
(*STOP)
(*STOP-COLD)
(*INTERACTIVE)
(Default)
DEFINITIONS:
*RESTART-STOP
This subkeyword specifies that the current timestep be completed, all
specified output data be written to the output and SR2 files and a restart
record also be written to the SR2 files.
*WRST-CONT
This subkeyword specifies that an impromptu restart be written, i.e. all
specified output data be written to the output and SR2 files and a restart
record also be written to the SR2 files. The run then continues.
*STOP
This subkeyword specifies that the simulation run be terminated
immediately. The current timestep is not to be completed. However, prior to
termination, all buffers will be flushed and files closed to prevent file
corruption.
User's Guide IMEX
*STOP-COLD
****** for the PC only ******
This subkeyword specifies that the simulation run be stopped cold,
immediately and unconditionally. The action is that of a raw Control-C
without any trapping. This is the equivalent of Control-Break in interactive
mode. This option is needed as the action on any of the other options is not
instantaneous on the PC, hence may take too long.
*INTERACTIVE
This subkeyword specifies that the user will be prompted to enter the action
to be taken when a Unix signal 2 is detected.
Signal file:
If the user places a file in the data directory with the extension .signal and the same name
(without the extension) as the data set, the simulator running that data set will act as if it has
received an interrupt. If the signal file is empty, the action the simulator will take will default
to the last defined action. If the action was not previously defined, the simulator will treat the
existence of the file as a *RESTART-STOP signal (the default).
If the file contains a single integer in the range 1 to 4, then the simulator will
Integer
Action
Continue Simulation Run
Write Restart and Quit
Tidy up and Quit Now
Write impromptu Restart
DEFAULTS:
Optional keyword. If it is not specified, then the user is prompted interactively which of the
above four actions is to be taken. If *INTERRUPT is specified without an action, the default
action is *RESTART-STOP.
CONDITIONS:
When a Control-C is encountered by the simulator during a run with an action of either
*RESTART-STOP or *WRST-CONT, all required restart information is written out. In
addition, grid, well and sector output will be written to both the output and SR2 files.
When this additional output information is written out, user defined output frequency controls
remain unchanged and will behave as if the Control-C was not encountered.
User's Guide IMEX
ASCII Form of SR2 (Optional)
*SR2ASCII
PURPOSE:
*SR2ASCII indicates that an ASCII copy of the main-results-file (mrf) will be created in
addition to the binary file. The extension for this file is .asc
FORMAT:
*SR2ASCII
DEFAULTS:
Optional keyword. No Defaults
CONDITIONS:
User's Guide IMEX
Precision of SR2 (Optional)
*SR2PREC
PURPOSE
*SR2PREC flags the precision with which floating point data is written to the binary SR2
file. Use *SINGLE for 4 bytes and *DOUBLE for 8 bytes. *SINGLE is not available with
*XDR *OFF.
Use of *SINGLE will cause the result of a restart run to vary slightly from the same timesteps
of the first run, due to the fact that the lower order digits of the starting values have been
truncated.
FORMAT:
*SR2PREC ( *SINGLE | *DOUBLE )
DEFAULTS:
If *SR2PREC is absent, then *SR2PREC *DOUBLE is assumed.
CONDITIONS:
User's Guide IMEX
Controlling Data File Listing Using NOLISTLIM (Optional)

*NOLISTLIM
PURPOSE:
*NOLISTLIM specifies that the default 20 lines of keyword echoing under the *LIST
keyword be extended so that all values of the keyword are echoed regardless of the number of
lines required to do so. This can create large files if used, but allows the user to view all of
the data.
FORMAT:
*NOLISTLIM
DEFAULTS:
Optional keywords. Defaults to a list limit of 20 lines per keyword
CONDITIONS:
*NOLISTLIM must appear on a line by itself in the Input/Output Control Section.
EXPLANATION:
simulation run. When keywords containing large amounts of data are listed under *LIST,
only the first 20 lines of data are echoed. In order to echo all lines of data (this can result in
large files) the *NOLISTLIM keyword in the Input/Output Control section is required.
See *LIST and *NOLIST in the Keyword Data Entry System section.
User's Guide IMEX
Input Data Units (Optional)
*INUNIT
PURPOSE:
*INUNIT specifies the input data units.
FORMAT:
*INUNIT
(*SI)
(*FIELD)
(*LAB)
(*MODSI)
DEFINITIONS:
*SI
This option specifies SI units for input data.
*FIELD
This option specifies FIELD units for input data.
*LAB
This option specifies LAB units for input data.
Please note: Do not use the analytical aquifer model when using *LAB
units. The use of minutes rather than days as a time unit makes the
aquifer model invalid.
*MODSI
This option specifies modified SI units for input data.
DEFAULTS:
Optional keyword. Default is *SI.
CONDITIONS:
input-data-file.
EXPLANATION:
Input data may be entered using one of the four options shown in the following Units Table.
Examples:
*INUNIT *FIELD
User's Guide IMEX
UNITS TABLE
*SI
*FIELD
*LAB
*MODSI
Time
days
days
mins
days
Liquid Volumes
m3
bbl
cm3
m3
Gas Volume
m3
ft3
cm3
m3
Pressure
kPa
psi
kPa
kg/cm2
Absolute Permeability
md
md
md
md
Viscosity
mPa-s
cp
mPa-s
mPa-s
Mass Density
kg/m3
lbm/ft3
g/cm3
kg/m3
Length, Distance
ft
cm
Solution Gas/Oil Ratio
m3/m3
scf/STB
cm3/cm3
m3/m3
Polymer Concentration
kg/m3
lb/STB
g/cm3
kg/m3
Porosity
fraction
fraction
fraction
fraction
Saturation
fraction
fraction
fraction
fraction
Temperature
deg C
deg F
deg C
deg C
Volume Factors
m3/m3
RB/STB
cm3/cm3
m3/m3
Gas Expansion Factor
m3/m3
scf/RB
m3/m3
m3/m3
m3/m3
RB/scf
cm3/cm3
m3/m3
dynes/cm
dynes/cm
dynes/cm
dynes/cm
Oil, Water Formation
Gas Formation
Volume Factor
Surface Tension
User's Guide IMEX
Here is a list of conversion units used by IMEX. To convert user input units to SI units
multiply by the numbers given below the unit in the table. To convert from SI, multiply by
the inverse of the numbers given in the table.
Liquid Volume
User Unit
Conversion Factor
Gas Volume
User Unit
Conversion Factor
Pressure
User Unit
Conversion Factor
Viscosity
User Unit
Conversion Factor
Permeability
User Unit
Conversion Factor
Density
User Unit
Conversion Factor
Length
User Unit
Conversion Factor
Mass
User Unit
Conversion Factor
Time
User Unit
Conversion Factor
Solution Gas/Oil Ratio
User Unit
Conversion Factor
Formation Vol. Factors (oil,
water)
User Unit
Conversion Factor
Gas Expansion Factor
User Unit
Conversion Factor
m3
1.0
bbl
0.1589873
cm3
1.0E-06
m3
1.0
m3
1.0
ft3
0.02831685
cm3
1.0E-06
m3
1.0
kPa
1.0
psi
6.894757
kPa
1.0
kg/cm2
98.0665
mPa-s
1.0
cp
1.0
mPa-s
1.0
mPa-s
1.0
md
1.0
md
1.0
md
1.0
md
1.0
kg/m3
1.0
lbm/ft3
16.01846
g/cm3
1000.0
kg/m3
1.0
m
1.0
ft
0.3048
cm
.01
m
1.0
kg
1.0
lbm
0.4535924
g
.001
kg
1.0
days
1.0
days
1.0
minutes
6.944E-04
days
1.0
m3/m3
1.0
scf/STB
0.1780444
cm3/cm3
1.0
m3/m3
1.0
m3/m3
1.0
RB/STB
1.0
cm3/cm3
1.0
m3/m3
1.0
m3/m3
1.0
scf/RB
0.178044
cm3/cm3
1.0
m3/m3
1.0
User's Guide IMEX
Output Data Units (Optional)
*OUTUNIT
PURPOSE:
*OUTUNIT specifies the output data units.
FORMAT:
*OUTUNIT
(*SI)
(*FIELD)
(*LAB)
(*MODSI)
DEFINITIONS:
*SI
This option specifies SI units for output data.
*FIELD
This option specifies FIELD units for output data.
*LAB
This option specifies LAB units for output data.
*MODSI
This option specifies modified SI units for input data.
DEFAULTS:
Optional keyword. Default is *INUNIT.
CONDITIONS:
This keyword must appear in the Input/Output Control keyword group, at the start of the data
file.
EXPLANATION:
Output data may be displayed using one of the four options shown in the Units Table (see
*INUNIT)
Examples:
*OUTUNIT *FIELD
User's Guide IMEX
Data Range Checking (Optional)
*RANGECHECK
PURPOSE:
*RANGECHECK controls the data range check feature.
FORMAT:
*RANGECHECK
(*ON)
(*OFF)
DEFINITIONS:
*ON
Turn on the range check feature.
*OFF
Turn off the range check feature.
DEFAULTS:
Optional keyword. Default is *ON.
CONDITIONS:
input-data-file.
EXPLANATION:
Most input data is examined to determine if it is within an expected range of numbers.
Specifying *RANGECHECK *OFF will disable the noncritical data range checking.
*RANGECHECK *OFF also will suppress the printing of all "warning" messages. Error
messages will always be printed.
Examples:
*RANGECHECK *OFF
User's Guide IMEX
Maximum Number of Error Messages (Optional)
*MAXERROR
PURPOSE:
*MAXERROR specifies the maximum number of error messages before the simulation
terminates.
FORMAT:
*MAXERROR num
DEFINITIONS:
num
An integer indicating the maximum number of error messages.
DEFAULTS:
Optional keyword. Default is 20.
CONDITIONS:
input-data-file. The value for num must be between 1 and 100.
EXPLANATION:
During data input, when a syntax or range error occurs, the simulator will print an error
message, then attempt to continue scanning the input data.
Simulation is stopped if there are errors in the initialization data. Thus, initialization is not
done and the well data is not read. If initialization is done but there are errors in the well
data, then simulation is stopped at this point. In both cases, the run is terminated before the
*MAXERROR value is reached.
Some syntactical errors may result in the simulator getting so confused that it would print out
dozens of meaningless error messages. Thus, if the number of error messages exceeds the
maximum number of errors, the simulator immediately stops.
Examples:
** Stop simulation if 50 errors encountered.
*MAXERROR 50
User's Guide IMEX
Restart Timestep (Optional)
*RESTART
PURPOSE:
*RESTART indicates that this is a restart run.
FORMAT:
*RESTART num
DEFINITIONS:
num
An integer to specify the timestep number after which to restart the
simulation.
DEFAULTS:
Optional keyword. If this keyword is not present in the input-data-file, then no restart records
are read and the simulation begins at time zero.
If *RESTART is present, but no timestep number, the default is to restart at the last timestep
in the input index-results-file.
CONDITIONS:
EXPLANATION:
Restart records store a "snap-shot" of reservoir conditions at a particular time. Using a restart
record you can restart a simulation from some mid point in a run. This allows you to try
different well production strategies, produce more detailed output, or make other changes
without the expense of repeating the entire simulation run.
Examples:
** Restart at the 10th timestep.
*RESTART 10
** Restart at the last timestep in the restart file.
*RESTART
User's Guide IMEX
Restart Record Writing Frequency (Optional)
*WRST, *REWIND
PURPOSE:
*WRST controls the frequency of writing restart records to the index-results-file and to either
the main-results-file or the rewindable-results-file.
*REWIND controls the frequency of rewinding the rewindable-results-file.
FORMAT:
*WRST
*REWIND
(*FREQ-TIME
(*FREQ
(*TIME)
(*TNEXT)
(num)
freq)
freq)
DEFINITIONS:
*FREQ-TIME
Subkeyword indicating that restart records are to be written at both the
timestep frequency specified by 'freq' and the subsequent recurrent *TIME or
*DATE keywords.
*FREQ
Subkeyword indicating that restart records are to be written at only the
timestep frequency specified by 'freq'.
freq
An integer to specify the frequency (timesteps). Data is written at the given
frequency. If 'freq' is missing after *FREQ or *FREQ-TIME then *TIME is
the default action. If both *FREQ-TIME and *FREQ are missing and 'freq'
appears by itself after *WRST then *FREQ-TIME is the default.
*TIME
Write data to the two ( or three) results files every time specified by
subsequent recurrent *TIME or *DATE keywords.
*TNEXT
Write data to the two (or three) results files on the next time specified by the
next recurrent *TIME or *DATE keywords. No further output is written
until the next *WRST card is read.
num
Frequency of rewinding the rewindable-restart-file. num is the maximum
number of restart records allowed to accumulate in the rewindable-restart-file
before it is rewound. num must be greater than 0. If num is 1, then only the
last restart is available.
User's Guide IMEX
DEFAULTS:
Optional keywords. If the *WRST keyword is not present in the input-data-file, the default is
no restart record written; if *WRST is present, the default is to write a restart at every time
specified by subsequent recurrent *TIME or *DATE keywords in the input-data-file.
If *REWIND is not present the rewindable-restart-file is not used. If 'num' is absent, then
num = 1 is assumed.
CONDITIONS:
This keyword may appear in the Input/Output Control keyword group and may also occur as
part of recurrent data. Thus, the frequency of records in the restart file may be changed
during the simulation.
EXPLANATION:
If *REWIND (num) is used, restart records are written not to the main-results-file but to the
rewindable-results-file. This file is rewound after 'num' restart records are written into it.
Thus at the end of the run it can contain up to 'num' restart records from which the simulation
can be restarted.
Examples:
** Write restart record every 10 timesteps.
*WRST 10
** Write restart at the every time change.
*WRST
*REWIND 5 ** rewind the rewindable-results-file
** after five restarts have been written.
** Write restart on date Jan 01 2001
*WRST *TNEXT
*DATE 2001 01 01
User's Guide IMEX
Output Printing Frequency (Optional)
*WPRN
PURPOSE:
*WPRN controls the frequency of writing data to the output file.
FORMAT:
*WPRN
(*WELL)
(*GRID)
(*SECTOR)
(*ITER)
(freq)
(*TIME)
(*TNEXT)
(freq)
(*TIME)
(*TNEXT)
(freq)
(*TIME)
(*TNEXT)
(*ALL)
(*MATRIX)
(*NEWTON)
(*BRIEF)
(*NONE)
DEFINITIONS:
*WELL freq
Write well results to the output file every freq timesteps, where freq is an
integer. If freq is zero, no well results are written.
*WELL *TIME
Write well results to the output file at every time specified by subsequent
recurrent *TIME or *DATE keywords in the input-data-file.
*WELL *TNEXT
Write well results to the output file on the next time specified by the next
recurrent *TIME or *DATE keywords in the input-data-file. No further
output is written until the next *WPRN *WELL card is read.
*GRID freq
Write grid results to the output file every freq timesteps, where freq is an
integer. If freq is zero, no grid results are written.
*GRID *TIME
Write grid results to the output file at every time specified by subsequent
recurrent *TIME or *DATE keywords in the input-data-file.
User's Guide IMEX
*GRID *TNEXT
Write grid results to the output file on the next time specified by the next
recurrent *TIME or *DATE keywords in the input-data-file. No further
output is written until the next *WPRN *GRID card is read.
*SECTOR freq
Write sector, flux sector and lease line results to the output file every freq
timesteps, where freq is an integer. If freq is zero, no sector and lease plane
results are written.
*SECTOR *TIME
Write sector, flux sector and lease line results to the output file at every time
specified by subsequent recurrent *TIME or *DATE keywords in the inputdata-file.
*SECTOR*TNEXT
Write sector, flux sector and lease line results to the output file on the next
time specified by the next recurrent *TIME or *DATE keywords in the
input-data-file. No further output is written until the next *WPRN
*SECTOR card is read.
*ITER *ALL
Provide full details of matrix solution and Newton iterations at every
timestep.
*ITER *MATRIX
Provide details of matrix solution only at every timestep.
*ITER *NEWTON
Provide details of Newton iterations at every timestep.
*ITER *BRIEF
Provide a summary of timestep convergence behavior.
*ITER *NONE
Provide no information on timestep behavior.
DEFAULTS:
Well results are written every timestep. Grid results are written every *TIME or *DATE
card. Sector and lease plane results are written every *TIME or *DATE card. Timestep
summary output defaults to *WPRN *ITER *BRIEF
CONDITIONS:
part of recurrent data. Thus, the amount of detail in the output file may be changed during
the simulation.
User's Guide IMEX
P/Z Sector Output:

In gas-water, black-oil and condensate models which specify a PVT region Temperature
(*TRES) for each PVT region defined, average P/Z is output on a sector and field basis.
Average P/Z is available weighted in each Sector by both pore volume and Hydrocarbon pore
volume.
EXPLANATION:
Examples:
** Write well results at every time change.
*WPRN *WELL *TIME
** Write grid results every 10 timesteps.
*WPRN *GRID 10
** Write sector and lease plane results every time
** change.
*WPRN *SECTOR *TIME
** Write details of Newton iterations
** at every timestep.
*WPRN *ITER *NEWTON
** Write well results only on the next
** time change or date card.
*WPRN WELL *TNEXT
User's Guide IMEX
Items in Output Print File (Optional)
*OUTPRN
PURPOSE:
*OUTPRN identifies what information is written to the output file.
FORMAT:
*OUTPRN
(*WELL)
(*GRID)
(*TABLES)
(*RES)
(*WELL-SECTOR)
(*FLUX-SECTOR)
(*ALL)
(*RESERVOIR)
(*LAYER)
(*BRIEF)
(*ALL)
(*NONE)
(*EXCEPT) (grid_list)
(grid_list)
(*ALL)
(*NONE)
(*ALL)
(*NONE)
(*EXCEPT) (res_list)
(res_list)
(*SORT-WELL-NAME)
(*SORT-WELL-NUM)
(*NONE)
(*ALL) (sc-rc selector) (ijk-con selector)
(*INFLUX)
(*INTERSEC)
(*NONE)
DEFINITIONS:
*WELL
This subkeyword specifies that well results will be written to the output file.
*GRID
This subkeyword specifies that grid results will be written to the output file.
*TABLES
This subkeyword specifies that input property tables will be printed at the
start of the simulation run.
*RES
This subkeyword specifies that input reservoir properties will be printed at
the start of the simulation run. Original volumes in place will always be
printed after input reservoir properties unless *RES *NONE is specified.
User's Guide IMEX
*WELL-SECTOR
This keyword specifies that well in sector information will be printed out.
Well in sector information consists of two tables. The first is ordered by
sectors (sector table) and lists all active wells within a sector (an active well
has at least one perforation open). The second table is ordered by well (well
table) and lists all sectors associated with each active perforation of every well.
Wells which have all perforations closed or have not had any perforations
defined will not be listed in either table. Individual perforations that are no
longer active will not be listed in the well table. Perforations which have had
their production/injection zeroed due to use of the *XFLOW-MODEL
*ZERO-FLOW keyword will not be listed in the well table for as long as
their production/injection is zeroed. Active perforations of wells that are
shut in will be listed in the well table, but shut in wells are not listed in
the sector table.
The table information is printed out every timestep to a separate file with
extension sct, the file will be created if when scanning the data IMEX finds
the *OUTPRN *WELL-SECTOR keywords. The amount of printout can be
controlled by using the *NONE subkeyword in recurrent data.
For example, the user can initially define *OUTPRN *WELL-SECTOR
*NONE and have no printout until within the recurrent data a *OUTPRN
*WELL-SECTOR keyword is found with subkeyword *SORT-WELLNAME or *SORT-WELL-NUM. Output is again turned off at the next
time/date the *OUTPRN *WELL-SECTOR *NONE keyword is found.
If *OUTPRN *WELL-SECTOR is not found, no tables are written. If
*OUTPRN *WELL-SECTOR is found without subkeywords *SORT-WELLNAME or *SORT-WELL-NUM, *SORT-WELL-NAME is the default.
This output option will normally be used only on selected runs to help debug
well placement within sectors.
*FLUX-SECTOR
This subkeyword specifies that flux sector summary will be printed out. The
output frequency is controlled by *WPRN *SECTOR
*ALL
Write all possible variables to the output file. For the *WELL option this
means all of *BRIEF, *RESERVOIR and *LAYER. For *FLUX-SECTOR
option, it means both influx and inter-sector flux will be printed out.
*RESERVOIR
Write a summary of well variables at reservoir conditions to the output file.
If available this will also cause voidage replacement information to be
printed out at the field and group level
User's Guide IMEX
*LAYER
Write a summary of layer variables at surface conditions to the output file.
*BRIEF
Write a summary of well variables to the output file at surface conditions
only. No layer information is printed.
*NONE
No variables will be written to the output file.
*EXCEPT
This subkeyword identifies an exception list of variables to be written to the
output file (valid for *GRID only).
*SORT-WELL-NAME
Sorts wells in both the sector and well tables in the sct file by well name
(*OUTPRN *WELL-SECTOR *SORT-WELL-NAME)
*SORT-WELL-NUM
Sorts wells in both the sector and well tables in the sct file by well number
(*OUTPRN *WELL-SECTOR *SORT-WELL-NUM)
grid_list
A series of character strings to identify which variables to write (or not to
write when used with *EXCEPT) to the output file (valid for *GRID only).
The grid_list for *GRID is:
*SO
*SG
*SW
*SS
*PCONC
*PADSORP
*PRES
*DATUMPRES
*OILPOT
*BPP
*SSPRES
*WINFLUX
*IMEXMAP
*KRO
*KRW
*KRG
*KRS
*VISO
*VISW
*VISG
Oil saturation
Gas saturation
Water saturation
Solvent saturation
Polymer concentration
Polymer adsorption
Pressure
Datum pressure
Oil potential
Bubble point pressure
Solvent saturation pressure
Aquifer water influx
Implicit / explicit block map
Oil relative permeability
Water relative permeability
Gas relative permeability
Solvent relative permeability
Oil viscosity
Water viscosity
Gas viscosity
User's Guide IMEX
*VISS
*MASDENO
*MASDENW
*MASDENG
*MASDENS
*TRMOI
*TRMOJ
*TRMOK
*SWCON
*SWCRIT
*SOIRW
*SORW
*SGCON
*SGCRIT
*SLCON
*SOIRG
*SORG
*APIGRAV
*POROS
*KRSETN
*RFG
*RFO
*RFW
*RFS
*PCOW
*PCOG
*SBDZ
*EG
*BO
*RS
*RV
*DPP
*SEAWF
Solvent viscosity
Oil density
Water density
Gas density
Solvent density
Oil phase transmissibility in
the I-, J-, and K-direction
respectively
Connate water saturation Swcon
Critical water saturation Swcrit
Irreducible oil saturation Soirw (oil-water table)
Residual oil saturation after water flood (Sorw)
Connate gas saturation Sgcon
Critical gas saturation Sgcrit
Connate liquid saturation Slcon (gas-liquid table)
Irreducible oil saturation Soirg (gas-liquid table)
Residual oil saturation after gas flood (Sorg)
Light oil volume fraction (STC) (API tracking option)
Porosity at current pressure (includes compressibility/
compaction effects)
Relative Permeability Set Number Map
Gas resistance factor for non-Darcy flow (1+ Forchheimer
number)
Oil resistance factor for non-Darcy flow (1+ Forchheimer
number)
Water resistance factor for non-Darcy flow (1+ Forchheimer
number)
Solvent resistance factor for non-Darcy flow (1+ Forchheimer
number)
Water-oil capillary pressure (water-gas capillary pressure
when using *MODEL *GASWATER)
Oil-gas capillary pressure
Subsidence
Gas expansion factor
Oil formation volume factor
Solution gas ratio
Condensate oil content
Dew point pressure
Seawater Volume Fraction (Seawater Injection Model Only)
The following keywords are not output automatically when *GRID *ALL or *GRID
*EXCEPT is used. The keywords must appear in a grid_list.
*KRWSCL
*KROWSCL
*KRGSCL
*KROGSCL
User's Guide IMEX
Krw at irreducible oil (Soirw)

Krow at connate water (Swcon)
Krg at connate liquid (Slcon)
Krog at connate gas (Sgcon)
*PCWSCL
*PCGSCL
*SORMAX
*SORH
Pcow, Pcwg or Pclg (or J Function equivalent) at connate

water (Swcon)
Pcog (or J Function equivalent) at connate liquid (Slcon)
Maximum residual oil saturation of the imbibition process
(oil-water table, Krow hysteresis model or trapped oil
hysteresis only)
Current residual oil saturation of the imbibition process (oilwater table, Krow hysteresis model or trapped oil hysteresis
only)
res_list
A series of character strings to identify which variables to write (or not to
write when used with *EXCEPT) to the output file (valid for *RES only).
The res_list for *RES is:
*GRNUM
*SECTOR
*DI
*DJ
*DK
*NTG
*PAYDEPTH
*POR
*PV
*HCPV
*PERMI
*PERMJ
*PERMK
*TRANSI
*TRANSJ
*TRANSK
*DIFRAC
*DJFRAC
*DKFRAC
*TRSMIS
*RS
*SWCON
*SWCRIT
*SOIRW
*SORW
*SORMAX
*SORH
Grid block numbers

Sector numbers
I-direction grid block length
J-direction grid block length
K-direction grid block length
Net-to-gross ratio
Block center depth
Gross porosity
Pore volume
Hydrocarbon pore volume
Permeability
in the I-, J-, and K-direction
respectively
Transmissibility multipliers
in the I-, J-, and K-direction
respectively
Fracture spacing for dual porosity
or dual permeability system in the
I-, J-, and K-direction respectively
Transmissibility
Solution gas Rs
hysteresis only)
Current residual oil saturation of the imbibition process (oilwater table, Krow hysteresis model or trapped oil hysteresis
only)
User's Guide IMEX
*SGCON
*SGCRIT
*SLCON
*SOIRG
*SORG
*PVTSETN
*DPO
*DPW
*DPG
*NDBETCR
*WSRFTN
*GSRFTN
*CRSETN
*KRWSCL
*KROWSCL
*KRGSCL
*KROGSCL
*PCWSCL
*PCGSCL
*PCWSHF
*JFWSHF
*FRWIDTHCOR
*PTHRESHI
*PTHRESHJ
*PTHRESHK
sc-rc selector
*SC
*RC

PVT set number map
Offsets for oil, water, and
gas phase equilibrium with DEPTH_ave
respectively
Non-Darcy correction factor
Water-oil (water-gas) surface tension term
Gas-oil surface tension term
Compaction rock type (set number) map
water (Swcon)
Pcow (or Pcgw or Pclg) shift
Water J function shift
Fracture width correction for velocity dependent polymerwater viscosity table
Pressure gradient threshold in the I direction
Pressure gradient threshold in the J direction
Pressure gradient threshold in the K direction
Flux sector summary contains values in surface condition

Flux sector summary contains values in reservoir condition
If neither present, both *SC and *RC are selected.

ijk-con selector
*SUM
*ICON
*JCON
*KCON
Flux sector summary contains summation of connections in all three

directions
Flux sector summary contains flow across connections in I direction
Flux sector summary contains flow across connections in J direction
Flux sector summary contains flow across connections in K direction
If none of above selectors present, *SUM is selected.
User's Guide IMEX
DEFAULTS:
Optional keyword. If it is not present in the input-data-file, the defaults are:
*OUTPRN
*OUTPRN
*WELL
*GRID
*OUTPRN
*OUTPRN
*TABLES
*RES
*OUTPRN
*OUTPRN
*WELL-SECTOR
*FLUX-SECTOR
*BRIEF
*SO *SG *SW *SS *PCONC *PADSORP *PRES
*OILPOT *BPP *SSPRES *WINFLUX *IMEXMAP
*ALL
*SECT *DI *DJ *DK *NTG *PAYDEPTH *POR
*PERMI *PERMJ *PERMK *TRANSI *TRANSJ
*TRANSK
*NONE
*NONE
CONDITIONS:
The keywords *OUTPRN *WELL, *OUTPRN *GRID *OUTPRN *WELL-SECTOR and
*OUTPRN *FLUX-SECTOR may appear in the Input/Output Control keyword group and may
also occur as part of recurrent data. Thus, the amount of detail in the output files may be
changed during the simulation.
EXPLANATION:
An example of *OUTPRN *GRID, when using the list option is:
*OUTPRN *GRID *SO *SG *SW *PRES *OILPOT *BPP *IMEXMAP
*OUTPRN *WELL *BRIEF will only print the surface information.
*OUTPRN *WELL *LAYER will print both the surface and the layer
information.
*OUTPRN *WELL *RESERVOIR will print both the surface and reservoir
information.
*OUTPRN *WELL *ALL will print the surface, the reservoir and the
layer information.
When *OUTPRN *WELL *RESERVOIR or *OUTPRN *WELL *ALL is specified, well

mobility-weighted grid block pressure, grid block datum pressure, and drawdown are printed
to the output file. These mobility-weighted quantities are weighted averages of values for the
individual completions, with the weighting factors proportional to the total mobilities of the
phases in the completion grid blocks.
User's Guide IMEX
Simulation Results File Writing Frequency (Optional)
*WSRF
PURPOSE:
*WSRF controls the writing of well and/or grid information to the index-results-file and the
main-results-file ( the SR2 file system).
FORMAT:
*WSRF
(*WELL)
(*GRID)
(*SECTOR)
(freq)
(*TIME)
(*TNEXT)
(freq)
(*TIME)
(*TNEXT)
(freq)
(*TIME)
(*TNEXT)
DEFINITIONS:
*WELL freq
Write well results to the SR2 file system every freq timesteps, where freq is an
integer. If freq is zero the well results are not written to the SR2 file system.
*WELL *TIME
Write well results to the SR2 file system at every time specified by
subsequent recurrent *TIME or *DATE keywords in the input-data-file.
*WELL *TNEXT
Write well results to the SR2 file system on the next time specified by the
next recurrent *TIME or *DATE keywords in the input-data-file. No further
output is written until the next *WSRF *WELL card is read.
*GRID freq
Write grid results to the SR2 file system every freq timesteps, where freq is an
integer. If freq is zero the grid results are not written to the SR2 file system.
*GRID *TIME
Write grid results to the SR2 file system at every time specified by
subsequent recurrent *TIME or *DATE keywords in the input file.
*GRID *TNEXT
Write grid results to the SR2 file system on the next time specified by the
next recurrent *TIME or *DATE keywords in the input-data-file. No further
output is written until the next *WSRF *GRID card is read.
User's Guide IMEX
*SECTOR freq
Write sector, flux sector and lease line results to the SR2 file system every
freq timesteps, where freq is an integer. If freq is zero, sector results are
written only at time 0.
*SECTOR *TIME
Write sector, flux sector and lease line results to the SR2 file system at every
time specified by subsequent recurrent *TIME or *DATE keywords in the input
file.
*SECTOR *TNEXT
Write sector, flux sector and lease line results to the SR2 file system on the
next time specified by the next recurrent *TIME or *DATE keywords in the
input-data-file. No further output is written until the next *WSRF *SECTOR
card is read.
DEFAULTS:
*WSRF
*WSRF
*WSRF
*WELL
*GRID
*SECTOR
1
*TIME
*TIME
CONDITIONS:
part of recurrent data. Thus, the amount of detail and frequency of records in the SR2 file
system may be changed during the simulation.
EXPLANATION:
*WSRF controls the writing of well, grid, sector and lease plane information to the SR2 file
system. The SR2 files system consists of up two three files: the index-results-file, the mainresults-file and the rewindable-results-file. These files are required for restart runs and by the
graphics post-processor "RESULTS".
The SR2 files also contain the grid information related to sectors defined on the *SECTOR or
*SECTORARRAY cards. Thus it is possible to use "RESULTS" to probe which sectors a
block is in. In addition it is possible to use "RESULTS" to only view blocks belonging to a
specified sector. The blocks not in the specified sector would be invisible.
P/Z Sector Output:
In gas-water, black-oil and condensate models which specify a PVT region Temperature
(*TRES) for each PVT region defined, average P/Z is output on a sector and field basis.
Average P/Z is available weighted in each Sector by both pore volume and Hydrocarbon pore
volume.
Examples:
*WSRF *WELL 1
*WSRF *GRID *TIME
*WSRF *SECTOR *TNEXT
User's Guide IMEX
Items in Simulation Results File (Optional)
*XDR, *OUTSRF
PURPOSE:
*OUTSRF identifies what information is written to the index-results-file and the mainresults-file (the SR2 file system).
FORMAT:
FORMAT:
*XDR (*ON | *OFF)
*OUTSRF
*OUTSRF
(*WELL)
(*DOWNHOLE)
(*BLOCKP)
(*LAYER)
(*GRID)
(*RES)
(*SPECIAL
(*FLUX-SECTOR)
(*ON | *OFF)
(*ALL) (*DOWNHOLE)
(*NONE)
(*ALL)
(*NONE)
(*EXCEPT) (grid_list)
(grid_list)
(*ALL | *NONE)
i j k grid_list)
(*ALL | *NONE) (sc-rc selector) (ijk-con
selector)
DEFINITIONS:
*XDR
The binary (data) file may be written in external data representation (XDR)
format as well as the binary format native to your platform. Use of XDR
allows the SR2 binary file(s) to be written on one platform and read on
another. For example, the SR2 files can be generated on a UNIX workstation
server and then accessed with RESULTS or the Report Writer on a PC. If
the SR2 is in XDR format, then the keyword "XDR" will appear near the top
of the index file (irf).
*GRID
This subkeyword specifies that grid results will be written to the SR2 file
system.
*WELL
This subkeyword specifies that well results will be written to the SR2 file
system.
*RES
This subkeyword specifies that input reservoir properties will be printed at the
start of the simulation run.
User's Guide IMEX
*DOWNHOLE
This subkeyword specifies that well performance values will be written to the
SR2 file system at reservoir conditions in addition to surface conditions. If
available voidage replacement ratios will be recorded to the SR2 file system
for display using Results Graph.
*LAYER
This subkeyword specifies that well performance values will be written for
all layers of wells specified to the SR2 file system. The default is *NONE
(see DEFAULTS below). Using *ALL specifies that layer information will
be written for all wells to the SR2 file system.
Use of this option, especially with *ALL, can increase the size of the SR2
file substantially.
*BLOCKP
The pressure of the block containing the bottomhole layer as defined in the
description for keyword *PERF. If *LAYER is enabled, then block pressure
of each layer is written as well.
*SPECIAL
This subkeyword specifies a grid block location and a variable. The grid
variable value at the given location is written to the SR2 file system.
*FLUX-SECTOR
This subkeyword specifies that flow values of flux sectors will be written
out. The output frequency is controlled by *WSRF *SECTOR. The output
can be found under the LEASE origin in RESULTS GRAPH.
If *FLUX-SECTOR *ALL is specified, both influx and inter-sector flux will
be written out. *FLUX-SECTOR *NONE indicates nothing will be written to
the SR2 file except if there is *LEASE keyword defined. Please see
CONDITIONS below for details.
*ALL
Write all possible variables to the SR2 file system.
*NONE
No variables will be written to the SR2 file system.
*EXCEPT
This subkeyword identifies an exception list of variables to be written to the
SR2 file system (valid for *WELL and *GRID only).
ijk
Grid block location of *SPECIAL variable.
User's Guide IMEX
grid_list
A series of keywords to identify which grid variables to write (or not to write
when used with *EXCEPT) to the SR2 file system (valid for *GRID only).
For *OUTSRF *SPECIAL the grid_list can contain only one variable.
Valid keywords in the grid_list for *GRID include:
*SO
*SG
*SW
*SS
*PCONC
*PADSORP
*PRES
*DATUMPRES
*OILPOT
*BPP
*SSPRES
*WINFLUX
*KRO
*KRW
*KRG
*KRS
*VISO
*VISW
*VISG
*VISS
*MASDENO
*MASDENW
*MASDENG
*MASDENS
*SWCON
*SWCRIT
*SOIRW
*SORW
*SGCON
*SGCRIT
*SLCON
*SOIRG
*SORG
*APIGRAV
*POROS
*KRSETN
*PERM
(*PERMEABILITY)
User's Guide IMEX
Oil saturation
Gas saturation
Water saturation
Solvent saturation
Polymer adsorption
Pressure
Datum pressure
Oil potential
Bubble point pressure
Solvent saturation pressure
Aquifer water influx
Oil relative permeability
Water relative permeability
Gas relative permeability
Solvent relative permeability
Viscosity
Water viscosity
Gas viscosity
Solvent viscosity
Oil density
Water density
Gas density
Solvent density
Light oil volume fraction (STC) (API tracking option)
Porosity at current pressure (includes compressibility
compaction effects)
Relative permeability set number map
Permeability in each direction. This option is only available
when compaction is used and permeability multipliers vary
with pressure or when permeability is entered in the recurrent
data section
FLOW VECTORS
*FLUXSC
(*FLUX)
*FLUXRC
*VELOCSC
(*VELOC or *VELOSC)
*VELOCRC
(*VELORC)
STREAMLINES
*STRMLN
*RFG
*PCOW
*PCOG
*SBDZ
*EG
*BO
*RS
*RV
*DPP
*SEAWF
Flux and velocity vectors of oil, water and gas at surface

conditions and/or reservoir conditions.
Due to the large amount of data required to construct flow
vectors in Results, it is recommended that only a single flux
or velocity at either surface or reservoir conditions be output.
To reduce file size, flow output is not, by default, included
when *GRID *ALL (or EXCEPT) is chosen. Flow Vector
output must be explicitly selected.
Due to the large number of calculations required to
recalculate flow vectors it is also recommended that the
*WSRF *TNEXT option be used to explicitly select (and
hopefully reduce) the timesteps when Flow Vectors are
written to the SR2.
Triggers the output of Block Face based phase velocities to
be used to construct streamlines in Results3D.
To reduce SR2 file size, face velocity output is not, by
default, included when *GRID *ALL (or EXCEPT) is
chosen. Streamline output must be explicitly selected.
Due to the large number of calculations required to calculate
face velocities it is also recommended that the *WSRF
*TNEXT option be used to explicitly select (and hopefully
reduce) the timesteps when Streamline information is written
to the SR2.
Gas resistance factor for non-Darcy flow (1+ Forchheimer
number)
Water-oil capillary pressure (water-gas capillary pressure
when using *MODEL *GASWATER)
Oil-gas capillary pressure
Subsidence (per block and summed over layers)
Gas expansion factor
Oil formation volume factor
Solution gas ratio
Condensate oil content
Dew point pressure
Seawater Volume Fraction (Seawater Injection Model
Only)
The following keywords are not output automatically when *GRID *ALL or *GRID
*EXCEPT is used. The keywords must appear in a grid_list.
*KRWSCL
*KROWSCL
*KRGSCL
*KROGSCL

User's Guide IMEX
*PCWSCL

water (Swcon)
hysteresis only)
Current residual oil saturation of the imbibition process
hysteresis only)
*PCGSCL
*SORMAX
*SORH
sc-rc selector
*SC
*RC
Flux sector summary contains values in surface condition

Flux sector summary contains values in reservoir condition
If neither present, both *SC and *RC are selected.

ijk-con selector
*SUM
*ICON
*JCON
*KCON
Flux sector summary contains summation of connections in all three

directions
Flux sector summary contains flow across connections in I direction
Flux sector summary contains flow across connections in J direction
Flux sector summary contains flow across connections in K direction
If none of above selectors present, *SUM is selected.

DEFAULTS:
*OUTSRF
*GRID
*SO *SG *SW *SS *PCONC *PADSORP

*PRES *OILPOT *BPP *SSPRES
*WINFLUX
*OUTSRF *WELL *LAYER *NONE

*OUTSRF *RES *ALL
*XDR *ON
*OUTSRF
*FLUX-SECTOR *ALL *SC *SUM
CONDITIONS:
the simulation. *OUTSRF *RES, *OUTSRF *SPECIAL and *OUTSRF *FLUX-SECTOR
cannot appear in the Recurrent Data section of the input-data-file.
If *LAYER or *BLOCKP well information is requested then mobility weighted datum
pressure is also output.
The mobility weighted datum pressure is a weighted average of the datum pressures in the
grid blocks in which the well is completed, with the weighting factors proportional to the
total phase mobility in the completion grid block.
User's Guide IMEX
*TRMOI, *TRMOJ, and *TRMOK are not available for output to the SR2 file using the
*OUTSRF keyword. *TRMOI, *TRMOJ, and *TRMOK are only available for display in the
output file using the *OUTPRN keyword
The *PERM and all of the flow vector keywords cannot be used as subkeywords of the
*SPECIAL option. *PERM is used to view permeabilities which vary with pressure (see the
compaction keyword *CROCKTAB and *CROCKTABH) or with time if permeability is
redefined in the recurrent data section. If permeability does not vary with pressure or has not
been redefined in the recurrent data section *PERM will produce no output. *PERMI,
*PERMJ and *PERMK in the res_list can be used to view constant permeabilities.
*APIGRAV, *RFG, oil transmissibility arrays (*TRMOI, *TRMOJ, *TRMOK), saturation
endpoint arrays (*KRSETN, *SWCON, *SWCRIT, *SGCON, *SGCRIT, *SOIRW, *SORW,
*SLCON, *SORG, *SOIRG) and relative permeability endpoint arrays (*KRWSCL,
*KROWSCL, *KRGSCL, *KROGSCL, *PCWSCL and *PCGSCL) cannot be used as
subkeywords of the *SPECIAL option.
Because flux sectors and lease lines use the same LEASE origin in the SR2, the output
control for flux sector affects both. If a lease line is defined in the dataset, it automatically
switches on *OUTSRF *FLUX-SECTOR *ALL with *SC *SUM being selected. On other
hand, if *RC, *ICON, *JCON and *KCON are selected for flux sectors, corresponding values
for lease lines will be output as zero.
EXPLANATION:
An example of *OUTSRF, when using the list option is:
*OUTSRF *GRID *SO *SG *SW *PRES *OILPOT *BPP
User's Guide IMEX
Items in Diary File (Optional)
*OUTDIARY
PURPOSE:
*OUTDIARY controls some of the information written to the output-diary-file.
FORMAT:
*OUTDIARY
(*BRIEF | *MAXLOC | *WELLSTATUS)

(*PRESAQ | *PRESNOAQ)
(*HEADER freq)
DEFINITIONS:
*BRIEF
This subkeyword specifies that the grid block location (indices) where
maximum saturation or pressure changes occur will not be written to the
output-diary-file. The oil/gas/water production/injection rates will be written
with up to 8 significant figures. The water-cut will be written with up to 5
significant figures.
*MAXLOC
This subkeyword specifies that the grid block location (indices) where maximum
saturation or pressure changes occur will be written to the output-diary-file. The
oil/gas/water production/injection rates will be written with up to 5 significant
figures. The water-cut will be written with up to 3 significant figures.
*WELLSTATUS
This subkeyword specifies that the number of producing wells, injecting
wells, and shut in wells and the maximum changes of saturations and
pressure will be written to the output-diary-file. The oil/gas/water
production/injection rates will be written with up to 6 significant figures.
The water-cut will be written with up to 3 significant figures.
*PRESAQ
This subkeyword specifies that the total pore volume average pressure,
including the aquifer blocks, will be written to the output-diary-file.
*PRESNOAQ
This subkeyword specifies that the total pore volume average pressure,
excluding the aquifer blocks (blocks where water saturation equals 1), will be
written to the output-diary-file.
*HEADER
This subkeyword writes the table header to the output-diary-file every freq
timesteps. If *HEADER is not present the table header will be written every
20 timesteps.
User's Guide IMEX
freq
This integer specifies the number of timesteps after which the table header
will be written to the output-diary-file. If this integer is missing then the
table header will be written every 20 timesteps.
DEFAULTS:
Optional keywords. If not present in the input-data-file, the defaults are:
*OUTDIARY *BRIEF *PRESAQ *HEADER 20
CONDITIONS:
This keyword may appear in the Input/Output Control keyword group and the Recurrent Data
section.
EXPLANATION:
When the keyword *MAXLOC is present the grid blocks locations where maximum
saturation and pressure changes occur are written to the diary file. Then, due to space
limitations, the oil, gas and water production and injection rates and the water-cut have to be
written with less significant figures.
Only one of the three keywords *BRIEF, *MAXLOC and *WELLSTATUS should be used
for a given timestep. Likewise, *PRESAQ and *NOPRESAQ are exclusive.
The table header is written to the output-diary-file every "freq" timesteps or whenever the
subkeywords *BRIEF *MAXLOC *WELLSTATUS *PRESAQ or *PRESNOAQ are
changed.
Examples:
** Write grid block locations with maximum changes
** to the output-diary-file.
*OUTDIARY *MAXLOC
** Write table header after 50 timesteps.
*OUTDIARY *HEADER 50
** Average pressure does not include aquifer blocks.
*OUTDIARY *PRESNOAQ
** Write well status of the entire field
*OUTDIARY *WELLSTATUS
User's Guide IMEX
Production Split Output (Optional)
*PSPLIT
PURPOSE:
*PSPLIT Turns on limited production split output.
FORMAT:
*PSPLIT
(*ON)
(*OFF)
DEFINITIONS:
*PSPLIT
Turns on the output of surface production split information to the output and
SR2 files.
*PSPLIT *ON
Explicitly turns on the output of surface production split information to the
output and SR2 files.
*PSPLIT *OFF
Explicitly turns off the output of surface production split information to the
output and SR2 files. (Default)
DEFAULTS:
Optional keyword. If it is not present in the input-data-file, the default is not to print the split
information.
CONDITIONS:
This keyword may appear in the Input/Output Control keyword group, but not in the recurrent
data section, as it must not be changed during a simulation.
EXPLANATION:
The use of surface production streams divides up oil and gas phase information into surface
split based information. The splits are defined below.
1. Black Oil Split - Liquid at surface, Liquid at reservoir conditions
2. Condensate Split - Liquid at surface, Gas at reservoir conditions
3. Free Gas Split - Gas at surface, Gas at reservoir conditions
4. Solution Gas Split - Gas at surface, Liquid at reservoir conditions
For a black oil model there are two splits reported: the solution gas split and the free gas split
(which sum to the total gas produced at surface). For the condensate option, all four of the
splits listed above are reported.
Production splits from wells and from sectors (sum of wells in sector) are output to the ascii
and SR2 output files.
User's Guide IMEX
Debug Output (Optional)
*DEBUG
PURPOSE:
*DEBUG identifies what debug information is written to the output file.
FORMAT:
*DEBUG
(*JDUMP ts_range iter_range block_range)

(*VDUMP ts_range)
(*WELMAN *ON | *OFF | *WELLBORE)
(*PDUMP *ON | *OFF)
(*ECHO-RESTART *ON | *OFF)
(*SMOOTHKRPC*ON | *OFF)
(*XFLOW *ON | *OFF)
(*PARAMS)
(LSOLVER *ON | *OFF)
DEFINITIONS:
*JDUMP
Outputs the entire Jacobian Matrix in a format that the AIMSOL driver can
read.
ts_range
Range of timesteps in the format i1:i2, where the Jacobian or grid map is to
be printed.
iter_range
Range of Newtonian iterations in the format i1:i2 where the Jacobian matrix
is to be printed.
block_range
Range of grid blocks in the format i1:i2, for which the Jacobian is to be
printed.
*VDUMP
This keyword dumps the grid map from subroutine OUTIN after each
Newtonian iteration. This is very useful for tracking down convergence
problems.
*WELMAN
This keyword activates the debug information from the well-management
module.
*ON
Turn debug information on.
User's Guide IMEX
*OFF
Turn debug information off.
*WELLBORE
Print detailed information on wellbore hydraulic calculations.
*ECHO-RESTART
This keyword activates the keyword echo from the index-results-file during
restart.
*ON
*OFF
*SMOOTHKRPC
This keyword activates the debug information for *SMOOTH options
(*SWT, *SLT, and *SGT keywords). If this is enabled, then three files are
written:
*SMOOTHKRPC-IN, *SMOOTHKRPC-OUT and *SMOOTHKRPC-SUM.
These three files have default endings of .kri, .kro and .krs. The .kri file
contains the input relative permeability and capillary pressure tables. The .kro
file contains the internally calculated tables. The .krs file contains a summary
of the *SMOOTH calculations. If *OUTPRN *TABLES *ALL is enabled
(the default) and *SMOOTHKRPC is enabled then both the input tables and
the internally calculated tables are written to the output file. If *OUTPRN
*TABLES *ALL is enabled (the default) and *SMOOTHKRPC is not enabled
then only the internally calculated tables are written to the output file.
*ON
*OFF
*XFLOW
Activated debug information for the *XFLOW-MODEL keyword. Cycle by
cycle information on layer crossflow status is printed out if debug information
is turned on
*ON
User's Guide IMEX
*OFF
*PARAMS
When using run time dimensioning, print out the 'Virtual' parameter file
generated for IMEX. These values are calculated and used to run time
dimension IMEX and determine common block size. Use of this keyword
and use of the *DIM keyword can allow the user to tune the dimensioning of
the IMEX simulator.
*LSOLVER
Turn on summary output of the number of solver failures.
DEFAULTS:
Optional keyword. If it is not present in the input-data-file, the defaults are not to print the
debug information.
CONDITIONS:
the simulation.
EXPLANATION:
An example of *DEBUG, when using the *JDUMP option is:
** Dump derivatives at timestep 7 for iterations
** 2 and 3; blocks 244, 245 and 246
*DEBUG *JDUMP 7 2:3 244:246
This keyword must be in the Input/Output Control section or the Recurrent Data section.
There may be more than one *DEBUG *JDUMP keywords.
An example of *DEBUG, when using the *VDUMP option is:
** Dump primary variables at timesteps 7, 8 and 9
** during Newtonian iterations.
*DEBUG *VDUMP 7:9
This keyword must be in the Input/Output Control section or the Recurrent Data section.
There may be more than one *DEBUG *VDUMP keywords.
User's Guide IMEX
Fundamental Grid Definition (Required)
*GRID
PURPOSE:
*GRID defines the fundamental (main) grid, and marks the beginning of the reservoir
description.
FORMAT:
*GRID
(*CART)
(*VARI)
(*CORNER)
(*RADIAL)
ni nj nk
ni nj nk
ni nj nk
ni nj nk (*RW rw)
DEFINITIONS:
*CART
Keyword indicating a rectangular Cartesian grid, with uniform depth/uniform
thickness layers.
*VARI
Keyword indicating a rectangular grid allowing variable depth/variable
thickness layers.
*CORNER
Keyword indicating a corner point grid, as described in the following.
*RADIAL
Keyword indicating a radial-angular cylindrical grid.
ni
Number of grid blocks in the I direction.
nj
Number of grid blocks in the J direction.
nk
Number of grid blocks in the K direction.
User's Guide IMEX
Reservoir Description 283
*RW
The radius of the wellbore located in the innermost radial grid block(s) for
the *RADIAL grid.
rw
Radius of the wellbore (m | ft). Should be nonzero.
DEFAULTS:
Required keyword. No defaults.
If *RW is missing or 0.0, a default of 0.0762 m (0.25 ft) is supplied.
CONDITIONS:
This keyword should be the first keyword in the Reservoir Description keyword group.
EXPLANATION:
The keyword *GRID indicates the grid type and the number of grid blocks within the
fundamental (main) grid system. There is no default and one of the aforementioned
subkeywords must be entered.
The *GRID keyword defines a grid consisting of ni * nj * nk blocks. The blocks are
rectangular in shape for Cartesian and variable depth/variable thickness grids. They should
be close to rectangular for corner point grids. Radial-angular cylindrical grids consist of
stacked circular rings, where the rings may have angular subdivisions.
A wellbore volume of radius "rw" is removed from the innermost ring in all layers of
*RADIAL grids. This gives the innermost ring a circular inner boundary that connects to the
wellbore. The wellbore volume is not included in the innermost rings.
I, J, and K indices are used to identify the blocks, where I runs in the range of 1 to ni, J runs
in the range of 1 to nj and K runs in the range of 1 to nk. The notation (I,J,K) will sometimes
be used to denote a block. Blocks are ordered (numbered) with I increasing fastest, J next
fastest, and K slowest.
For Cartesian and variable depth/variable thickness grids, I corresponds to the "x" direction, J
to the "y" direction, and K to the "z" direction, where "x", "y" and "z" refer to a standard
(right handed) coordinate system in the reservoir.
Corner point cells have all their corners specified in terms of an "x-y-z" triple defined by the
incoming data. However, the I-J-K numbering system is still important because the default
search methods for determining connections assume that incrementing the I or J index for
each cell, and searching the resulting stack of potential neighbours by sweeping through K,
should find all contacting cells. Thus, I-J planes should roughly correspond to geological
layers indexed by K, even if the I axis and the "x" axis, or the J axis and the "y" axis, are not
aligned.
For radial-angular cylindrical grids, I corresponds to the "r" (radial) direction, J to the
"angular" (theta) direction, and K to the "z" direction.
284 Reservoir Description
User's Guide IMEX
By default, K increases vertically UPWARDS (against gravity) for all non-corner point grids.
This assumption can be reversed, making the K index increase downwards with gravity, by
using the *KDIR *DOWN keywords (see the *KDIR keyword following).
K always increases DOWNWARDS for corner point grids (meaning *KDIR *DOWN is
always assumed for corner point grids). This setting cannot be changed.
A summary follows:
Grid Type
Layer Ordering
*CART
*VARI
*CORNER
*RADIAL
x
x
x
r
y
y
y
theta
z
z
z
z
upward (default)
upward (default)
always (downwards)
upward (default)
The default cases in the above table can be modified by use of the *KDIR *DOWN keywords.
The grid can be tilted with respect to the gravity vector in some cases by use of the *DIP
keyword (see the *DIP keyword following).
Rectangular:
If a Cartesian or variable depth/variable thickness grid is examined with the I index
increasing towards the viewer, J should increase from left to right and K should increase
upwards, provided that *KDIR has not been used (or *KDIR *UP is set). If a Cartesian or
variable depth/variable thickness grid with *KDIR *DOWN, or a corner point grid, is
examined with the J index increasing towards the viewer, I should increase from left to right
and K should increase downwards.
The coordinate system for a Cartesian or variable depth/variable thickness grid with the no
KDIR specified (default), or *KDIR *UP specified, appears as follows:
K = "z"
DEFAULT *KDIR
(or *KDIR *UP)
J = "y"
90 o
Gravity
I = "x"
User's Guide IMEX
or viewing the above from a different vantage point:
K = "z"
J = "y"
Gravity
DEFAULT *KDIR
(or *KDIR *UP)
90 o
I = "x"
The coordinate system for a Cartesian or variable depth/variable thickness grid with *KDIR
*DOWN specified, or a corner point grid, appears as follows:
I="x"
90 o
Gravity
Corner point
or *KDIR *DOWN
J=:"y"
K="z"
Radial:
*RADIAL radial-angular cylindrical grids have rings indexed by I, counting from the
innermost outwards. Each ring may be split into "nj" sectors indexed by J, counting
counterclockwise around the K axis with the first sector lying over the I axis. Layers are
indexed by K, starting at the reservoir bottom and counting upwards if no *KDIR keyword
appears (or if *KDIR *UP is set), and counting downwards if *KDIR *DOWN is set.
Note that the first ring of *RADIAL grids is not subdivided even if nj exceeds 1, and will
exclude a volume of radius "rw"; the latter volume is not assigned an I index value. Thus, the
inner radius of the first (I = 1) grid block is given by "rw" and its outer radius is given by the
sum of "rw" and the length of the first block in the I direction.
User's Guide IMEX
The coordinates for a radial-angular cylindrical grid with the no KDIR specified (default), or
*KDIR *UP specified, appears as follows:
K
DEFAULT *KDIR
(or *KDIR *UP)
W is the Wellbore
s
RING I = 1
RING I = 2
s
s
"y"
J=2
Gravity
s
J=1
*GRID *RADIAL
"x"
and when *KDIR *DOWN is set:
RING I = 1
RING I = 2
s
s
.
"y"
J=1
.
.
"x"
s
J=2
W is the Wellbore
*KDIR *DOWN
Gravity
.
.
*GRID *RADIAL
Grid block orderings are further described in Appendix D. FIGURE 1a shows Cartesian and
variable depth/variable thickness grid systems that are NOT corner point grids, FIGURE 1b
shows a corner point grid, and FIGURE 2 is for a *RADIAL radial-angular cylindrical system.
The grid block nodes are the points where the pressures are evaluated. They are placed at the
block centres for rectangular grids. The nodes for *RADIAL blocks are placed in accordance
with an assumption made about the discretization of the accumulation terms in the simulator.
The *RADIAL node is chosen as the point where the block pressure is equal to its volumeweighted average for the steady-state flow of a radially-directed incompressible fluid through
the block. Details can be found in O.A. Pedrosa and K. Aziz, "Use of a Hybrid Grid in
Reservoir Simulation", SPERE, November 1986, pp. 611-621.
User's Guide IMEX
Corner Point:
Corner point grids are made up of blocks which are defined by their eight corner points. Each
corner point is described by giving its three coordinates, an "x"-, "y"- and "z"-coordinate,
locating it in the subsurface. The "x"- and "y"- coordinates have usually been projected
against a horizontal reference surface. The "z"- coordinate is usually the depth of the corner
point measured downwards from that surface. Both positive or negative depths are valid,
depending on the location of the reference surface with respect to the reservoir, although
positive values are most common.
It takes 24 numerical values to determine a general corner point block. Certain keywords
place assumptions on the corner point data however, so that it is not always necessary to read
24 values per block to define the grid. Details of corner point input are given later (see, for
instance, the *COORD and *ZCORN keywords).
The following is a model for a corner point block, giving labels for its corners:
N
W
NE-T
NW-T
SE-T
SW-T
NW-B
NE-B
SW-B
SE-B
The block is the volume contained within the 6 faces, where each face is made by connecting the
corner points with line segments as shown, and filling in the faces with a nonlinear (bilinear)
interpolation. Faces need not be planar. The numerical results will be better if the cell shape is
kept close to a rectangle however. The cell's barycentre is used as the pressure node.
IMEX performs some tests to see if cells are not too distorted, however examining grids in
advance with a visualization package is always recommended. Intrusions of one cell into
another should be avoided.
Cells can make flow connections in two different ways. The usual situation is when four
corners defining one face on one cell match (within a tolerance) the four corners defining a
face on another cell. Projections (onto both sides) of the common face will be used in the
transmissibility calculation for flow between the cells.
The other situation involves vertical faulting. If one of the side faces (not the top or bottom)
of a cell is planar, and it overlaps a planar side face on another cell, and the two left and the
two right edges of each face are collinear (that is, they lie on the same line), then a flow
connection will be made based on the overlap area. An example follows with the shared
edges marked:
User's Guide IMEX
< Right Edge
Left Edge >
Cell 1
Cell 2
Note that a cell's vertical faces need not always be planar although this is often the case.
Corner point grids defined using the *COORD keyword (see later) involve cells that share
vertical edges, and if the *COORD lines are parallel, their side faces will be planar. Note
also that the face overlap need only occur to within a tolerance.
The tolerance level for the above cases is controlled by *CORNER-TOL (see following).
Note that if cells intrude on each other to too large a degree, then corner points and faces
cannot make contact and flow will not occur.
Examples:
a) Rectangular Cartesian grid with ten blocks in the "x" direction, five blocks in the
"y" direction, and four blocks in the "z" direction (ni=10, nj=5, nk=4). Enter:
*GRID *CART 10 5 4
b) Variable depth/variable thickness rectangular grid with ten blocks in the "x"
direction, three blocks in the "y" direction, and one layer in the "z" direction
(ni=10, nj=3, nk=1). Enter:
*GRID *VARI 10 3 1
c) Corner point grid with 20 blocks in the "x" direction, 20 blocks in the "y"
direction, and 5 layers (ni=20, nj=20, nk=5). Enter:
*GRID *CORNER 20 20 5
d) Radial-angular cylindrical grid with 15 blocks in the radial direction, no radial
subdivisions, and 5 layers (ni=15, nj=1, nk=5). Enter:
*GRID *RADIAL 15 1 5
User's Guide IMEX
K Direction Index (Optional)
*KDIR
PURPOSE:
*KDIR controls whether increasing K means going deeper or shallower in the reservoir.
FORMAT:
*KDIR
(*UP)
(*DOWN)
DEFINITIONS:
*UP
Indicates that the K index increases upward, so larger K means shallower
grid blocks.
*DOWN
Indicates that the K index increases downward, so larger K means deeper
grid blocks.
DEFAULTS:
Optional keyword. Default is *KDIR *UP, except for corner point grids when *KDIR
*DOWN is the default. (The corner point grid setting can not be changed.)
CONDITIONS:
This keyword, if present, must be in the Reservoir Description keyword group.
EXPLANATION:
By default, the K index increases upward, except for corner point grids, when it increases
downwards. See the *GRID keyword discussion for more details.
User's Guide IMEX
Examples:
*KDIR *UP
K=3
Gravity
K=2
K=1
I=1
I=2
I=3
I=4
*KDIR *DOWN
K=1
Gravity
K=2
K=3
I=1
I=2
*KDIR *DOWN
I=3
I=4
(I-K View)
K=1
Gravity
K=2
K=3
I=1
I=2
I=3
I=4
(I-J View)
J=1
J=2
J=3
I=1
User's Guide IMEX
I=2
I=3
I=4
Block Dimensions for the I Direction (Conditional)
*DI
PURPOSE:
*DI signals input of an array of grid block lengths for the I direction. For rectangular grids,
the values are block widths measured in the I direction and for radial-angular cylindrical
grids, the values are block widths measured in the radial direction.
ARRAY:
*DI
DEFAULTS:
Conditional keyword. No defaults.
CONDITIONS:
This keyword must be in the Reservoir Description keyword group. It is required for all grids
except corner point grids, for which it is optional.
Several different methods are available for specifying block dimensions for corner point
grids, including the use of the *DI keyword. See the *ZCORN keyword for a further
discussion of corner point input.
All array reading options are valid; however, all blocks with the same I index must have the
same block length. The most commonly used array reading subkeywords are the *IVAR and
*CON options.
EXPLANATION:
The keyword *DI defines the dimensions of the grid blocks in the I direction.
Examples:
a) I direction grid increments for a Cartesian grid with ni=10 are:
1000,1000,1500,400,400,400,400,400,1000,1000.
Use:
*DI *IVAR 2*1000 1500 5*400 2*1000
b) I direction grid increments for a Cartesian grid are all 1200. Use:
*DI *CON 1200
-or*DI *CON
1200
c) The widths of ni=10 radial-angular cylindrical blocks measured radially are:
2.00
50.55
2.32
109.21
5.01
235.92
10.84
509.68
23.40
1101.08
User's Guide IMEX
Use:
*DI *IVAR
2.0
50.55
2.32
109.21
5.01
235.92
10.84
509.68
23.4
1101.08
Note that the first block's inner radius is "rw" and its outer radius is "rw"+2.
The acceptable range of values for block lengths in the I direction is:
min
max
User's Guide IMEX
SI
m
1.0E-4
1.0E+20
Field
ft
3.28E-4
3.28E+20
Lab
cm
1.0E-2
1.0E+22
Block Dimensions for the J Direction (Conditional)
*DJ
PURPOSE:
*DJ signals input of an array of grid block lengths for the J direction. For rectangular grids, the
values are block widths measured in the J direction and for radial-angular cylindrical grids, the
values are angular extents for portions of the subdivided rings, expressed in degrees.
ARRAY:
*DJ
DEFAULTS:
CONDITIONS:
except corner point grids, for which it is optional.
Several different methods are available for specifying block dimensions for corner point
grids, including the use of the *DJ keyword. See the *ZCORN keyword for a further
discussion of corner point input.
All array reading options are valid; however, all blocks with the same J index must have the
same block length/angular extent. The most commonly used array reading subkeywords are
the *JVAR and *CON options.
EXPLANATION:
The keyword *DJ defines the dimensions of the grid blocks in the J direction. Note that
angular extents should be expressed in degrees.
Examples:
a) J direction grid increments for a Cartesian grid with nj=10 are:
2000,2000,2500,4000,1500,1500,400,400,1000,2000.
Use:
*DJ *JVAR
2*2000 2500 4000 2*1500 2*400 1000 2000
b) J direction grid increments for a Cartesian grid are all 2200. Use:
*DJ *CON 2200
-or*DJ *CON
2200
c) A *RADIAL grid has no angular subdivisions (nj = 1). Use:
*DJ *CON 360
User's Guide IMEX
The acceptable range of values for block lengths in the J direction is:
min
max
SI
m
1.0E-4
1.0E+20
Field
ft
3.28E-4
3.28E+20
Lab
cm
1.0E-2
1.0E+22
The acceptable range of values for angles is:
min
max
User's Guide IMEX
SI
degrees
0.0
360.0
Field
degrees
0.0
360.0
Lab
degrees
0.0
360.0
Block Dimensions for the K Direction (Conditional)
*DK
PURPOSE:
*DK signals input of an array of (gross) grid block thicknesses measured in the K direction.
ARRAY:
*DK
DEFAULTS:
CONDITIONS:
except corner point grids. K direction grid block lengths should not be specified with the
*DK keyword for corner point grids; see the corner point keyword descriptions following,
such as *ZCORN.
All array reading options are valid; however, recall that all blocks with the same K index
must have the same thickness for Cartesian grids; that is, the layers in Cartesian grids have
uniform thicknesses. Different thicknesses in the same layer (same K) are permitted for
*GRID *VARI and *GRID *RADIAL however.
Blocks can be assigned a zero thickness if they are to be considered as pinched out (see the
discussions for *PINCHOUTARRAY and *PINCHOUT-TOL keyword following).
EXPLANATION:
This keyword defines the dimensions of the grid blocks in the K direction. These dimensions
are typically gross grid block thicknesses.
Examples:
a) K direction grid increments for a Cartesian grid with nk=8 are:
20,20,25,40,15,45,45,45.
Use:
*DK *KVAR
2*20 25 40 15 3*45
b) K direction grid increments for a Cartesian grid are all 22. Use:
*DK *CON 22.0
-or*DK *CON
22.0
User's Guide IMEX
The acceptable range of values for block lengths in the K direction is:
min
max
User's Guide IMEX
SI
m
1.0E-4
1.0E+20
Field
ft
3.28E-4
3.28E+20
Lab
cm
1.0E-2
1.0E+22
Depth to the Centre of a Grid Block (Conditional)
*DEPTH
PURPOSE:
*DEPTH indicates input of a reservoir depth for a single grid block. This depth is usually to
be measured to the block's centre, unless *TOP appears (see below).
FORMAT:
*DEPTH
(*TOP)
(*CENTRE)
i j k depth
DEFINITIONS:
*TOP
Subkeyword indicating that the depth is to the top (centre of the top face) of
the reference block.
*CENTRE
Subkeyword indicating that the depth is to the centre of the reference block.
i
I direction index of the reference block.
j
J direction index of the reference block.
k
K direction index of the reference block.
depth
Depth to the centre (or top if *TOP is used) of the reference block in the
reservoir (m | ft). The value may be of any sign.
DEFAULTS:
Conditional keyword. No defaults. *CENTRE is assumed if *TOP does not appear.
CONDITIONS:
This keyword must be in the Reservoir Description keyword group. One of *DEPTH,
*DTOP, *PAYDEPTH or *DEPTH-TOP must be specified for *GRID *CART, *GRID
*VARI, or *GRID *RADIAL. This keyword should not be used with corner point grids. If
depth modifications are required for corner point grids, the *PAYDEPTH or *DEPTH-TOP
keyword can be used.
EXPLANATION:
Depths are measured downwards from a horizontal reference surface. The I, J, K indices
describe a grid block whose depth is known, the depth being measured to the centre of the
grid block. The value may be positive or negative depending on the location of the reference
surface, although positive values are most common.
User's Guide IMEX
At least one depth is required for any simulation. Note that corner point grids use special
keywords to provide their depth information, or can obtain depth information from the actual
corner point coordinates.
Depths are assigned to all blocks based on the depth provided by this keyword. This
calculation is made based on the blocks' thicknesses (*DK keyword) and the dip angles
provided by the *DIP keyword (see *DIP keyword description following).
Example:
*DEPTH 1 1 1 2000.0
The acceptable range of values for depth is:
min
max
User's Guide IMEX
SI
m
-1.0E+4
1.0E+4
Field
ft
-32,808.0
32,808.0
Lab
cm
-1.0E+6
1.0E+6
Depth to the Tops of Grid Blocks (Conditional)
*DTOP
PURPOSE:
*DTOP indicates input of a number of depths that provide the depth to the centre of the top
face of each grid block in the top layer of the grid.
ARRAY:
*DTOP
DEFAULTS:
CONDITIONS:
*VARI, or *GRID *RADIAL. Use of this keyword, or *PAYDEPTH, is recommended for
*GRID *VARI. *PAYDEPTH or *DEPTH-TOP, not *DTOP, can be used for corner point
grids if depth modifications are required.
If this keyword is used with *GRID *CART, the values in the *DTOP array must all be the same.
No array qualifiers or array reading options are permitted for this particular array keyword.
A fixed number of values (ni * nj) is always expected.
EXPLANATION:
This keyword is usually used to define the depths of grid blocks for a variable depth/variable
thickness grid (*GRID *VARI). A total of ni * nj depth values must be entered. The values
are to be measured downwards from a horizontal reference surface to the centre of the tops of
the grid blocks in the upper-most layer. The values may be positive or negative depending on
the location of the reference surface. They are to be entered row by row with the I index
changing fastest, and the J index slowest.
Note that the K index assumed for this array will be that of the uppermost layer; that is, it will
be K = nk if *KDIR does not appear in the data set, or if *KDIR *UP has been specified, or it
will be K = 1 if *KDIR *DOWN appears.
Depths are assigned to all blocks based on the depths provided by this keyword and the
blocks' thicknesses (*DK keyword).
See FIGURE 5 in Appendix D.
Example:
A variable depth/variable thickness grid with ni=6, nj=4 and nk=2 might use the following:
*DTOP
1000.0
1070.0
1000.0
1070.0
1300.0
1090.0
1200.0
1100.0
1250.0
1080.0
1110.0
1100.0
1100.0
1110.0
1200.0
1170.0
1200.0
1120.0
1200.0
1070.0
1070.0
1200.0
1190.0
1070.0
User's Guide IMEX
The acceptable range of values for depths is:
min
max
User's Guide IMEX
SI
m
-1.0E+4
1.0E+4
Field
ft
-32,808.0
32,808.0
Lab
cm
-1.0E+6
1.0E+6
Depths to Centre of Pay (Conditional)
*PAYDEPTH
PURPOSE:
*PAYDEPTH indicates input of depths to the centre of the net pay for each grid block in the
reservoir. (Net pay is assumed to be centered in the grid block.)
ARRAY:
*PAYDEPTH
DEFAULTS:
CONDITIONS:
*VARI, or *GRID *RADIAL. Use of this keyword, or *DTOP, is recommended for *GRID
*VARI.
*PAYDEPTH can be specified for corner point grids. The *PAYDEPTH values will
override depths calculated from the "z" components of the corner point locations.
If this keyword is used with *GRID *CART, the depths in each layer (blocks with the same
K index) must agree, and the depth differences between layers must be consistent with the
gross thicknesses (*DK keyword).
EXPLANATION:
This keyword defines the depths to the pay of each individual grid block. All array qualifiers
and array reading options are allowed for specifying the ni * nj * nk values.
The values are to be measured downwards from a horizontal reference surface to the centre of
the grid block, which is where the net pay is assumed to be positioned. The values may be
positive or negative depending on the location of the reference surface in the reservoir,
although positive values are most common.
Since it is assumed that the (net) pay is centered in the block, the *PAYDEPTH array's
depths can be directly assigned to each block's node.
When used for corner point grids, the paydepth values override depths calculated from the "z"
components of the corner point locations. If a *PAYDEPTH value is not set for certain cells
(as when *PAYDEPTH is used with the *IJK option and not all cells are touched) then
depths for the remaining cells will revert to those taken from the "z" components. The actual
corner point locations are not altered by *PAYDEPTH, and grid visualizations are
unaffected. Only the "Depth to Centers" array in the output echo (use *OUTPRN *RES
*ALL) shows the results of using *PAYDEPTH with corner point grids. Use of
*PAYDEPTH with corner point grids works like a vertical position modifier for the cells.
User's Guide IMEX
Example:
*PAYDEPTH *ALL
1000.0 1300.0
1070.0 1090.0
1000.0 1200.0
1070.0 1100.0
2000.0 2300.0
2070.0 2090.0
2000.0 2200.0
2070.0 2100.0
1250.0
1080.0
1110.0
1100.0
2250.0
2080.0
2110.0
2100.0
1100.0
1110.0
1200.0
1170.0
2100.0
2110.0
2200.0
2170.0
1200.0
1120.0
1200.0
1070.0
2200.0
2120.0
2200.0
2070.0
1070.0
1200.0
1190.0
1070.0
2070.0
2200.0
2190.0
2070.0
min
max
User's Guide IMEX
SI
m
-1.0E+4
1.0E+4
Field
ft
-32,808.0
32,808.0
Lab
cm
-1.0E+6
1.0E+6
Depths to Top of Block (Conditional)
*DEPTH-TOP
PURPOSE:
*DEPTH-TOP indicates input of depths to the top of each grid block in the reservoir.
ARRAY:
*DEPTH-TOP
DEFAULTS:
CONDITIONS:
*DTOP *DEPTH-TOP or *PAYDEPTH must be specified for *GRID *CART, *GRID
*VARI, or *GRID *RADIAL. Use of this keyword, *PAYDEPTH or *DTOP, is
recommended for *GRID *VARI.
*DEPTH-TOP can be specified for corner point grids. The *DEPTH-TOP values will
override depths calculated from the "z" components of the corner point locations.
If this keyword is used with *GRID *CART, the depths in each layer (blocks with the same
K index) must agree, and the depth differences between layers must be consistent with the
gross thicknesses (*DK keyword).
EXPLANATION:
This keyword defines the depths to the top of each individual grid block. All array qualifiers
and array reading options are allowed for specifying the ni * nj * nk values.
The values are to be measured downwards from a horizontal reference surface to the top of
the grid block. The values may be positive or negative depending on the location of the
reference surface in the reservoir, although positive values are most common.
When used for corner point grids, the depth to top values override depths calculated from the
"z" components of the corner point locations. If a *DEPTH-TOP value is not set for certain
cells (as when *DEPTH-TOP is used with the *IJK option and not all cells are touched) then
depths for the remaining cells will revert to those taken from the "z" components. The actual
corner point locations are not altered by *DEPTH-TOP, and grid visualizations are
unaffected. Only the "Depth to Centers" array in the output echo (use *OUTPRN *RES
*ALL) shows the results of using *DEPTH-TOP with corner point grids. Use of *DEPTHTOP with corner point grids works like a vertical position modifier for the cells.
User's Guide IMEX
Example:
*DEPTH-TOP *ALL
1000.0 1300.0
1070.0 1090.0
1000.0 1200.0
1070.0 1100.0
2000.0 2300.0
2070.0 2090.0
2000.0 2200.0
2070.0 2100.0
1250.0
1080.0
1110.0
1100.0
2250.0
2080.0
2110.0
2100.0
1100.0
1110.0
1200.0
1170.0
2100.0
2110.0
2200.0
2170.0
1200.0
1120.0
1200.0
1070.0
2200.0
2120.0
2200.0
2070.0
1070.0
1200.0
1190.0
1070.0
2070.0
2200.0
2190.0
2070.0
min
max
User's Guide IMEX
SI
m
-1.0E+4
1.0E+4
Field
ft
-32,808.0
32,808.0
Lab
cm
-1.0E+6
1.0E+6
Grid Tilt Angles (Conditional)
*DIP
PURPOSE:
*DIP specifies the input of dip angles.
FORMAT:
*DIP
*DIP
idip
idip
jdip
(for *GRID *CART)

(for *GRID *RADIAL)
DEFINITIONS:
idip
Tilt angle in degrees of the I axis above the horizontal.
jdip
Tilt angle in degrees of the J axis above the horizontal. Not used for *GRID
*RADIAL.
DEFAULTS:
CONDITIONS:
This keyword must be in the Reservoir Description keyword group. This keyword is optional
with *GRID *CART and *GRID *RADIAL, and is not used with *GRID *VARI or *GRID
CORNER. The *DIP keyword should not be specified when the *DTOP, *PAYDEPTH or
*DEPTH-TOP keywords are used to input depths.
The "jdip" parameter is not required and should not be input for *GRID *RADIAL.
EXPLANATION:
The *DIP keyword allows a grid to tilt with respect to the gravity direction. It is generally
used with *DEPTH for *GRID *CART and *GRID *RADIAL grids. *GRID *VARI and
*GRID *CORNER have sufficient depth input flexibility provided by alternate keywords, so
that *DIP is not required for these grids. (The effect of using the *DEPTH and *DIP
keywords could be duplicated by using an externally generated array of depths entered via the
*PAYDEPTH or *DEPTH-TOP keywords.)
The angles "idip" and "jdip" are measured in degrees and should lie between -90 to +90
degrees.
A figure for calculating "idip" follows, where the reference horizontal axis "H H H" is at right
angles to the gravity vector and lies in the plane of the gravity vector and the I = "x" axis.
This example shows a negative dip angle (I lies below H):
User's Guide IMEX
H
H
NOTE: dipi has a

negative sign.
H
H
dipi
Gravity
V
I
A similar construction involving the J axis obtains "jdip".

For *RADIAL grids, it is assumed that the centre of the first sector lies in the plane defined
by the K axis and the gravity vector, and that the I axis is drawn down the centre of the first
sector. Thus, a rotated version of the above figure provides a description of how to measure
"idip".
See FIGURE 3 in Appendix D for further information regarding Cartesian grids, and
FIGURE 4, also in Appendix D, for radial-angular cylindrical grids.
Example:
*DIP
1.0 0.0
The acceptable range of values for dip angles is:
min
max
User's Guide IMEX
SI
Field
Lab
-90.0
90.0
-90.0
90.0
-90.0
90.0
Corner Point Depths for Corner Point Grids (Conditional)

*ZCORN
PURPOSE:
*ZCORN signals input of an array of corner point depths for corner point grids.
ARRAY:
*ZCORN
DEFAULTS:
CONDITIONS:
This keyword must be in the Reservoir Description keyword group and should only be used
with corner point grids (*GRID *CORNER). This keyword should be combined with *DI
and *DJ, or with *COORD, or with *XCORN and *YCORN, to define all the corner point
locations for a corner point grid.
The *RG qualifier can be used with this keyword.
EXPLANATION:
See the general corner point discussion given with the *GRID *CORNER keyword for
discussions of the notation used here.
The *ZCORN keyword causes the reading of all depths ("z"-coordinates) of the 8*ni*nj*nk
corner points required to define the grid. The depths should be input as described in the
following algorithm.
Algorithm for *ZCORN Ordering:
Operations should be done in the order shown. Note that the text before each task ([...])
describes how often to carry out that task.
Do the following for K = 1, ..., nk: [
Do the following for J = 1, ..., nj: [
Write NW-T and NE-T depths for block ( 1,J,K),
...
Write NW-T and NE-T depths for block (ni,J,K).
Write SW-T and SE-T depths for block ( 1,J,K),
...
Write SW-T and SE-T depths for block (ni,J,K).
]
Write NW-B and NE-B depths for block ( 1,J,K),
...
Write NW-B and NE-B depths for block (ni,J,K).
Write SW-B and SE-B depths for block ( 1,J,K),
...
Write SW-B and SE-B depths for block (ni,J,K).
] ]
User's Guide IMEX
This completes the algorithm.

The "x"- and "y"-coordinates of the corner points must be provided by the *DI and *DJ
keywords, or by the *COORD keyword, or by the *XCORN and *YCORN keywords. If the
*DI and *DJ keywords are used, it will be assumed that corner point NW-T of block (1,1,1)
is at "x"-coordinate 0.0 and "y"-coordinate 0.0 with increments to be provided by the *DI and
*DJ keywords.
Examples:
Provide corner point depths for a ni = 4, nj = 2, nk = 1 *CORNER grid for a reservoir dipping
in the "x"-coordinate direction whose layers are a constant 10 units thick.
*ZCORN
2000
2000
2000
2000
2010
2010
2010
2010
2001
2001
2001
2001
2011
2011
2011
2011
2001
2001
2001
2001
2011
2011
2011
2011
2002
2002
2002
2002
2012
2012
2012
2012
2002
2002
2002
2002
2012
2012
2012
2012
2003
2003
2003
2003
2013
2013
2013
2013
2003
2003
2003
2003
2013
2013
2013
2013
2004
2004
2004
2004
2014
2014
2014
2014
The acceptable range of values for corner points depths is:
min
max
User's Guide IMEX
SI
m
-1.0E+20
1.0E+20
Field
ft
-3.28E+20
3.28E+20
Lab
cm
-1.0E+22
1.0E+22
Lateral Corner Point Locations for Corner Point Grids

(Conditional)
*XCORN, *YCORN
PURPOSE:
*XCORN signals input of an array of corner point "x"-coordinate locations for corner point
grids.
*YCORN signals input of an array of corner point "y"-coordinate locations for corner point
grids.
ARRAY:
*XCORN
-or*YCORN
DEFAULTS:
CONDITIONS:
These keywords must be in the Reservoir Description keyword group and should only be used
with corner point grids (*GRID *CORNER). Both keywords should appear, and be combined
with *ZCORN to define all the corner point locations for a corner point grid.
EXPLANATION:
See the general corner point discussion given for the *GRID *CORNER keyword for
The *XCORN and *YCORN keywords each cause reading of all the (ni+1)*(nj+1)*(nk+1)
"x"- and "y"-coordinate values required to define the lateral locations of all points defining a
*CORNER grid. The values should be input as described in the following algorithm.
Algorithm for *XCORN/*YCORN Ordering:
Do the following for K = 1, ..., (nk + 1): [
Do the following for J = 1, ..., (nj + 1): [
Do the following for I = 1, ..., (ni + 1): [
I, J, K are less than ni, nj, nk, respectively:
write the "x"- (or "y"-) coordinate of the NW-T point;
J is less than nj, K is less than nk, and I = ni:
write the "x"- (or "y"-) coordinate of the NE-T point;
I is less than ni, K is less than nk, and J = nj:
write the "x"- (or "y"-) coordinate of the SW-T point;
User's Guide IMEX
I is less than ni, J is less than nj, and K = nk:

write the "x"- (or "y"-) coordinate of the NW-B point;
I is less than ni, and J = nj, K = nk:
write the "x"- (or "y"-) coordinate of the SW-B point;
J is less than nj, and I = ni, K = nk:
write the "x"- (or "y"-) coordinate of the NE-B point;
K is less than nk, and I = ni, J = nj:
write the "x"- (or "y"-) coordinate of the SE-T point;
I = ni, J = nj, K = nk:
write the "x"- (or "y"-) coordinate of the SE-B point;
where the choice of "x"- or "y"- is determined by
whether *XCORN or *YCORN is being written.
] ] ]
This completes the algorithm. Note that I is ranging fastest, and K slowest, in the above; J is
intermediate.
Examples:
Provide the *XCORN and *YCORN data for a ni = 4, nj = 2, nk = 1 *CORNER grid. Note that
the "x"- direction grid spacing is uniformly 100 units and the "y"-direction grid spacing is
uniformly 200 units.
*XCORN
0 100
0 100
0 100
0 100
0 100
0 100
*YCORN
0
0
200 200
400 400
0
0
200 200
400 400
200
200
200
200
200
200
300
300
300
300
300
300
400
400
400
400
400
400
0
200
400
0
200
400
0
200
400
0
200
400
0
200
400
0
200
400
The acceptable range of values for corner point coordinates is:
min
max
User's Guide IMEX
SI
m
-1.0E+20
1.0E+20
Field
ft
-3.28E+20
3.28E+20
Lab
cm
-1.0E+22
1.0E+22
Line-Based Corner Point Locations for Corner Point Grids

(Conditional)
*COORD
PURPOSE:
*COORD signals input of an array of "x"- and "y"- coordinate corner point location
information for corner point grids.
ARRAY:
*COORD
DEFAULTS:
CONDITIONS:
with corner point grids (*GRID *CORNER). This keyword should be combined with
*ZCORN to define all the corner point locations for a corner point grid.
*COORD is especially useful when faults are present.
EXPLANATION:
See the general corner point discussion given for the *GRID *CORNER keyword for
The *COORD keyword causes the reading of information defining the "x"- and "y"coordinate locations for all corner points defining a *CORNER grid. *COORD expects that
the corner points lie on lines that are predominantly vertical. There should be (ni+1)*(nj+1)
such lines, and since the definition of each line requires the specification of two points, each
itself requiring the specification of three coordinates, *COORD expects to read 2 * 3 *
(ni + 1) * (nj + 1) values as described in the following algorithm.
Algorithm for *COORD Ordering:
Do the following for J = 1, ..., (nj + 1): [
Do the following for I = 1, ..., (ni + 1): [
Firstly, ...
If I and J are less than ni and nj respectively, write the "x"-, "y"-, "z"-coordinates of
a point that lies on a vertical line through the NW corner of block (I,J,1). This could
be the "-B" or "-T" corner, or block (I,J,K)'s corner for any K, as all these points
should be collinear.
If I = ni and J is less than nj, write the NE corner. If I is less than ni and J = nj, write
the SW corner. If I = ni and J = nj, write the SE corner.
User's Guide IMEX
Secondly, ...
If I and J are less than ni and nj respectively, write the "x"-, "y"-, "z"-coordinates of
another point that lies on a vertical line through the NW corner of block (I,J,1).
This point should differ from the previous one only in its "z"- coordinate.
If I = ni and J is less than nj, write the NE corner. If I is less than ni and J = nj,
write the SW corner. If I = ni and J = nj, write the SE corner.
] ]
Note that I is ranging fastest, J slowest in the above.
As *COORD data only provides lines on which corner points must lie, *ZCORN array data is
still required to locate the corner points along the lines.
Examples:
Provide *COORD data for a ni = 4, nj = 2, nk = 1 *CORNER grid. Note that the "x"direction grid spacing is uniformly 100 units and the "y"-direction grid spacing is uniformly
200 units. (This example appears the same regardless of the value for nk.)
*COORD
0
200
400
0 0
0 0
0 0
0
200
400
0
0
0
1
1
1
100
300
0
0
0
0
100
300
0
0
1
1
0
200
400
200 0
200 0
200 0
0
200
400
200
200
200
1
1
1
100
300
200
200
0
0
100
300
200
200
1
1
0
200
400
400 0
400 0
400 0
0
200
400
400
400
400
1
1
1
100
300
400
400
0
0
100
300
400
400
1
1
The acceptable range of values for corner point coordinates is:
min
max
User's Guide IMEX
SI
m
-1.0E+20
1.0E+20
Field
ft
-3.28E+20
3.28E+20
Lab
cm
-1.0E+22
1.0E+22
Complete Corner Point Locations for Corner Point Grids

(Conditional)
*CORNERS
PURPOSE:
*CORNERS signals input of a complete array of corner point locations for corner point grids.
ARRAY:
*CORNERS
DEFAULTS:
CONDITIONS:
with corner point grids (*GRID *CORNER). This keyword should not be combined with any
other array-based corner point keywords. This keyword provides a complete array of all
coordinate values required for all the corner points.
EXPLANATION:
See the general corner point discussion given with the *GRID *CORNER keyword for
This keyword causes the processing of 3*(8*ni*nj*nk) values, with the first group of
8*ni*nj*nk values giving all the "x"-coordinates of all corner points, the second group giving
all the "y"-coordinates, and the third group giving all the "z"-coordinates. Each group uses
the same corner point ordering (as presented below), which is also the ordering used by the
*ZCORN keyword. Only the choice of coordinate direction changes from group to group.
Note that the third group of 8*ni*nj*nk values is the same array that would be input using the
*ZCORN keyword.
Algorithm for *CORNERS Ordering:
Do the following three times with:
1) "values" replaced by ""x"-coordinate values";
2) "values" replaced by ""y"-coordinate values";
3) "values" replaced by ""z"-coordinate values", the latter also being the depths:
[
User's Guide IMEX
Do the following for K = 1, ..., nk: [

Write NW-T and NE-T values for block ( 1,J,K)
...
Write NW-T and NE-T values for block (ni,J,K)
Write SW-T and SE-T values for block ( 1,J,K)
...
Write SW-T and SE-T values for block (ni,J,K)
]]
Write NW-B and NE-B values for block ( 1,J,K)
...
Write NW-B and NE-B values for block (ni,J,K)
Write SW-B and SE-B values for block ( 1,J,K)
...
Write SW-B and SE-B values for block (ni,J,K)
]]]
This technique for corner point input will exhibit duplication in the first two groups of
8*ni*nj*nk values, due to the fact that corner points must lie on vertical lines.
Examples:
Provide *CORNERS data for a ni = 4, nj = 2, nk = 1 grid in a reservoir dipping in the "x"coordinate direction. Note that the single layer is 10 units thick and that the "x"-direction
grid spacing is 100 units and the "y"-direction grid spacing is 200 units.
*CORNERS
0
100
0
100
0
100
0
100
0
100
0
100
0
100
0
100
0
0
200
200
200
200
400
400
0
0
200
200
200
200
400
400
2000
2000
2000
2000
2010
2010
2010
2010
User's Guide IMEX
2001
2001
2001
2001
2011
2011
2011
2011
100
100
100
100
100
100
100
100
0
200
200
400
0
200
200
400
200
200
200
200
200
200
200
200
0
200
200
400
0
200
200
400
200
200
200
200
200
200
200
200
0
200
200
400
0
200
200
400
300
300
300
300
300
300
300
300
0
200
200
400
0
200
200
400
300
300
300
300
300
300
300
300
0
200
200
400
0
200
200
400
400
400
400
400
400
400
400
400
0
200
200
400
0
200
200
400
2001
2001
2001
2001
2011
2011
2011
2011
2002
2002
2002
2002
2012
2012
2012
2012
2002
2002
2002
2002
2012
2012
2012
2012
2003
2003
2003
2003
2013
2013
2013
2013
2003
2003
2003
2003
2013
2013
2013
2013
2004
2004
2004
2004
2014
2014
2014
2014
The acceptable range of values for corner points are:
min
max
SI
m
-1.0E+20
1.0E+20
Field
ft
-3.28E+20
3.28E+20
Lab
cm
-1.0E+22
1.0E+22
User's Guide IMEX
Local Refined Grid (Optional)
*REFINE
PURPOSE:
*REFINE defines specifications for local grid refinement.
(Note that refinement means that certain blocks in a range on the fundamental (main) grid,
this range being described by the *RANGE keyword following, will each be replaced by a
refined grid. Each refined grid will be made up of several small blocks that will together fill
the space occupied previously by a parent fundamental block.)
The *REFINE keyword can be used with *GRID *CART, *GRID *VARI or *GRID
*CORNER but not with *GRID *RADIAL.
FORMAT:
*REFINE
*HYBRID
nir njr nkr (*IDIR | *JDIR | *KDIR)

*RW rw
(*ALPHAI alphai)
(*ALPHA alpha)
-or*REFINE nir njr nkr

-or*REFINE i1:i2 j1:j2 k1:k2 into nir njr nkr
-or*REFINE i1 j1 k1 / i2 j2 k2 { ... { / in jn kn } } into nir njr nkr
-orREFINE i1:i2 j1:j2 k1:k2
into nir njr nkr *HYBRID (*IDIR | *JDIR |

*KDIR)
*RW rw (*ALPHAI alphai) (*ALPHA
alpha)
-or*REFINE i1 j1 k1 / i2 j2 k2
{ ... { / in jn kn } }
User's Guide IMEX
into nir njr nkr *HYBRID (*IDIR | *JDIR |

*KDIR)
*RW rw
(*ALPHAI alphai)
(*ALPHA alpha)
DEFINITIONS:
*HYBRID
Indicates the use of a hybrid refined grid. This is a refined grid made up of
stacked rings with optional angular subdivisions, much like a radial- angular
cylindrical grid. This refinement is associated with a wellbore of radius "rw"
(as specified by the *RW subkeyword following) assumed to be running
through the centres of the innermost rings. The wellbore's volume is not
included in any of the simulator's grid blocks.
*HYBRID rings will be circular if the parent block's permeabilities
perpendicular to the well are the same (isotropic case). Otherwise, the rings
will have ellipses for boundaries with aspect ratios related to the ratio of
parent block permeabilities. If *HYBRID does not appear, the refined grid
will be made up of rectangular blocks.
nir
If *HYBRID is specified, then nir is the number of rings to be used. The
rings will have curved sides (circles or ellipses) with ring number 1 always
having a circular inner boundary of radius "rw". The last ring, ring number
nir, will be curved on its inner boundary, but its outer boundary will be made
up of four flat sides that match up with neighbouring fundamental grid
blocks or other refined grids. These flat sides correspond to the four sides of
the parent grid block that lie in the directions perpendicular to the well.
If the refinement is rectangular, than nir is the number of grid blocks to be
inserted in the I direction in the refinement.
njr
If *HYBRID is specified, njr must be 1 or 4. If njr is equal to 4, each ring
will be broken up into four pieces, each associated with 90 degrees of
angular extent as measured in the appropriate angular coordinate. Note that
ring number 1 is never quartered however, not even when njr is 4. If njr is 1,
the rings are not subdivided.
If the refinement is rectangular, than njr is the number of grid blocks to be
inserted in the J direction in the refinement.
nkr
If *HYBRID is specified, then nkr is the number of layers used in the stacks
of rings; that is, nkr is the number of layers used in the direction along the
wellbore. If the refinement is rectangular, then nkr is the number of grid
blocks to be inserted in the K direction in the refinement.
*IDIR
The wellbore is aligned with the I direction on the fundamental grid. Thus,
the radial/elliptical rings used with the *HYBRID refined grid will be set up
perpendicular to the I direction.
User's Guide IMEX
*JDIR
The wellbore is aligned with the J direction on the fundamental grid. Thus,
perpendicular to the J direction.
*KDIR
The wellbore is aligned with the K direction on the fundamental grid. Thus,
perpendicular to the K direction.
*RW
Indicates input of a well radius. Required for *REFINE *HYBRID.
rw
Radius of the wellbore (m | ft). Should be nonzero.
*ALPHAI
Used only for isotropic *HYBRID cases, this parameter will be defaulted if it
does not appear. The *ALPHAI parameter defines the ratio of the outer
radius of the first ring to "rw" for *HYBRID refined grids.
*ALPHA
Used only for isotropic *HYBRID cases, this parameter will be defaulted if it
does not appear. The *ALPHA parameter defines the ratio of successive
outer radii for rings i to i-1, for each i running from 2 through nir-1, for
*HYBRID refined grids. (The condition is not applicable to ring nir, as it is
truncated to have four planar sides so as to fit its neighbours.)
alphai
Value for *ALPHAI. Should exceed 1.
alpha
Value for *ALPHA. Should exceed 1.
i1:i2
Indicates the I beginning and ending indices of the fundamental grid region
which will be assigned refinements.
j1:j2
Indicates the J beginning and ending indices of the fundamental grid region
k1:k2
Indicates the K beginning and ending indices of the fundamental grid region
User's Guide IMEX
i1 j1 k1
Indicates the I, J, K location in the fundamental grid region which will be
assigned refinements.
i2 j2 k2
Indicates the I, J, K location in the first level of refinement of the (i1,j1,k1)
fundamental grid region which will be assigned refinements.
in jn kn
Indicates the I, J, K location in the (n-1) level of refinement of the (i1,j1,k1)
DEFAULTS:
Optional keyword. The default is no refined grids.
If *REFINE *HYBRID appears with no values, the default for nir is 2 (two rings), the default
for njr is 1 (no angular subdivisions), and the default for nkr is 1 (no layering).
If *REFINE appears with no values and no *HYBRID subkeyword, the default dimensions
are nir = njr = nkr = 3, meaning 3 rectangular subdivisions are used in each direction.
The default for "rw" is 3 inches (.25 ft or .0762 m) if *RW does not appear, or if *RW 0.0
appears.
The wellbore direction will default to *KDIR (vertical) if no *IDIR, *JDIR, or *KDIR
appears for *REFINE *HYBRID.
The default value for "alphai" is chosen so that the cross-sectional area contained within the
outer boundary of the first ring(s) is (1/(nir+1))**2 of the total area available, the latter being
the cross-sectional area of the parent block being replaced.
The default value for "alpha" is chosen so that if the outer most ring (ring nir) was allowed to
be circular with an outer radius equal to that of ring nir-1 multiplied by "alpha" (so that it was
treated like the other rings), its area would equal the total area available. Of course, as noted
above, the outer most ring is truncated to have flat sides so that neighbouring blocks can be
properly fitted.
For anisotropic media, the values for *ALPHAI and *ALPHA are calculated internally using
much the same criteria as discussed above for the isotropic case, except that elliptical
geometries are used.
CONDITIONS:
The *REFINE keyword may be located in the Reservoir Description keyword group. It may
also occur as part of recurrent data. Thus refined blocks may appear later in the simulation.
Any well perf's which use the refined blocks or any of coarser blocks which contain them,
must be defined with *REFINE before they are perfed with PERFRG.
*RANGE keyword(s) should follow immediately after the *REFINE keywords.
Note that neighbouring *HYBRID grids must share the same well direction, and if the well
through one *HYBRID grid is to be continued into another *HYBRID grid, both grids must
have the same nir and njr. For multilevel refinement, only the finest grid may be hybrid.
User's Guide IMEX
The *REFINE keyword can be used with *GRID *CART, *GRID *VARI or *GRID
*CORNER but not with *GRID *RADIAL.
Use *PERFV with *REFINE *HYBRID.
EXPLANATION:
Regular Refinement
A *REFINE keyword, with accompanying subkeywords, and *RANGE keyword(s), indicate
that a local refined grid option is being used. *REFINE and its parameters determine the
amount of refinement, and *RANGE indicates where the refinement is to take place. (The
*RANGE keyword is described later.) Refinements are Cartesian (rectangular) unless
*HYBRID is specified.
*REFINE i1:i2 j1:j2 k1:k2 INTO (nir njr nkr) or *REFINE i1 j1 k1 / i2 j2 k2 { ... {/ in jn kn}
} INTO (nir njr nkr) specifies the starting and ending indices for a group of fundamental grid
blocks or a single multilevel refinement block which are to be refined.
In addition, the *REFINE INTO option allows the user to specify the refined grid size
on the same keyword the range is defined. *REFINE INTO is therefore a combination of
both the standard *REFINE and *RANGE keywords.
Each grid is assigned a number by the simulator. Grids are numbered starting with the
fundamental grid which is grid number 1, followed by grids created by **REFINE INTO
keywords. Each *REFINE INTO keyword creates grids by running through the parent
blocks (I,J,K) with I running fastest from i1 to i1, j next fastest, and K running slowest from
k1 to k2. Thus, each *RANGE keyword creates (i2-i1+1) * (j2-j1+1) * (k2-k1+1) new grids
which are numbered in succession.
Example:
The following describes 4 grids in total:
*GRID *VARI 5 5 1
...
...
*REFINE 1:2 2 1 INTO 2 2 3
*REFINE 2 2 1 / 2 2 2 INTO 3 3 1
...
...
Grid 1 is the fundamental grid, another grid has parent (1,2,1), another has parent (2,2,1) and
the last grid has parent refined grid (2,2,2) in parent (2,2,1).
Example:
A grid system consists of 4 blocks in the I direction, 4 blocks in the J direction and 2 blocks
in the K direction. Two columns of grid blocks are to be refined, and each of these blocks is
to contain a 2 x 2 x 2 grid. The refined grid blocks (1,1,1) and (1,1,2) in the fundamental grid
blocks (1,1,1) and (1,1,2)are to be further refined into 2 x 2 x 1 grids.
User's Guide IMEX
The data may appear as follows:

*GRID *CART 4
:
*REFINE 2 2 2
*RANGE 1 1
*RANGE 4 4
*REFINE 2 2 1
*RANGE 1 1
*RANGE 1 1
*RANGE 1 1
*RANGE 1 1
4 2
1:2
1:2
1
1
2
2
/
/
/
/
1
1
1
1
1
1
1
1
1
2
1
2
An I-J view of the grid would be:

J=4
J=3
J=2
J=1
I=1
I=2
I=3
I=4
I=3
I=4
A cross-section for J=1 would be:

K=2
K=1
I=1
I=2
The acceptable range of values for well radius is:
min
max
SI
m
0.0
1.0
Field
ft
0.0
3.280
Lab
cm
0.0
3.280
The acceptable range of values for the *ALPHA and *ALPHAI ratios are:
min
max
SI
m
1.0
1.0E+20
Field
ft
1.0
3.28E+20
Lab
cm
1.0
1.0E+22
User's Guide IMEX
Hybrid Refinement
The hybrid grid option refers to refining a parent block from a Cartesian grid into a local
cylindrical grid whose "axial" direction may be oriented in either the global I, J or K
direction. There are nr divisions in the radial direction, of which the outermost is formed to
fit the shape of the parent block. The hybrid's angular direction is divided into either 1 or 4
divisions; the innermost radial division is always a full circle. The hybrid's axial direction is
divided into nz uniform sections. It is anticipated, but not required, that a well will be placed
inside the innermost radial division.
The *HYBRID option may affect results especially when a process is influenced strongly by
near-wellbore phenomena, e.g., cyclic steam stimulation. The well can be horizontal or
vertical. The wellbore and the corresponding hybrid grid axis must go through the centre of
the grid block. Thus, the hybrid grid's local "Z-axis" may be in the global I ,J or K direction
depending on the well direction.
Perpendicular to this axis the permeability may be equal (isotropic case) or not equal
(anisotropic case). The aspect ratio of grid dimensions normal to the axial direction should
not be too different from the square root of the corresponding absolute permeability ratio.
Deviations of more than a factor of 1.25 can lead to large errors and so is not allowed.
ISOTROPIC CASE:
This is normally the case for a hybrid grid whose axis is in the vertical direction. The grid
dimensions must be within a factor of 1.25 of square.
ANISOTROPIC CASE:
This is normally the case for a hybrid grid whose axis is in a horizontal direction. The aspect
ratio of block size normal to the axial direction should be within a factor of 1.5 to 2 of the
square root of the ratio of the corresponding absolute permeabilities. For example, a well
horizontal in the X-direction with Ky = 10 Kz should have a grid aspect ratio of about
delta_Y/delta_Z = square_root(10).
Two rules apply to adjacent parent blocks refined with *HYBRID:
1. For hybrid grids adjacent in the hybrid's axial direction, nr and ntheta must be the
same, and nz may be different. This case is typical for modelling a hybrid grid
around a well that passes through more than one parent block.
2. For hybrid grids adjacent in a direction other than the hybrid's axial direction, only
nz must be the same.
Hybrid Grid Orientations
Normally, the user will need to know only which of *IDIR, *JDIR or *KDIR was specified in
order to interpret the position of the individual hybrid grid blocks relative to the surrounding
fundamental blocks. However, the precise meaning of the hybrid grid's local J and K indices
is needed in order to assign non-uniform properties and conditions, and to interpret in detail
the textual output.
In the following, x, y, z, I, J and K refer to the fundamental grid and similar primed (')
symbols refer to the local cylindrical grid.
In each of the following orientation cases, both a "GLOBAL VIEW" and a "LOCAL VIEW"
are shown. In the "LOCAL VIEW" the point-of-view is on the hybrid z' axis looking in the
User's Guide IMEX
negative z' direction, i.e., z' points toward the viewer. Note that the only difference between
the "LOCAL VIEW" of the cases is the relation to the global coordinates.
Well in I Direction (*IDIR): x = xo + z',
y = yo + y',
GLOBAL VIEW
k'=1
z = zo x'
LOCAL VIEW
k'=2
J+1
K+1
K-1
J'=2
y'
J
y
x'
J'=1
J'=3
z'
J'=4
x
J-1
Adjacent block is
connected to
(I,J,K-1)
J'=1
(I,J+1,K)
J'=2
Well in J Direction (*JDIR): x = xo + x',
(I,J,K+1)
J'=3
y = yo + z',
(I,J-1,K)
J'=4
z = zo y'
LOCAL VIEW
GLOBAL VIEW
I-1
K-1
I+1
k'=2
k'=1
J'=2
y'
J'=1
x'
J'=3
y
z'
J'=4
K+1
x
(I-1,J,K)
J'=3
(I,J,K-1)
J'=2
(I+1,J,K)
J'=1
Adjacent block is
connected to
Well in K Direction (*KDIR): x = xo + x',
y = yo + y',
GLOBAL VIEW
(I,J,K+1)
J'=4
z = zo + z'
LOCAL VIEW
I-1
J+1
I+1
k=3
y
y'
k=2
z
k=1
J'=1
x'
J'=3
x
J'=2
J
z'
J'=4
z
J-1
x
Adjacent block is
connected to
(I+1,J,K)
J'=1
(I,J+1,K)
J'=2
(I-1,J,K)
J'=3
(I,J-1,K)
J'=4
User's Guide IMEX
Direction Dependent Data

The specification of direction-dependent data differs slightly from fundamental or regular
refined grids. In the data entry keywords there is no explicit way to refer to a hybrid grid's
local directions. For example, for entering permeability there are PERMI, PERMJ and
PERMK but nothing explicitly for R, Theta and Z. The method used to refer to the hybrid
grid direction is as follows.
For each orientation described above, there is correspondence between the I,J,K direction
labels and the hybrid grid's local radial, angular and axial directions.
Orientation
*IDIR
*JDIR
*KDIR
Radial
K
I
I
Angular
J
K
J
Axial
I
J
K
Take the *IDIR orientation for example. You would use PERMK to modify permeability in
the hybrid grid's local radial direction, PERMJ to modify the angular direction and PERMI to
modify the axial direction. You would examine K Direction Block Size to find the block size
in the hybrid grid's local radial direction, J Direction to find the angular size and I Direction
to find the axial size.
This correspondence is reported in the textual output file in the grid summary section, for
each hybrid grid. It applies to all direction dependent input (except block size)
-
permeabilities
transmissibility multipliers
block sizes
permeabilities
transmissibility multipliers
transmissibilities
and output
Note that connection-based quantities such as transmissibility have an explicit Radial and
Angular direction printout, but the Axial direction values are found through this direction
correspondence.
Block Sizes
The block sizes reported for the hybrid grid blocks are similar to those found for a cylindrical
grid system. The radial and axial block sizes have standard definitions. The angular
direction block size is the midpoint arc length, so that the product of the block sizes in the
three directions is equal to the block volume (without volume modifier).
The only exception to this is the outermost radial block which acts as an interface between
the radial grid and the surrounding Cartesian grid. The angular block size is the parent
block's size in the corresponding direction instead of the midpoint arc length. The radial
block size is an average value which gives the block volume (without volume modifier).
For the case with no angular subdivisions (ntheta = 1) the radial block size is based not on the
entire outermost block volume, but the fraction associated with the outer face of interest.
This fraction of the total volume is the same as the block's volume for the ntheta = 4 case.
User's Guide IMEX
Example: Hybrid refinement where nr = 2 and ntheta = 1

*REFINE 1 1 1 *INTO 2 1 1 *HYBRID *KDIR
Example: nr = 3 and ntheta = 4, with axial direction in the X-direction

*REFINE 1 1 1 *INTO 3 4 1 *HYBRID *IDIR
User's Guide IMEX
Refined Grid Location (Conditional)
*RANGE
PURPOSE:
*RANGE specifies the input of location(s) where the refinement(s) defined by the previous
*REFINE keyword are to act. (*RANGE is not required if *INTO was used with the
*REFINE keyword.)
FORMAT:
*RANGE i1:i2 j1:j2 k1:k2
-or*RANGE i1 j1 k1 / i2 j2 k2 { ... { / in jn kn } }
DEFINITIONS:
i1:i2
j1:j2
k1:k2
i1 j1 k1
Indicates the I, J, K location in the fundamental grid region which will be
assigned refinements.
i2 j2 k2
Indicates the I, J, K location in the first level of refinement of the (i1,j1,k1)
in jn kn
Indicates the I, J, K location in the (n-1) level of refinement of the (i1,j1,k1)
DEFAULTS:
No defaults.
CONDITIONS:
This keyword must be in the Reservoir Description keyword group. This offers an older,
alternate form for the *INTO syntax of the *REFINE keyword. One or more *RANGE
keyword(s) must follow a *REFINE keyword, while *RANGE keywords can only follow
*REFINE or other *RANGE keywords.
User's Guide IMEX
EXPLANATION:
*RANGE specifies the starting and ending indices for a group of fundamental grid blocks or
a single multilevel refinement block which are to be refined. Further discussions and
examples are provided with the description of the *REFINE keyword.
Each grid is assigned a number by the simulator. Grids are numbered starting with the
fundamental grid which is grid number 1, followed by grids created by *RANGE keywords.
Each *RANGE keyword creates grids by running through the parent blocks (I,J,K) with I
running fastest from i1 to i1, j next fastest, and K running slowest from k1 to k2. Thus, each
*RANGE keyword creates (i2-i1+1) * (j2-j1+1) * (k2-k1+1) new grids which are numbered
in succession.
Example:
The following describes 4 grids in total:
*GRID *VARI
:
*REFINE 2 2
*RANGE 1:2
*REFINE 3 3
*RANGE 2 2
:
5 5 1
3
2 1
1
1 / 2 2 2
Grid 1 is the fundamental grid, another grid has parent (1,2,1), another has parent (2,2,1) and
the last grid has parent refined grid (2,2,2) in parent (2,2,1).
User's Guide IMEX
Dual Porosity (Optional)
*DUALPOR
PURPOSE:
*DUALPOR indicates the use of a dual porosity model related to the Warren and Root model
in some or all of the simulator's grid blocks.
FORMAT:
*DUALPOR
DEFAULTS:
Optional keyword. No default.
CONDITIONS:
This keyword must be located in the Reservoir Description keyword group, before the
*REFINE and *RANGE keywords.
Only one of *DUALPOR, *DUALPERM, *SUBDOMAIN or *MINC may be specified.
EXPLANATION:
This keyword indicates that a Warren and Root style dual porosity option be used in the
simulator. This option allows each simulator block to have up to two porosity systems, one
called the matrix porosity and the other called the fracture porosity. Each porosity can have
its own porosity value and its own permeabilities, as well as other distinct properties. Matrix
properties are described using the *MATRIX keyword while fracture properties are described
using the *FRACTURE keyword.
In the Warren and Root model, inter-block flows are calculated in much the same manner as
they would be in the standard (no *DUALPOR keyword) model. These flows are governed
by the fracture properties. However, an additional set of matrix-fracture flows are calculated
when *DUALPOR is specified. These flows are governed by the matrix properties. The
calculation of this matrix-fracture fluid transfer is described when the *SHAPE keyword is
defined.
Thus, *DUALPOR allows one matrix porosity and one fracture porosity per grid block,
where the matrix is connected only to the fracture in the same grid block. Fracture porosities
are connected to other neighbouring fracture porosities in the usual manner. The presence of
both fracture and matrix porosities in a block, or just a fracture porosity or a matrix porosity,
is under user control (see the *POR and *NULL keywords). Of course, if *DUALPOR was
set but no blocks were assigned a matrix porosity, results like that of the standard simulator
(no *DUALPOR) would be obtained.
Property definition for *DUALPOR systems usually requires the use of pairs of definitions
for most items, one carrying a *MATRIX identifier and the other a *FRACTURE identifier.
Further details are explained in the descriptions for the individual properties.
See J. E. Warren and P. J. Root, "The Behaviour of Naturally Fractured Reservoirs", SPEJ,
Sept. 1963, pp. 245-255 (Trans. AIME, 234) for further details.
User's Guide IMEX
Dual Permeability (Optional)
*DUALPERM
PURPOSE:
*DUALPERM indicates the use of a dual porosity model related to the Warren and Root
model in some, or all, of the simulator's grid blocks. Moreover, inter-block fracture to
fracture flows are augmented by inter-block matrix to matrix flows. The matrix to fracture
flows within blocks remain.
FORMAT:
*DUALPERM
DEFAULTS:
CONDITIONS:
Only one of *DUALPOR, *DUALPERM, *SUBDOMAIN, or *MINC may be specified.
EXPLANATION:
The description given above for the *DUALPOR keyword should be studied first, as this
option is closely related.
The *DUALPERM option uses the same calculations as the *DUALPOR option, except that
inter-block matrix to matrix flows are also calculated in addition to the expected inter-block
fracture to fracture flows and the matrix to fracture fluid transfer within blocks.
Thus, *DUALPERM allows one matrix porosity and one fracture porosity per grid block,
where the matrix is connected to the fracture in the same grid block. Fracture porosities are
connected to neighbouring fracture porosities, and the same holds true for neighbouring
matrix porosities.
Property definition for *DUALPERM systems usually requires the use of pairs of definitions
for most items, one carrying a *MATRIX identifier and the other a *FRACTURE identifier.
Further details are explained in the descriptions for the individual properties.
Dual permeability is often important in reservoirs with free gas and large variations in depth
for which only the vertical (K direction) matrix to matrix inter-block flows are important. If
this is so, use zero transmissibility modifiers in the I and J directions (see the *TRANSI and
*TRANSJ keywords following).
User's Guide IMEX
Dual Porosity Subdomain Method (Optional)
*SUBDOMAIN
PURPOSE:
*SUBDOMAIN indicates the use of a dual porosity model using the subdomain method.
FORMAT:
*SUBDOMAIN idiv
DEFINITIONS:
idiv
Number of subdivisions for each matrix blocks, typically chosen from the
range of 2 to 5.
DEFAULTS:
CONDITIONS:
*REFINE and *RANGE keywords. Only one of *DUALPOR, *DUALPERM,
*SUBDOMAIN, or *MINC may be specified.
The use of grid refinement is supported, but if the grid is refined vertically and the refined
block thickness is less than *DKFRAC, *DKFRAC will be set to the refined block thickness.
This will result in an altered capillary holdup for the refined region. The user should be aware
of how vertical refinement will alter *DKFRAC and capillary holdup.
When using the Subdomain option, the Grid Module which initializes grid block properties
and interblock transmissibilities must not be called from Recurrent Data. This prevents the
user from setting transmissibility multipliers in recurrent data and defining any grid
refinements/modifications in recurrent data.
EXPLANATION:
This option allows each simulator block to have up to two porosity systems, one called the
matrix porosity and the other called the fracture porosity. Moreover, the *SUBDOMAIN
option splits up the matrix porosity vertically into "idiv" equal thickness pieces. Inter-block
fracture to fracture, and matrix to fracture flows within a block, are calculated. Also, matrix
to matrix flow between the matrix layers within a block are calculated.
The *SUBDOMAIN method allows the modelling of segregating flows within the matrix
porosity of a block.
Each block will have its own porosity value and its own permeabilities, as well as other
distinct properties. Matrix properties are described using the *MATRIX keyword while
fracture properties are described using the *FRACTURE keyword. Property definition
usually requires the use of pairs of definitions for most items, one carrying a *MATRIX
identifier and the other a *FRACTURE identifier. Further details are explained in the
descriptions for the various properties.
User's Guide IMEX
Within a subdomain block, *DKFRAC (not block thickness) is used to calculate capillary
holdup. *DKFRAC may be less than the block thickness, but can be no greater than block
thickness. If *DKFRAC greater than the thickness of a Subdomain block is encountered
*DKFRAC for that block will be set equal to the block thickness.
For details of this method, please see J. R. Gilman, "An Efficient Finite-Difference Method
for Simulating Phase Segregation in the Matrix Blocks in Dual-Porosity Reservoirs", SPERE,
July 1986, pp. 403-413.
User's Guide IMEX
Assigning Fractional Volumes to Subdomain blocks

(Optional)
*FRACVOL
PURPOSE:
By default the *SUBDOMAIN idiv option generates subdomains of thickness Spatial
block thickness/idiv. Similarly the *MINC idiv option generates nested rings of equal
volume.
FRACVOL allow the user to alter the size of individual subdomain blocks (*SUBDOMAIN)
or nested rings (*MINC) relative to each other. This is used to improve subdomain capillary
holdup calculations and is crucial in the proper modeling of reinfiltration.
FORMAT:
*FRACVOL frac_1 frac_2 frac_idiv
DEFINITIONS:
Frac_1
Fractional volume of first subdomain block / outermost ring block
Frac_2
Fractional volume of second subdomain block / second ring block
Frac_idiv
Fractional volume of last subdomain block / last ring block
DEFAULTS:
Optional keyword. No default. If omitted subdomain volumes / ring volumes are all set
equal to 1.0/idiv
CONDITIONS:
This keyword must be located in the Reservoir Description keyword group. *SUBDOMAIN
or *MINC must be specified. Values of *FRACVOL as low as 0.01 have been used without
difficulty.
EXPLANATION:
It allows the user to adjust the volume/size of stacked subdomain or nested ring blocks
relative to each other according to *FRACVOL values.
In order to obtain the correct capillary holdup when using the Subdomain model it is optimal
to give the first and last division the smallest possible fractional volume which does not
significantly influence CPU time. In the case of a very small number of subdomain divisions
when *TRANSD > 0.0 and *SD_REINF > 0.0 it may be sufficient to reduce the size of the
last subdomain division only. See Keywords *TRANSD and *SD_REINF.
Examples:
The following provides examples of *FRACVOL usage:
User's Guide IMEX
*SUBDOMAIN 5
*FRACVOL 0.02 0.32 0.32 0.32 0.02
** or if the number of subdomains is limited and
** TRANSD > 0.0 and SD_REINF > 0.0
*SUBDOMAIN 4
*FRACVOL 0.33 0.33 0.32 0.02
User's Guide IMEX
Subdomain to Subdomain Flow Transmissibility Multipliers

(Optional)
*TRANSD
PURPOSE:
*TRANSD indicates input of an array of Inter-Subdomain (vertical) transmissibility
multipliers and enables the subdomain to subdomain vertical matrix connection.
ARRAY:
*TRANSD
DEFAULTS:
Optional keyword. Defaults: 0.0
CONDITIONS:
This keyword may only be in the Reservoir Description keyword group and is only valid
when the *SUBDOMAIN dual porosity option is in effect.
EXPLANATION:
In the standard Subdomain dual porosity model, the fracture and matrix are modeled as two
separate grid systems. For a single spatial block, the fracture is modeled using a single grid
block while the matrix is modeled using a vertical stack of grid blocks.
Within a single spatial block, the matrix stack can only develop fluid flow within the stack
and between each block of the stack and the fracture block.
The use of the TRANSD array allows the matrix stacks in each subdomain to be connected
vertically together by an additional vertical connection per spatial block. The last matrix
block in a subdomain is connected to the first matrix block in the block below it. This allows
gravity drainage to occur directly from subdomain to subdomain.
There is capillary continuity between subdomains unless *SD_REINF is used. See keywords
FRACVOL and SD_REINF.
Examples:
The following provides examples of TRANSD usage:
** Turn on direct gravity drainage between subdomains

** for all blocks
*TRANSD *CON 1.0
** Turn on direct gravity drainage between block 2,3,6
** and 2,3,7 only
*TRANSD *IJK 2:2 3:3 6:6 1.0
The suggested range of values for transmissibility multipliers is:

min
max
User's Guide IMEX
SI
0.0
1000.0
Field
0.0
1000.0
Lab
0.0
1000.0
Dual Porosity MINC Method (Optional)
*MINC
PURPOSE:
*MINC indicates the use of a dual porosity model using the multiple-interacting-continua
(MINC) approach.
FORMAT:
*MINC idiv
DEFINITION:
idiv
Number of subdivisions for each matrix blocks, typically chosen from the
range of 2 to 5.
DEFAULT:
CONDITIONS:
Only one of *DUALPOR, *DUALPERM, *SUBDOMAIN, or *MINC may be specified.
EXPLANATION:
This option allows each simulator block to have up to two porosity systems, one called the
matrix porosity and the other called the fracture porosity. Moreover, the *MINC option splits
up the matrix porosity into "idiv" nested rings. Inter-block fracture to fracture, and matrix to
fracture flows within a block, are calculated. Also, matrix to matrix flow between the matrix
rings within a block are calculated.
The *MINC method allows the modelling of some transient behavior within the matrix
porosity of a block.
Each porosity will have its own porosity value and its own permeabilities, as well as other
distinct properties. Matrix properties are described using the *MATRIX keyword while
fracture properties are described using the *FRACTURE keyword. Property definition
usually requires the use of pairs of definitions for most items, one carrying a *MATRIX
identifier and the other a *FRACTURE identifier. Further details are explained in the
descriptions for the various properties.
For details of this method, please refer to K. Pruess and T. N. Narasimhan, "A Practical
Method for Modelling Fluid and Heat Flow in Fractured Porous Media", SPEJ, Feb. 1985,
pp. 14-26.
User's Guide IMEX
Shape Factor Calculation (Conditional)
*SHAPE
PURPOSE:
*SHAPE describes which type of shape factor is to be used in calculating matrix to fracture
flows for various dual porosity models.
FORMAT:
*SHAPE
(*GK)
(*WR)
DEFINITIONS:
*GK
This subkeyword indicates the use of a Gilman and Kazemi style formulation
for the shape factor.
*WR
This subkeyword indicates the use of a Warren and Root style formulation
for the shape factor.
DEFAULTS:
Conditional keyword. Default: *SHAPE *GK.
CONDITIONS:
This keyword must be located in the Reservoir Description keyword group.
Only relevant when one of *DUALPOR, *DUALPERM, *SUBDOMAIN, or *MINC is
specified.
EXPLANATION:
As a general rule, fluid flow between small porous regions is proportional to a
transmissibility that consists of a sum of terms that are each a surface area multiplied by a
permeability and divided by a regional dimension perpendicular to the area (assuming the
regions are of similar size). By multiplying the numerator and denominator of such terms by
the regional dimension, the terms become volumes multiplied by permeabilities, which are
divided by the squares of regional dimensions.
If the regions are now taken to be disconnected portions of the matrix porosity in a grid
block, separated by thin fracture planes, the regional dimensions become the fracture
spacings. Assuming that there are many similar such matrix portions connected to fractures
that pervade the grid block, a total transmissibility for matrix to fracture flow within the block
can be obtained by scaling up the disconnected matrix portion volumes to the total matrix
porosity volume. The shape factor to be discussed here concerns what then happens to the
other terms making up the total transmissibility.
User's Guide IMEX
The Warren and Root formulation has matrix to fracture flow within a grid block proportional
to:
W & R Transmissibility = 20/3 * k * (( 1/Lx + 1/Ly + 1/Lz)**2) * M-Vol
whereas the Gilman and Kazemi formulation has:
G & K Transmissibility = 4 * k *(1/(Lx**2) + 1/(Ly**2) + 1/(Lz**2)) * M-Vol
where the above assumes fracture planes exist in all three directions with fracture spacings
Lx, Ly, Lz (which are set using the *DIFRAC, *DJFRAC and *DKFRAC keywords
following), the permeability is k, and the total matrix volume is M-Vol. Further details for
this and other cases are given in the references noted before.
The *SHAPE keyword selects which of the above is to be used. The Gilman and Kazemi
factor (*SHAPE *GK) is recommended due to the straightforward way it incorporates
various anisotropies.
Note that the basic transmissibility formulas are used even when the fracture spacings exceed
the grid block size. These cases correspond to dividing up the matrix to fracture flow over
several grid blocks and are modelled with the usual matrix to fracture flows in the individual
blocks.
User's Guide IMEX
Matrix-Fracture Transfer Calculation (Conditional)
*TRANSFER
PURPOSE:
*TRANSFER indicates the type of matrix-fracture model for treating different phases in
dual-porosity or dual-permeability reservoirs.
FORMAT:
*TRANSFER
ntran
DEFINITIONS:
ntran
The matrix-fracture model number:
= 0; No special treatment
= 1; Complete phase segregation model
DEFAULTS:
Conditional keyword. If the keyword *TRANSFER is not found the default is *TRANSFER
0. If the keyword *TRANSFER is found without an explicitly defined NTRAN, the default is
*TRANSFER 1.
CONDITIONS:
This keyword must be located in the Reservoir Description keyword group. This keyword is
only required for naturally fractured reservoirs as specified by either the *DUALPOR or
*DUALPERM keyword.
EXPLANATION:
Normally, when NTRAN=0, the matrix fracture transfer is calculated following either Warren
and Roots method or Gilman and Kazemi's approach (see the *SHAPE keyword for more
details).
For example, using the *SHAPE *GK option, the matrix-fracture transfer term jmf for phase
j would be calculated as
jmf = V
K rj j
j
(Pjm Pjf )
j = o, w, g
where
j = reservoir condition density of phase j
Pg = Po + Pcog
Pw = Po - Pcow
K
Ky
K
= 4 x +
+ z
L2
L2y
L2z
x
User's Guide IMEX
with V being the grid block volume and Lx, Ly and Lz the fracture spacing (*DXFRAC,
*DYFRAC, *DZFRAC) in the x, y and z directions respectively.
The matrix and fracture blocks are assumed to be at the same depth, thus no gravity term is
included above. This equation is used when NTRAN=0. Generally, the matrix capillary
pressure is much larger than the fracture capillary pressure and the above expression for the
transfer term jmf cannot correctly model the gravity drainage process.
The gravity effects needed to model the gravity drainage process can be approximately
included by using the NTRAN=1 option. This assumes complete gravity segregation of the
oil, water and gas phases as follows:
omf = V
K ro o
( Pom Pof
o
gmf = V
K rg g
Sg
1
Pgm Pgf +

1 S org S wc
2
g
Sg
1
ogm h
ogf h
1 S org S wc
2
wmf = V
K rw w
w
1
S w S wc
Pwm Pwf
2 1 S orw S wc
1
S w S wc
+
2 1 S orw S wc
wom h
m
wof h
where
og
wo
h
=
=
=
(reservoir condition oil density reservoir condition gas density) *g

(reservoir condition water density reservoir condition oil density) *g
height of matrix element in the direction of gravity (h is normally *DZFRAC
unless the block thickness is less than *DZFRAC in which case h block
thickness)
and subscripts
f
m
=
=
fracture property
matrix property
Note that normally m = f and the additional gravity terms just accounts for the difference
in the height of the fluid columns.
User's Guide IMEX
Allowing Addition Matrix-Fracture Connections (Optional)

*DPCONNECT
PURPOSE:
The DPCONNECT keyword allows the grid module to produce extra connection between
blocks in a dual porosity/permeability system when either the matrix or the fracture is
missing from one of the two blocks being connected.
FORMAT:
*DPCONNECT 1
DEFAULTS:
Optional keyword. No default. If omitted normal matrix-fracture connects are made.
CONDITIONS:
This keyword must be located in the Reservoir Description keyword group. *DUALPOR or
*DUAL PERM must be specified.
EXPLANATION:
When enabled, a dual porosity or dual permeability block with a missing fracture will connect
the matrix to both the matrix and fracture of surrounding blocks. In addition a dual porosity
or dual permeability block with a missing matrix will connect the fracture to both the matrix
and fracture of surrounding blocks.
User's Guide IMEX
Fracture Spacing (Conditional)
*DIFRAC, *DJFRAC, *DKFRAC
PURPOSE:
*DIFRAC indicates the input of fracture spacings for the I direction.
*DJFRAC indicates the input of fracture spacings for the J direction.
*DKFRAC indicates the input of fracture spacings for the K direction.
ARRAY:
*DIFRAC
*DJFRAC
*DKFRAC
DEFAULTS:
Conditional keywords. No defaults.
CONDITIONS:
These keywords must be in the Reservoir Description keyword group. These keywords are
required with *DUALPOR, *DUALPERM, *SUBDOMAIN, and *MINC.
Setting value(s) to zero for the *DIFRAC keyword means that there are no fracture planes
perpendicular to the I axis in those blocks, and similarly for the other directions. (Setting
values to 0 corresponds to an infinite, and hence ineffective, spacing in that direction.)
If a block is assigned a zero value for each of *DIFRAC, *DJFRAC, and *DKFRAC, then
the block's fracture porosity will be declared null and will not participate in any simulator
calculations.
EXPLANATION:
Fracture spacings are used to calculate the matrix to fracture transfer coefficient as detailed
above for the *SHAPE keyword (see also Appendix A).
Fracture spacings should be measured from centre line to centre line in the appropriate
direction. The basic transmissibility formulas (see the *SHAPE keyword) are valid even
when the fracture spacings exceed the grid block size, so large fracture spacings should not
be a concern (except that weak matrix to fracture coupling occurs).
*DIFRAC and *DJFRAC fracture spacings are required for *GRID *RADIAL situations.
For such grids, *DIFRAC should be thought of as measuring spacings in the "x"-direction
(corresponding to the 0 degree axis) and *DJFRAC to spacings in the "y"-direction. Spacings
are generally inherited from parent blocks for *HYBRID grids and hence, are automatically
available.
The acceptable range of values for fracture spacing is:
min
max
SI
m
0.0
1.0E+4
Field
ft
0.0
32,808.0
Lab
cm
0.0
1.0E+6
User's Guide IMEX
Null Block Indicator (Optional)
*NULL
PURPOSE:
*NULL indicates the input of an array of null block indicators which can be used to mark
individual porosities as non-participating in dual porosity models, or entire blocks as nonparticipating.
ARRAY:
*NULL
DEFAULTS:
Optional keyword. Default: All blocks are active (participating); no blocks are null.
CONDITIONS:
This keyword must be in the Reservoir Description keyword group.
The numerical values in the incoming array must be zeroes (0) for null or inactive blocks, or
ones (1) for active or participating blocks.
EXPLANATION:
All blocks are considered active unless a block is specifically designated as being null. Null
blocks do not participate in any of the simulator's calculations. Moreover, they are a barrier
to any flow.
The *NULL keyword provides one method for making blocks null; use the following flags in
the *NULL array:
0 = null block
1 = active block.
Other methods are also available for making blocks null; see the *POR keyword, for instance.
Note that if a block is marked null, setting nontrivial properties, including a nonzero porosity,
will NOT cause the block to become active again. These properties will just be ignored. If a
dual porosity model is being used, selective participation of the two porosities can be
controlled with the *NULL keyword. Using *NULL with no *MATRIX/*FRACTURE
qualifier nulls the block (including both porosities) and makes it a barrier to flow. Nulling
the block once with one of *MATRIX or *FRACTURE, and then again with the other
qualifier has the same effect. Using *NULL *MATRIX and setting each of *DIFRAC,
*DJFRAC, and *DKFRAC equal to 0 for a block accomplishes the same task. Using just
*NULL *MATRIX or *NULL *FRACTURE makes only one of the porosities nonparticipating. Flow can occur to the other porosity as required.
Note that a pinched out status set by the *PINCHOUTARRAY keyword over-rides *NULL
settings. See the description of *PINCHOUTARRAY following. *NULL settings over-ride
pinch out setting generated by use of the *PINCHOUT-TOL keyword, or zero thickness
situations.
User's Guide IMEX
Example: Cause certain blocks in a 10 x 10 x 3 system to be null:

*NULL *IJK
1:10
1:10
1:5 10:10
1:4
9:9
1:3
8:8
1:2
7:7
1:1
6:6
1:3
1:3
1:3
1:3
1:3
1:3
1
0
0
0
0
0
User's Guide IMEX
Porosity (Required)
*POR
PURPOSE:
*POR indicates input of porosities, where zero values can be used to mark individual
porosities as null (non-participating) in dual porosity models, or entire blocks as null.
ARRAY:
*POR
DEFAULTS:
Required keyword. No defaults.
CONDITIONS:
EXPLANATION:
The *POR array assigns blocks their porosity values. The fluid (pore) pressure at which
these porosities are to apply is given using the *PRPOR keyword (see following).
Dimensionless fractions are expected (values between 0.0 and 1.0), NOT percentages. Zero
values have special significance in setting blocks null.
When a dual porosity model is not being used, a block becomes null (non-participating) when
it is assigned a zero porosity.
Zero values can mark individual porosities as null (non-participating) in dual porosity
models. This should be done using the *MATRIX and *FRACTURE qualifiers. An entire
block becomes null when both its porosities are null.
Note that in dual porosity models, fracture porosity is the fraction of void space in the
fracture system considered with respect to a volume of the bulk reservoir rock. Matrix
porosity is the fraction of void space in a piece of unfractured matrix material, examined
independently of any fractures.
Example:
A grid with 5 blocks in the I direction, 3 blocks in the J direction, and a single layer is used to
model a dual porosity system. Only those blocks with J=2 are fractured. Block (1,1,1) is not
part of the reservoir. Porosities could be input as:
*POR *MATRIX *IJK
1:5
1:3 1
0.16
1
1
1
0.00
*POR *FRACTURE *IJK
1:5
1:3 1
0.00
1:5
2
1
0.04
The acceptable range of values for porosities is:
min
max
User's Guide IMEX
SI
dimensionless
0.0
1.0
Field
dimensionless
0.0
1.0
Lab
dimensionless
0.0
1.0
Rock Compressibility (Required)
*PRPOR, *CPOR
PURPOSE:
*CPOR signals the input of a rock compressibility value that will be used throughout the
entire model.
*PRPOR signals the input of a reference pressure for the rock compressibility. This pressure
is the fluid (pore) pressure at which the values input using *POR are to hold.
*CPOR and *PRPOR cause modifications to continuously occur in the input *POR porosity
values as fluid pressure in the reservoir rises and falls.
FORMAT:
*CPOR
(*MATRIX)
(*FRACTURE)
(*MATRIX)
(*FRACTURE)
*PRPOR
cpor
ref_pressure
DEFINITIONS:
cpor
Pressure dependence of formation porosity; that is, rock compressibility
(1/kPa | 1/psi | 1/kPa | 1/(kg/cm**2)).
ref_pressure
Reference pressure (kPa | psi | kPa | kg/cm**2) for calculating the effect of
rock compressibility.
DEFAULTS:
*CPOR will default to 0.0 (all units).
*PRPOR will default to 1 atmosphere (101.3 kPa | 14.7 psi | 101.3 kPa | 1 kg/cm**2).
CONDITIONS:
These keywords must be in the Reservoir Description keyword group.
EXPLANATION:
The functional form used in the simulator for the calculation of a block's porosity is:
por(p) = por_input * [1 + cpor * (p - prpor)]
where "por_input" is the porosity input for the block using the *POR keyword, "cpor" and
"prpor" are as input using the *CPOR and *PRPOR keywords, and "p" is the pore pressure
(in the oil phase) for the block.
Note that the porosities entered using the *POR keyword should be associated with, or
measured at, the reference pressure entered using *PRPOR.
When using dual porosity models, both matrix and fracture rock compressibilities and
reference pressures should be input. The *MATRIX and *FRACTURE subkeywords are
used to indicate the matrix and fracture dependence.
User's Guide IMEX
Example:
Porosity values for a dual porosity model were estimated to be near the values .11 and .02, for
matrix and fracture respectively, at atmospheric conditions. The rock compressibilities were
estimated to be 3.0E-6 for both porosities. The following data should be used:
*CPOR *MATRIX
*PRPOR *MATRIX
3.0E-6
14.7
*CPOR *FRACTURE
*PRPOR *FRACTURE
3.0E-6
14.7
*POR *MATRIX
*CON .11
*POR *FRACTURE *CON .04
The suggested range of values for rock compressibility is:
min
max
SI
1/kPa
0.0
1.0E-3
Field
1/psi
0.0
6.89x10-3
Lab
1/kPa
0.0
1.0E-3
Mod. SI
1/kg/cm**2
0.0
1.0E-5
Field
psi
0.0
14505.0
Lab
kPa
0.0
1.0E+5
Mod. SI
kg/cm**2
0.0
1.0E+2
and for reference pressure is:
min
max
SI
kPa
0.0
1.0E+5
Values outside of these ranges are accepted but a warning will be issued by the simulator.
User's Guide IMEX
Pore Volume Modifiers (Optional)
*VOLMOD
PURPOSE:
*VOLMOD indicates input of an array of pore volume modifiers.
ARRAY:
*VOLMOD
DEFAULTS:
Optional keyword. Default: 1.0
In dual porosity models, pore volume modifiers can be applied to the matrix and fracture pore
volumes separately by use of the *MATRIX and *FRACTURE qualifiers. However, the
default for fracture multipliers is still 1.0, even if *VOLMOD *MATRIX (or even just
*VOLMOD) appeared, and the default for matrix multipliers is still 1.0, even if *VOLMOD
*FRACTURE appeared.
It is recommended that *VOLMOD *MATRIX and *VOLMOD *FRACTURE be used
together, or not at all, for dual porosity models.
CONDITIONS:
Values for pore volume modifiers must be non-negative, and they may exceed 1.
EXPLANATION:
Pore volumes are calculated in the simulator based on the reservoir rock bulk volume, which
is calculated from the external grid block dimensions, and the porosity, which depends on
pore-pressure and compressibility data. The pore volumes accumulate fluids in the simulator
and are basic to all its calculations.
Pore volume modifiers are multiplicative and are used to modify the pore volume calculation
given above. These modifiers can be used to adjust pore volumes for a reservoir boundary
that passes through a block, thereby leaving a portion of the grid block out of the reservoir, or
for a flow unit boundary that may be encountered when simulating pattern floods. It is
recommended that pore volume modifiers be used for these situations, rather than adjusting
the porosities directly, so that unusual porosity values are avoided.
Note that multipliers larger than 1 may be used to associate volume external to the reservoir
with a grid block. For instance, total wellbore volume for a gas well can be associated with
the grid block through which the well penetrates by using a multiplier for that block.
Example:
The region on the right side of the following block is not part of the reservoir and constitutes
.4 of the volume of the grid block shown. Input the true average porosity for the portion of
the grid block that lies in the reservoir and assign a block volume multiplier of .6 to the block.
User's Guide IMEX
Reservoir
Portion
POR = 0.1
*POR
*VOLMOD
XXXXXX
XXXXXXX
XXXXXXXX
XXXXXXXXX
XXXXXXXXXX
XXXXXXXXXXX
Block (2, 3, 4)
*IJK ...
2 3 4 0.1
...
*IJK 2 3 4 0.6
Multipliers for the other blocks will default to 1.

The acceptable range of values for pore volume modifiers is:
min
max
User's Guide IMEX
SI
m
0.0
1.0E+4
Field
ft
0.0
32,808.0
Lab
cm
0.0
1.0E+6
Permeabilities (Required)
*PERMI, *PERMJ, *PERMK
PURPOSE:
*PERMI indicates input of an array of I direction permeabilities.
*PERMJ indicates input of an array of J direction permeabilities.
*PERMK indicates input of an array of K direction permeabilities.
ARRAY:
*PERMI
*PERMJ
*PERMK
DEFAULTS:
Required keywords. No defaults.
CONDITIONS:
These keywords must be in the Reservoir Description keyword group. They may also appear
in recurrent (well) data.
EXPLANATION:
Permeabilities are used to compute inter-block fluid flows and must be available for any
direction for which fluid may enter or leave a block. Note that values must be available for
both grid blocks in a communicating pair, as permeability averaging is carried out. Also, a
block's permeabilities are used for both the flows to its high side neighbours as well as its low
side. For instance, if block (I,J,K) has an I direction permeability "permi", this value enters
into calculating the flow to block (I+1,J,K) as well as to block (I-1,J,K).
If a dual porosity model is being used, values for permeabilities are required for both
*MATRIX and *FRACTURE. Matrix permeabilities are required for calculating the fluid
flow between the matrix and fracture systems, and for computing matrix to matrix flow for
*DUALPERM, while fracture permeabilities are required for calculating fracture to fracture
flows between blocks.
*MATRIX permeabilities should be the values measured from a piece of unfractured matrix
material, while *FRACTURE permeabilities should be entered as effective fracture
permeabilities; that is, the permeability of the fracture system with respect to a volume of
reservoir rock. Effective fracture permeabilities are what is usually measured during a well
test or what would be computed from an open channel flow model with multiplication by a
fracture porosity.
*PERMJ and *PERMK may be specified using the *EQUALSI array input option, provided
that a *PERMI array is entered first. *EQUALSI may not be used if a Dual Continua model
is used with grid refinement.
Example:
Permeability input for a *DUALPOR model may appear as follows:
User's Guide IMEX
** Specify horizontal permeability

*PERMI *FRACTURE *ALL
2500.
2200. 2150.
2300.
2200.
...
*PERMI *MATRIX *ALL
340.
315.
280.
...
260.
240.
** J direction permeabilities are equal to I

** direction values.
*PERMJ *EQUALSI
** Vertical permeability is one tenth of the
** horizontal permeability.
*PERMK *EQUALSI * 0.10
The acceptable range of values for permeability is:
min
max
User's Guide IMEX
SI
mD
0.0
1.0E+9
Field
mD
0.0
1.0E+9
Lab
mD
0.0
1.0E+9
Netpay (Optional)
*NETPAY
PURPOSE:
*NETPAY indicates input of an array of net pays which are to be converted internally to an
array of net-to-gross multipliers.
ARRAY:
*NETPAY
DEFAULTS:
Optional keyword. The default net-to-gross multiplier is 1.0 for grid blocks that are not
supplied with net pay values, or equivalently, net pay equals gross pay, the latter being
defined by *DK or corner point input.
In dual porosity models, net pay values can be applied to the matrix and fracture pore volumes
separately by use of the *MATRIX and *FRACTURE qualifiers. However, the default for
fracture multipliers is still 1.0, even if *NETPAY *MATRIX (or just *NETPAY) appeared, and
the default for matrix multipliers is still 1.0, even if *NETPAY *FRACTURE appeared.
It is recommended that *NETPAY *MATRIX and *NETPAY *FRACTURE be used
CONDITIONS:
Values for net pays must be non-negative, and may exceed the values input for gross pays.
EXPLANATION:
The keyword *NETPAY allows input of net thicknesses ( m | ft | cm ) which are converted to
net-to-gross ratios by dividing the input array values by the respective gross block
thicknesses, the latter being obtained from *DK or corner point input. These net-to-gross
ratios are used to modify the porosity array and permeability arrays in the I and J directions.
The net-to-gross ratios are used as multiplication modifiers as follows:
a)
b)
c)
por is replaced by
permi is replaced by
permj is replaced by
por*ntg
permi*ntg
permj*ntg
where "por" denotes the grid block's porosity as set using the *POR keyword, "permi" and
"permj" are the block's permeabilities as set using the *PERMI and *PERMJ keywords. "ntg" is
the net-to-gross ratio as derived from the net pay value assigned by the *NETPAY keyword
divided by the appropriate thickness obtained from the *DK array or corner point input.
Note that the permeability in the K direction is not altered. Transmissibility multipliers (see
the *TRANSK keyword following) are available for vertical flow adjustments due to full or
partial barriers caused by shales or other geological features.
User's Guide IMEX
The acceptable range of values for any derived net-to-gross ratio is:
min
max
User's Guide IMEX
SI
Field
Lab
0.0
1.0E+4
0.0
1.0E+4 0
0.0
1.0E+4
Netgross (Optional)
*NETGROSS
PURPOSE:
*NETGROSS indicates input of an array of net-to-gross multipliers.
FORMAT:
*NETGROSS
DEFAULTS:
Optional keyword. The default net-to-gross multiplier is 1.0.
In dual porosity models, net-to-gross multipliers can be applied to the matrix and fracture
pore volumes separately by use of the *MATRIX and *FRACTURE qualifiers. However, the
default for fracture multipliers is still 1.0, even if *NETGROSS *MATRIX (or just
*NETGROSS) appeared, and the default for matrix multipliers is still 1.0, even if
*NETGROSS *FRACTURE appeared.
It is recommended that *NETGROSS *MATRIX and *NETGROSS *FRACTURE be used
CONDITIONS:
This keyword must be in the Reservoir Description keyword group. This keyword should not
be used if *NETPAY was used (see above).
EXPLANATION:
The keyword *NETGROSS allows input of net-to-gross which are used to modify the
porosities and permeabilities in the I and J directions. The net-to-gross ratios are used as
multiplication modifiers as follows:
a)
por is replaced by
por
*ntg
b)
permi is replaced by
permi
*ntg
c)
permj is replaced by
permj
*ntg
where "por" denotes the grid block's porosity as set using the *POR keyword, "permi" and
"permj" are the block's permeabilities as set using the *PERMI and *PERMJ keywords, and
"ntg" is the incoming net-to-gross ratio for the block. These modifications are used internally
and do not appear in the output.
Note that the permeability in the K direction is not altered. Transmissibility multipliers (see
the *TRANSK keyword) are available for vertical flow adjustments due to full or partial
barriers caused by shales or other geological features. Transmissibility multipliers can still be
used to further alter the flows in the I and J directions.
The acceptable range of values is:
min
max
SI
Field
Lab
0.0
1.0E+4
0.0
1.0E+4 0
0.0
1.0E+4
User's Guide IMEX
Transmissibility Multipliers (Optional)
*TRANSI, *TRANSJ, *TRANSK
PURPOSE:
*TRANSI indicates input of an array of I direction transmissibility multipliers.
*TRANSJ indicates input of an array of J direction transmissibility multipliers.
*TRANSK indicates input of an array of K direction transmissibility multipliers.
ARRAY:
*TRANSI
*TRANSJ
*TRANSK
DEFAULTS:
In dual porosity models, transmissibility multipliers can be applied to both the matrix and
fracture pore volumes separately by use of the *MATRIX and *FRACTURE qualifiers. The
defaults for *TRANSI *MATRIX is 1.0, regardless of values set for *TRANSI
*FRACTURE, and the same holds if *MATRIX and *FRACTURE are reversed. The same
comments hold for *TRANSJ and *TRANSK.
Transmissibility multipliers may be altered in recurrent (well) data. Multipliers that are
altered will take on their assigned values, while all others will retain their existing values; that
is, the values set by any appearance of *TRANSI, *TRANSJ, and *TRANSK keywords in
the Reservoir Description, other values being defaulted to 1.0
CONDITIONS:
These keywords may be in the Reservoir Description keyword group. They may also appear
in recurrent (well) data.
EXPLANATION:
Flow between grid blocks is proportional to a cross-sectional inter-block flow area, an averaged
permeability value, and a divisor equal to the inter-block distance. These terms combine to
form a transmissibility which is calculated in the simulator. Before it is used, a transmissibility
multiplier, as set by a *TRANSI, *TRANSJ or *TRANSK keyword, is applied.
The multiplier defaults to 1.0, equivalent to applying no multiplier, if no data is present.
Transmissibility multipliers are dimensionless.
*TRANSJ and *TRANSK may be specified using the *EQUALSI array input option,
provided that a *TRANSI array is entered first.
Since two blocks enter into any inter-block flow calculation, a method is required for
deciding which block will contribute the multiplier.
User's Guide IMEX
Inter-block flow between blocks on a single grid:

This rule applies whether the grid is the (main) fundamental grid (grid 1), or any refined grid.
If flow between a pair of blocks is considered, and they both lie on the same grid, then it is
the block with the lowest I index for an I direction pair, or the lowest J index for a J direction
pair, or the lowest K index for a K direction pair, that supplies the multiplier; that is, a
directional multiplier applies to a block's interface with its neighbour with the higher index in
that direction. These rules apply even when faults are present (see *FAULT following).
Note that fault considerations only affect lateral (I and J direction) calculations.
This rule is altered for *GRID *RADIAL and *GRID *HYBRID when connecting block nj
to block 1 in direction J when nj exceeds 1; that is, when a subdivided ring is being closed. In
this case, the multiplier from block nj is used for the closure. Also, flow perpendicular to the
wellbore in *HYBRID grids uses a multiplier averaged over the two directions perpendicular
to the well.
Note that refined grids inherit the multipliers from their parent block, unless special values
are read for the refined grid directly (*RG qualifier).
Note that except for the special case of zero transmissibility multipliers, all refined blocks in
a locally refined block inherit the multipliers from their parent block. Interior blocks and
those at a refined grid refined grid interface would inherit the parents multipliers.
In the special case of a zero multiplier, the multiplier is inherited only by those refined blocks
on the appropriate interface.
If the user needs to modify transmissibilities of an interface to a value other than zero when
refined grids are involved, the use of the *RG keyword is required to explicitly refer to
refined blocks at the interface.
Flow between a refined grid and the fundamental:
Basically the same rules apply as for fundamental blocks, except when determining the I, J,
or K index of a refined block at a refined block fundamental block interface, refer to its
parents I, J, or K index. If the refine blocks parent has the lowest I (J, or K) index then the
multiplier or the refined block is used. If the adjoining fundamental block has the lowest
index then the multiplier of the adjoining fundamental block is used. This also applies to the
*TRANLI (J, K) keywords except that the fundamental block with the highest index is used.
Use of this rule and *TRANLI (J, K) make it possible to only refer to fundamental blocks
when defining non zero transmissibility multipliers between refined and fundamental blocks.
Flow between two refined grids:
Again the same rules apply. Refer to the I, J, or K index of the parent blocks for both refined
blocks.
Dual porosity models:
*MATRIX transmissibilities are applied to matrix to fracture flows for all double porosity
models, except for *DUALPERM (when no such multipliers exist). *MATRIX multipliers are
applied to matrix to matrix flows within a block for *SUBDOMAIN and *MINC.
If a *DUALPERM model is being used, the *MATRIX transmissibility multipliers are used for
modifying inter-block matrix to matrix flow in the same manner that single porosity multipliers
operate. There are no matrix to fracture flow multipliers available for this case.
User's Guide IMEX
It is the I and J direction multipliers that are often zeroed with the *DUALPERM model, leaving
the K direction multipliers nonzero. This choice is made because the most important dual
permeability effects are usually in the vertical direction, arising due to phase density differences.
Examples:
The following provides an example of standard transmissibility multiplier usage:
** Specify horizontal transmissibility multipliers
*TRANSI *FRACTURE *ALL
1.4 2*1.2 1.4 1.5 1.4
...
*TRANSI *MATRIX *ALL
1.2 1.3
1.4
1.1
...
1.2
1.4
*TRANSJ *EQUALSI
** Vertical transmissibility is one tenth of the
** horizontal transmissibility.
*TRANSK *EQUALSI * 0.10
Suppose block (2,2,2) contains a 3 x 2 x 1 refined grid. Then I direction transmissibility

multipliers can be applied to flows going in or out of the refined grid as follows:
*TRANSI *RG 1 1 1 *ALL
.8 1 .8 .8 1 .8
min
max
User's Guide IMEX
SI
Field
Lab
0.0
1000.0
0.0
1000.0
0.0
1000.0
Transmissibility Multipliers for Lower Indexed Block Faces

(Optional)
*TRANLI, *TRANLJ, *TRANLK
PURPOSE:
*TRANLI indicates input of I direction transmissibility multipliers for faces contacting lower
indexed blocks.
*TRANLJ indicates input of J direction transmissibility multipliers for faces contacting lower
indexed blocks.
*TRANLK indicates input of K direction transmissibility multipliers for faces contacting
lower indexed blocks.
FORMAT:
*TRANLI
*TRANLJ
*TRANLK
DEFAULTS:
CONDITIONS:
These keywords may be in the Reservoir Description keyword group or they may be in
recurrent (well) data.
EXPLANATION:
Flow between grid blocks is proportional to a cross-sectional inter-block flow area, an
averaged permeability value, and a divisor equal to the inter-block distance. These terms
combine to form a transmissibility which is calculated in the simulator. Before it is used, a
transmissibility multiplier is applied. The multiplier can be set using the *TRANSI,
*TRANSJ, or *TRANSK keywords (as described elsewhere) or the *TRANLI, *TRANLJ, or
*TRANLK keywords described here.
All transmissibility multipliers are dimensionless.
Transmissibility multipliers can be specified for any grid block. A default value of 1.0 will
be used for unspecified multipliers.
When transmissibility multipliers appear in recurrent (well) data, any block's multipliers may
be (re)assigned, BUT unreferenced blocks retain their values assigned earlier (which will be
1.0's if no other values were ever assigned).
Since two blocks enter into any inter-block flow calculation, a method is required for
deciding how blocks will contribute multipliers.
If flow between a pair of blocks is considered, it is the block with the highest I index for an I
direction pair, or the highest J index for a J direction pair, or the highest K index for a K
direction pair, that supplies multiplier values set by the *TRANLI, *TRANLJ, or *TRANLK
keywords, respectively. This behavior is the opposite of the assignment of multipliers based
on the *TRANSI, *TRANSJ, or *TRANSK keywords, where the lower indexed block in the
pair supplies the multiplier.
User's Guide IMEX
If both types of multipliers have been defined for a face, one coming from a *TRANLI,
*TRANLJ or *TRANLK value assigned to the higher indexed block, and a *TRANSI,
*TRANSJ or *TRANSK value assigned to the lower indexed block, then the following rules
are applied, in the order shown, to determine the final transmissibility:
1. if both values are 1, then no modification is performed (multiplier is 1);
2. if either value is 0, then no fluid flow is allowed (multiplier is 0);
3. if one value is 1, and the other is not 1, then modification is based on the non-unity
value (multiplier is the non-unity value);
4. if both values are not 1, then the arithmetic average of the two values is used
(multiplier is the average of the two values).
Thus, setting a zero *TRANLI, *TRANLJ, *TRANLK, *TRANSI, *TRANSJ, or *TRANSK
cuts off all fluid flow at a face.
These multipliers can be used to control flow between refined grids, or from the fundamental
grid to a refined grid. These multipliers apply even when faults are present (see *FAULT
following). Faults only use the *TRANLI and *TRANLJ multipliers.
The rules of how these low side multipliers apply to refined blocks is essentially the same
as those for *TRANSI (J, or K). The user should always refer to the I, J, or K indices of the
parent blocks and apply the same rules as he would on a fundamental grid to determine which
blocks multipliers are used.
These multipliers have no effect on flow between matrix and fracture in dual porosity models
(*DUALPOR and *DUALPERM). *MATRIX flow values are used for matrix to matrix
flow between different blocks in a DUAL PERMEABILITY model.
*TRANLI, *TRANLJ, or *TRANLK should not be used with *HYBRID grids.
*TRANLJ and *TRANLK may be specified using the *EQUALSI array input option,
providing that *TRANLI is entered first.
min
max
User's Guide IMEX
SI
Field
Lab
0.0
1000.0
0.0
1000.0
0.0
1000.0
Pinch Out Array (Optional)
*PINCHOUTARRAY
PURPOSE:
*PINCHOUTARRAY defines pinch outs using an array input format. (See also
*PINCHOUT-TOL) (This keyword replaces the older keyword *PINCHOUT.)
FORMAT:
*PINCHOUTARRAY
DEFAULTS:
Optional keyword. Default: No pinch outs.
CONDITIONS:
*PINCHOUTARRAY cannot appear in the same data set as the *PINCHOUT-TOL keyword.
Pinched out cells may be set using *PINCHOUTARRAY, or they can detected by a thickness
tolerance set by *PINCHOUT-TOL, but only one technique is allowed per data set. Note that,
regardless of the technique chosen, blocks with true zero thickness (*DK value of 0.0, or corner
point cells entered with top corner points that are equal to bottom corner points) will be treated
as pinched out, unless they had been flagged null using the *NULL keyword (see later).
All array qualifiers and array reading options are allowed for specifying the required ni * nj *
nk values. The array values should consist of 0's to indicate blocks that are pinched out and
1's to indicate blocks that are not pinched out. (This keyword is similar to the *NULL
keyword in that 1's are used to indicate active blocks and 0's are used to indicate special
blocks that do not fully participate in the simulation.)
Note that if a vertical stack of one or more corner point cells are pinched out, and they are
surrounded above and below by active corner point cells, then the blocks above and below
will not be connected to each other UNLESS the pinched out cells form an uninterrupted
stack of contacting blocks. The presence of a gap will break the connection, where
*CORNER-TOL gives the tolerance describing how close cells need to be before they are
deemed to make contact.
Pinching out blocks on *HYBRID refined grids is not recommended.
This keyword replaces the older *PINCHOUT keyword. Use of this older keyword is no
longer recommended.
EXPLANATION:
*PINCHOUTARRAY indicates the modelling of pinched out layers. Such blocks will not
participate in any of the simulator's flow calculations; that is, they will be inactive. However,
fluid will be permitted to pass through them in the vertical direction (only).
Pinched out blocks are used to remove layers from the simulator's calculations in certain
regions on a grid. Such layers may be required to model geological strata that exist in other
portions of the grid but are not present in the pinched out region. The *PINCHOUTARRAY
keyword corresponds to true geological pinch outs.
Blocks that are designated as pinched out allow fluid to pass through them vertically, but not
laterally, and fluid can pass through a stack of one or more pinched out blocks on a grid.
User's Guide IMEX
Pinched out blocks can also lie between active blocks on a grid and a refined grid region,
allowing fluid to pass through between the grids. Two *HYBRID grids can even be
connected vertically through intervening pinched out layers on their parent grid.
Pinched out blocks should have relatively small thicknesses (a *DK array value of nearly 0.0
or matching, or nearly matching, top and bottom corner points). since the transmissibility
calculations between cells above and below the pinched out stack do not use the intervening
stack's thickness.
A pinched out status set with *PINCHOUTARRAY over-rides an inactive setting using
*NULL. This means that if a block has been designated pinched out in a
*PINCHOUTARRAY list, fluid will pass through it regardless of it having been also
designated inactive in a *NULL list. However, a *NULL setting overrides zero thickness;
that is, a zero thickness block will not allow vertical fluid passage if it is flagged inactive
using the *NULL keyword. Note that zero thickness overrides zero pore volume; that is, a
block that is pinched out due to having zero thickness will allow fluid passage regardless of
whether it was assigned non-zero porosity values or not.
The simulator uses the following hierarchy when determining whether a block is pinched out
(allows vertical fluid passage) or is completely inactive. Note that Rule 1) overrides Rule 2),
which overrides Rule 3), which overrides Rule 4).
1. The block has been flagged as pinched out using the *PINCHOUTARRAY
keyword (a 0 value was assigned). This block will always be pinched out and this
state will not be overridden.
2. The block has been flagged as inactive using the *NULL keyword (a 0 value was
assigned). This block will not allow vertical fluid passage, unless overridden by
Rule 1).
3. A zero thickness block will be pinched out (allows vertical fluid passage) if not
overridden by Rules (1-2).
4. A zero pore volume block will be inactive and not allow any fluid passage unless
overridden by Rules (1-3).
Example:
To pinch out the second layer of a 100 X 100 X 9 model use the following:
*PINCHOUTARRAY *IJK 1:100 1:100 2:2
Note that the remaining blocks in the model need not be referred to and their state remains
unaltered.
User's Guide IMEX
Pore Volume Cut-Off Threshold (Optional)
*PVCUTOFF
PURPOSE:
*PVCUTOFF controls the level at which a block will be set null due to a small pore volume.
FORMAT:
*PVCUTOFF pvcut
DEFINITIONS:
pvcut
Pore volume (block volume multiplied by porosity) below which a block will
be considered to be null. Dimensions are (m3 | ft3).
DEFAULTS:
Optional keyword. Default is to examine the *POR values and any *NULL keyword input
for null block determination.
CONDITIONS:
EXPLANATION:
This option ensures that blocks with small pore volumes can be systematically removed from
the simulation. Such small pore volume blocks can hinder convergence and should not
remain in a simulation.
User's Guide IMEX
Pinchout Tolerance (Optional)
*PINCHOUT-TOL
PURPOSE:
*PINCHOUT-TOL controls the minimal thickness required to initiate an automatic pinched
out connection (see also *PINCHOUTARRAY).
FORMAT:
*PINCHOUT-TOL pnctol
DEFINITIONS:
pnctol
Minimal thickness required under which a block is removed from the
simulation and the block above is connected directly to the block below.
Dimensions are (m | ft).
DEFAULTS:
Optional keyword. The defaults are:
0.0010 (m | ft) for Corner Point grids on the fundamental grid only, if
*PINCHOUTARRAY does not appear;
0.0002 (m | ft) for non-Corner Point Grids on the fundamental grid only, if
*PINCHOUTARRAY does not appear;
0.0 (m | ft) otherwise.
Values at computer round-off levels are considered to be identically 0.0.

CONDITIONS:
*PINCHOUT-TOL cannot appear in the same data set as the *PINCHOUTARRAY keyword.
Pinched out cells may be set using *PINCHOUTARRAY, or they can detected by a thickness
tolerance set by *PINCHOUT-TOL, but only one technique is allowed per data set. Note
that, regardless of the technique chosen, blocks with true zero thickness (*DK value of 0.0, or
corner point cells entered with top corner points that are equal to bottom corner points) will
be treated as pinched out, unless they had been flagged null using the *NULL keyword.
Note that if a vertical stack of one or more corner point cells are pinched out. And they are
surrounded above and below by active corner point cells, then the blocks above and below
will not connect to each other UNLESS the pinched out cells form an uninterrupted stack of
contacting blocks. The presence of a gap will break the connection, where *CORNER-TOL
gives the tolerance describing how close cells need to be before they are deemed to make
contact.
Having pinching out blocks on *HYBRID refined grids is not recommended.
User's Guide IMEX
EXPLANATION:
Block whose thickness are less than pnctol are considered to be pinched out. When this
occurs, blocks above and blocks below the pinched out blocks are connected as if the pinched
out blocks did not exist. The minimum thickness pnctol at which this occurs is controlled
using *PINCHOUT-TOL.
Pinched out blocks are used to remove layers from the simulator's calculations in certain
regions on a grid. Such layers may be required to model geological strata that exist in other
portions of the grid but are not present in the pinched out region. The *PINCHOUT-TOL
and *PINCHOUTARRAY keywords correspond to true geological pinch outs.
Blocks that are designated as pinched out allow fluid to pass through them vertically, but not
laterally, and fluid can pass through a stack of one or more pinched out blocks on a grid.
Pinched out blocks can also lie between active blocks on a grid and a refined grid region,
allowing fluid to pass through between the grids. Two *HYBRID grids can even be
connected vertically through intervening pinched out layers on their parent grid.
*NULL settings override *PINCHOUT-TOL-generated pinch outs; that is, a small thickness
block will not allow vertical fluid passage if it is flagged inactive using the *NULL keyword.
Note that small thickness overrides zero pore volume; that is, a block that is pinched out due
to having small thickness will allow fluid passage regardless of whether it was assigned nonzero porosity values or not.
User's Guide IMEX
Corner Point Tolerance (Optional)
*CORNER-TOL
PURPOSE:
*CORNER-TOL controls the tolerance used to decide if two corner points are the same. It is
also used for miscellaneous tolerance checking for corner point applications.
FORMAT:
*CORNER-TOL cptol
DEFINITIONS:
cptol
Minimal spacing required to separate corner points and related quantities;
that is, corner points that are closer than "cptol" are deemed to be the same.
Dimensions are (m | ft).
DEFAULTS:
Optional keyword. Default is 0.050 (m | ft).
CONDITIONS:
EXPLANATION:
Corner points that lie within an (Euclidean) distance of "cptol" are considered to be the same
for the purpose of deciding whether the cells make contact. Points from neighbouring cells
(four points from the top of one cell and four points from the bottom of the other) that are
supposed to be touching to make a standard flow connection will be regarded as making
contact if they lie within a distance of "cptol" (in top-bottom pairs). Extensions of this notion
are used to determine connectivity around faults and when refined grids are present, and
"cptol" plays a similar role in thee situations.
If a vertical stack of one or more corner point cells are pinched out, and they are surrounded
above and below by active corner point cells, then the blocks above and below will not
connect to each other UNLESS the pinched out cells form an uninterrupted stack of
contacting blocks. The presence of a gap will break the connection, where "cptol" gives the
tolerance describing how close cells need to be before they are deemed to make contact.
"cptol" is also used when determining corner orderings and making some miscellaneous
checks of cell quality.
User's Guide IMEX
Faults (Optional)
*FAULT
PURPOSE:
*FAULT indicates the input of designations for grid blocks whose flow connections are to
take into account their exact position in the reservoir with respect to their lateral neighbours.
Each *FAULT keyword is expected to describe a group of grid blocks that together form a
geological fault block.
FORMAT:
*FAULT
throw
i1:i2 j1:j2
:
:
DEFINITIONS:
throw
Geologically speaking, "throw" is the difference in depth between a
geological fault block and neighbouring reservoir rock. (m | ft | cm). (See
FIGURE 6 in Appendix D.)
In the simulator, throws provide modifications to depth data given earlier
through use of the *DEPTH, *DTOP, *PAYDEPTH or *DEPTH-TOP
keywords. A zero throw is valid if the depth information is already complete
and only the identification of the grid blocks involved in the geological fault
block is required.
i1:i2 j1:j2
The indices, i1, i2, j1, and j2 locate grid block columns whose first index (I
index) lies between i1 and i2 inclusive, whose second index (J index) lies
between j1 and j2 inclusive, and whose third index (K index) lies between 1
and nk inclusive.
The grid block columns identified by successive lines of these indices will
make up a geological fault block.
DEFAULTS:
Optional keyword. Default: no faults.
CONDITIONS:
*FAULT should not be used with *GRID *CORNER. (Fault data can be entered directly for
corner point grids as was discussed earlier.)
EXPLANATION:
Geologic faults are formed when a portion of the reservoir is dislocated with respect to
another portion. These dislocated portions form geological fault blocks. Lateral flow cannot
follow the usual geological strata in these cases. To take this into account when modelling a
reservoir, it is necessary to be able to group grid blocks into fault blocks, and to take account
of these fault blocks when developing inter-block communication.
User's Guide IMEX
The fault model described here assumes that each geologic fault block can be described by
the grid blocks in a collection of grid block range descriptions which operate as noted above.
Note that fault blocks must extend through the entire reservoir. For convenience, a "throw"
value can be applied to the depths of all grid blocks in a fault block.
Note that throws can be positive, zero, or negative, and that they will be added directly to the
already existing depth values. Thus, comments given earlier regarding depth measurements
(see *DEPTH, *DTOP, *PAYDEPTH and *DEPTH-TOP keywords) will apply. If a full and
correct depth array was introduced earlier (using the *PAYDEPTH option, for instance), the
throw values can be set to 0.0. (If a grid block is assigned to more than one fault block,
throws accumulate in the order they are input.)
When it comes time to compute transmissibilities for lateral inter-block flow, and fault blocks
are present, special checking is carried out. For instance, if lateral flow into grid block (I,J,K)
from its positive I-direction is being considered, which is normally flow from block
(I+1,J,K), and *FAULT keywords appeared, the following is done.
If the high side of block (I,J,K) lies on the edge of a fault block (that is, (I,J,K) was identified
in a *FAULT list that did not refer to (I+1,J,K)), or any block in the adjacent column has its
low side on the edge of a (necessarily) different fault block (that is, (I+1,J,K) was identified
in a *FAULT list that did not refer to (I,J,K)), then block (I,J,K) will be connected to ANY
block of the form (I+1,J,KK) that has a positive vertical overlap with block (I,J,K).
Moreover, the transmissibility calculation will take into account the amount of actual overlap.
A similar calculation will be done for the low side of block (I,J,K), and for the J direction
cases.
Vertical transmissibility calculations are not affected by such fault considerations, as are
flows internal to fault blocks.
Thus, exact positioning at fault block boundaries governs inter-block flows, as grid blocks in
one fault block will no longer align with their usual lateral neighbours.
The acceptable range of values for throws area:
min
max
User's Guide IMEX
SI
m
1.0E-3
1.0E+3
Field
ft
.00328
3,280.0
Lab
cm
0.1
1.0E+5
Aquifers (Optional) see Appendix G
*AQUIFER, *AQPROP, *AQMETHOD,
*AQLEAK
PURPOSE:
*AQUIFER indicates the input of aquifer location information.
*AQPROP indicates the input of aquifer properties.
*AQMETHOD selects the numerical method to be used for calculating water influx from the
aquifer.
*AQLEAK determines the ability of water to leak back into the aquifer when under the
influence of a higher reservoir pressure.
FORMAT:
*AQUIFER
(*BOUNDARY)
(*BOTTOM)
(*REGION)
i1:i2 j1:j2 k1:k2
(*IDIR)
(*JDIR)
(*KDIR)
*AQPROP
Thickness Porosity Permeability Radius Angle (R-Ratio)
*AQMETHOD
*AQLEAK
(*CARTER-TRACY)
(*FETKOVITCH)
(*OLD)
(*ON)
(*OFF)
(*OLD)
DEFINITIONS:
*BOUNDARY
The aquifer is to be connected to all fundamental grid edge blocks (either I
equals 1 or ni, or J equals 1 or nj, or both) in the appropriate I or J direction
(ni = last block in x direction, nj = last block in y direction).
*BOTTOM
The aquifer is to be connected to the bottom of the reservoir; that is, the
bottom of all fundamental grid blocks in the deepest layer.
*REGION
The aquifer is to be connected to the region indicated by the following range
(i1:i2 j1:j2 k1:k2) and direction keywords (*IDIR, *JDIR or *KDIR)
keywords.
User's Guide IMEX
i1:i2
where the aquifer is to be connected.
j1:j2
k1:k2
*IDIR
The direction of inflow between the aquifer and the reservoir boundary is to
be in the I direction.
*JDIR
be in the J direction.
*KDIR
be in the K directions.
thickness
Aquifer thickness (m | ft | cm).
porosity
Aquifer porosity (fraction).
permeability
Aquifer permeability (md | md | md).
radius
Effective reservoir radius (m | ft | cm).
angle
Aquifer angle of influence (expressed as a fraction of a circle).
R-Ratio
Ratio of the aquifer's external radius to that the reservoir's effective radius
(fraction).
*CARTER-TRACY
Calculate water influx from the aquifer using the Carter-Tracy
approximation.
User's Guide IMEX
*FETKOVITCH
Calculate water influx from the aquifer using a formulation based on work by
Fetkovitch.
*AQMETHOD *OLD
Selects the aquifer calculation for simulator versions numbered "97.xx" and
earlier.
*AQLEAK
Indicates whether or not water is allowed to leak from the reservoir into the
aquifer at a block where the block pressure exceeds the adjacent aquifer
pressure. The behavior of the aquifer is modelled more accurately when
*ON follows *AQLEAK, and leakage is allowed, rather than when *OFF is
used, when no leakage is allowed. If *AQLEAK is not specified, this is
equivalent to *AQLEAK *OFF.
*AQLEAK *OLD
Selects the aquifer leak off calculation for simulator versions numbered
"98.xx" and earlier. The *OLD method, in some instances allowed aquifer
leakage even when leakage was not allowed. *AQLEAK *OFF (the default)
does not under any condition allow aquifer leakage when none was specified.
DEFAULTS:
Optional keywords. No defaults for *AQUIFER.
Defaults are assigned as follows:
Thickness:
average thickness for *BOUNDARY and *REGION
-orsquare root of the contact area (see below), for *BOTTOM
Porosity:
average porosity
Permeability:
average reservoir permeability, averaged over the aquifer flow directions
Radius:
the radius of the circle whose circumference, which when multiplied by the
thickness gives the contact area, for *BOUNDARY -or- the square root of
the (typical) area of a side of the reservoir, divided by pi (see below), for
*BOTTOM -or- the square root of the contact area divided by pi, for
*REGION
User's Guide IMEX
Angle:
full circle (1.0) for all grids except *RADIAL, when the sum of the outer
ring angular extents is used (after division by 360.0), for *BOUNDARY and
*REGION -or- angular extent of the bottom of the reservoir (after division
by 360.0; see below), for *BOTTOM
R-Ratio:
defaults to 100.0 for *AQMETHOD *FETKOVITCH
That is:
(aquifer external radius )
(reservoir effective radius )
= 100.0
In the above, "average" refers to a pore-volume weighted average taken over aquifercontacting cells, and "contact area" means the sum of the areas of all cell faces that are
defined to contact the aquifer, as specified by the *AQUIFER keyword.
*AQMETHOD defaults to *CARTER-TRACY using a dimensionless pressure influence
function for an infinite aquifer.
CONDITIONS:
Hydrocarbon to aquifer volume ratio Printout
The output is in the block Hydrocarbon Reservoir Information (including subgrids) and is
called Hydrocarbon to aquifer volume ratio.
An example is presented below:
Hydrocarbon to aquifer volume ratio:
1.84110E-02 rm3/rm3
The Aquifer volume includes both gridded and analytical aquifer volumes.
In order to give the correct analytical aquifer volume, the aquifers R-Ratio must be entered
for each aquifer (see keyword *AQPROP). This is true even if the aquifer is a Carter Tracy
Aquifer. The Carter Tracy aquifer model does not explicitly use the input R-Ratio in its
calculation as the R-Ratio is implicitly entered using the Aquifer pressure influence function
tables. However, without the explicit input of the R-Ratio, there is no way for the Simulator
to determine the R-Ratio to be used.
For Carter-Tracy aquifers, the R-Ratio is only used for printout purposes, but in order to be
meaningful, the R-Ratio input should be identical to the R-Ratio used to derive the pressure
influence function table in use. If R-Ratio is allowed to default, it will assume a value of
10.00.
The gridded aquifer volume used in this calculation is estimated by accumulating the volume
of grid blocks whose water saturation is at least 99.9% of *WOC_SW. *WOC_SW defaults
to 1.00 in IMEX.
User's Guide IMEX
EXPLANATION:
The aquifer modelling described here allows water influx (and reverse flow if *AQLEAK *ON
is enabled) to a reservoir from an aquifer, or several aquifers. Using an analytical aquifer
model can be more economical for simulation purposes than using grid blocks filled with water.
If great accuracy is required in modelling aquifers however, water filled blocks should be used.
Since outflow from the reservoir into the aquifer is modelled analytically when *AQLEAK
*ON is used, it is possible to have water outflow from a grid block attached to the aquifer that
has no mobile water. Thus, care must be exercised when using analytical aquifer models. It
is recommended that aquifers be modelled using grid blocks in cases where the reservoir
pressure is expected to increase significantly during the course of the simulation.
When dual porosity models are used, the aquifers will be connected to the fracture porosity of
grid blocks whenever possible. Otherwise, matrix porosity connections will be used.
The subkeyword *REGION indicates direct input of the aquifer boundary blocks. The
subkeywords *IDIR, *JDIR, and *KDIR indicate which face of the reservoir is in contact
with the aquifer. If more than one face is in contact with the aquifer, list each region on
subsequent lines.
*AQPROP specifies the aquifer properties for calculation of water influx from the aquifer.
*AQPROP defaults for *BOUNDARY aquifers envision a cylindrical reservoir with the
aquifer contacting the reservoir around the full cylinder. The defaults are constructed so that
the average thickness multiplied by the circumference gives the contact area calculated using
the *AQUIFER keyword.
*AQPROP defaults for *BOTTOM aquifers envision a square contact area with a side
dimension of L. Thus, L is equal to the square root of the bottom contact area. An average
reservoir thickness H is also used in these calculations. The aquifer is assumed to be bounded
by the edges of a wedge coming up to the bottom of the reservoir, as shown in the following
side view of the reservoir. The "D"'s define diagonals drawn through the corners and their
extensions, the "W"'s, define the wedge:
L
Reservoir
H
L
2 Arctan
Aquifer
User's Guide IMEX
where the wedge angle is 2*atan(L/H). Then, the theory for radial aquifers is applied in this
side view, with the value for Angle taken as 2*atan(L/H)/360, Thickness taken as L (viewed
as a dimension into the page), and Radius taken as the square root of (L*H/pi). This latter
choice is required as the reservoir must be circular to apply the theory. This defines defaults
for *AQUIFER *BOTTOM.
*AQPROP defaults for *REGION aquifers are much like those for *BOUNDARY except
that Radius is taken to be the square root of the contact area divided by pi.
If *AQMETHOD *CARTER-TRACY is specified, the simulator calculates water influx
using a Carter-Tracy approximation. For more information, refer to R. D. Carter and G. W.
Tracy, "An Improved Method for Calculating Water Influx", Trans., AIME, Vol. 219, (1960),
pp. 415-417. This method uses a dimensionless pressure influence function P(td), expressed
as a function of dimensionless time td. The function is defined for IMEX using a table (see
*AQFUNC below), along with an extrapolation method for dimensionless times that go
beyond the end of the table. If the internal infinite extent aquifer table is used, an analytical
expression is used for the extrapolation (see the Van Everdingen and Hurst reference
mentioned in the *AQFUNC section). Otherwise, linear extrapolation in dimensionless time
is used, which is appropriate for finite aquifers.
If *AQMETHOD *FETKOVITCH is specified, the simulator calculates water influx from
the aquifer using a formulation based on work by Fetkovitch (see, for instance, M. J.
Fetkovitch, "A Simplified Approach to Water Influx Calculations - Finite Aquifer Systems",
JPT, July 1971, pp. 814-828). This approach is able to model finite aquifers without having
to generate dimensionless pressure functions. Only a "R-Ratio" parameter is required.
Please be aware that for large aquifers, R-Ratio > 5, the Fetkovitch Method which ignores
transient effects, may not properly model aquifer influx. For larger aquifers use the CarterTracy Method.
If *AQMETHOD *OLD is specified, older calculations (from versions numbered "97.xx"
and earlier) are used. These versions used Carter-Tracy with a linear extrapolation for the
dimensionless aquifer pressure influence function. When the default infinite extent aquifer
function was being used, the linear extrapolation could result in an aquifer pressure drop that
was too large, which could result in an aquifer that could seem limited in extent. As noted
above, a standard analytical expression in log(td) is now used for large time extrapolation of
the infinite aquifer case. Some modifications of the defaults for *AQPROP parameters are
also introduced when *AQMETHOD *OLD is specified.
Block transmissibility multipliers affect the aquifer inflow. *BOUNDARY aquifers for
*RADIAL grids use *TRANSI multipliers. It the aquifer is of *BOTTOM type with layer
NK generally deeper than layer 1, then *TRANSK-defined multipliers are used. If layer 1 is
deeper than layer NK, then *TRANLK-defined multipliers are used. If the aquifer is of
*BOUNDARY or *REGION type, and I-direction contact is made on a cell with an I-index
I=1, then a *TRANLI-defined value is used. Similarly, if I-direction contact is made on a cell
with an I-index I=NI, then a *TRANSI-defined value is used. If aquifer contact is made on a
cell with an intermediate I-index, it is assumed that that cell has a null cell lying on one side
of it in the I-direction. For this case, the aquifer uses the *TRANSI/*TRANLI multiplier that
would have governed inter-block flow if the neighbour was not null. Similar comments hold
for the J- and K-directions.
User's Guide IMEX
Examples:
*AQUIFER *BOUNDARY
*AQPROP ...
-or*AQUIFER *BOTTOM
*AQPROP ...
Multiple use of *REGION is allowed to define complex aquifer connections.

*AQUIFER *REGION
3 4:6
7:9 *IDIR
*REGION 1:34
7:9 *JDIR
*REGION 4:7 5:7
9
*KDIR
*AQPROP
0. 0.3
150.0 0.0 .5
The acceptable range of values for the aquifer thickness is:
min
max
SI
m
0.0
1.0E+4
Field
ft
0.0
3.28E+4
Lab
cm
0.0
1.0E+6
User's Guide IMEX
Pressure Influence Function (Conditional) see Appendix G

*AQFUNC
PURPOSE:
*AQFUNC indicates input of a dimensionless pressure influence function as a table.
TABLE:
*AQFUNC
dimtime function
:
:
DEFINITIONS:
dimtime
Dimensionless time
function
Dimensionless pressure influence function
DEFAULTS:
Conditional keyword. See EXPLANATION for a discussion of the default table.
CONDITIONS:
This keyword must be in the Reservoir Description keyword group and must be used with
*AQUIFER.
EXPLANATION:
The simulator calculates water influx from (to) the aquifer using the Carter & Tracy
approximation. For more information, refer to R. D. Carter and G. W. Tracy, "An Improved
Method for Calculating Water Influx", Trans., AIME, Vol. 219, (1960), pp. 415-417. The
method requires the use of a dimensionless pressure influence function P(td) as a function of
dimensionless time td (see Appendix G).
The default dimensionless pressure function used is the one given in A. F. Van Everdingen
and W. Hurst, "The Application of the Laplace Transform to Flow Problems in Reservoirs",
AIME Dec. 1949, pp. 305-324, for a constant terminal rate solution and an infinite radial
aquifer. Influence functions for limited extent aquifers can also be found in this reference.
The default pressure influence functions and the corresponding dimensionless times for the
default case are tabulated below:
User's Guide IMEX
Dimensionless
Time, td
0.01
0.05
0.10
0.15
0.20
0.30
0.50
0.70
1.00
1.50
2.00
3.00
5.00
7.00
10.00
15.00
20.00
30.00
40.00
50.00
60.00
70.00
80.00
90.00
100.00
200.00
300.00
400.00
500.00
600.00
700.00
800.00
900.00
1000.00
Influence
Function, P(td)
0.112
0.229
0.315
0.376
0.424
0.503
0.616
0.702
0.802
0.927
1.020
1.169
1.362
1.500
1.651
1.829
1.960
2.147
2.282
2.388
2.476
2.550
2.615
2.672
2.723
3.064
3.263
3.406
3.516
3.608
3.684
3.750
3.809
3.860
Additional values may be found in the above reference.

An extrapolation method for dimensionless times that go beyond the end of the table is
required. If the above infinite extent table is used, an analytical expression is used for the
extrapolation (see the Van Everdingen and Hurst). Otherwise, linear extrapolation in
dimensionless time is used, which is appropriate for finite aquifers. Further discussion of
some compatibility issues relating to older versions of IMEX are given above under the
*AQUIFER section.
User's Guide IMEX
Sectors (Optional)
*SECTOR
PURPOSE:
*SECTOR controls the definitions of sectors, which are used to summarize regional reservoir
activity.i1:i2 j1:j2 k1:k2
FORMAT:
*SECTOR
'sector'
i1:i2
:
j1:j2
:
k1:k2
:
DEFINITIONS:
'sector'
Sector identification name, enclosed in single quotes (16 character
maximum).
i1:i2
Indicates the I beginning and ending indices of the grid region where the
sector is to be located.
j1:j2
Indicates the J beginning and ending indices of the grid region where the
k1:k2
Indicates the K beginning and ending indices of the grid region where the
DEFAULTS:
Optional keyword. The first sector is internally defined to include all active grid blocks and
is named: 'Entire Field'. Other sectors defined as above are assigned higher sector numbers.
CONDITIONS:
EXPLANATION:
Sectors can be used for printing the values of several quantities relating to the grid blocks that
were assigned to the sector. These quantities include:
User's Guide IMEX
Total and hydrocarbon pore volume
Fluids in place at surface and reservoir conditions
Average Pressure for the total pore volume
Average pressure with datum correction for total pore volume
Average pressure for the hydrocarbon pore volume
Average saturation for oil, gas, water or solvent
Average concentration for polymers

Percentage recovery, defined as:

% RC =
(cumulative production cumulative injection )

(initial material in place )
Cumulative injection and production values
Production and injection rates
The frequency of printing the table containing these quantities to the output-file is controlled
by the keyword *WPRN *SECTOR.
Grid blocks can belong to several sectors. Of course, all grid blocks are in the first sector by
default.
Example:
A grid system with some desired sector allocations is given below:
J=6
S1
S1
S1
S1/S3
S3
S3
S3
J=5
S1
S1
S1
S3
S3
S3
S2/S3
J=4
S1
S1
S1
S2
S2
S2
S2
S2
S2
S2
S2
S2
S2
S2
S2
J=3
J=2
S2
J=1
I=1
The following input will define those sectors:

*SECTOR
'area1'
*SECTOR
'area2'
'area2'
*SECTOR
'area2'
1:3
4
3:6
4:7
5:7
6:7
7
4:6
6
1
2
3
4
5
1
1
1
1
1
1
1
Note that there is considerable flexibility in the way that the *SECTOR keyword can be used.
See the following *SECTORARRAY keyword for an alternative method.
User's Guide IMEX
Sector Array (Optional)
*SECTORARRAY
PURPOSE:
*SECTORARRAY defines Sectors (see *SECTOR) using an array input format. Sectors
permit output to be printed on a regional basis.
FORMAT:
*SECTORARRAY 'Sector_Name'
DEFINITIONS:
'Sector_Name'
Sector name. (16 characters maximum)
DEFAULTS:
Optional keyword. The full FIELD sector, and MATRIX and FRACTURE sectors if dual
porosity or dual permeability is used, are always defined internally.
CONDITIONS:
This keyword must be in the Reservoir Description keyword group. All array qualifiers and
array reading options are allowed for specifying the required ni * nj * nk values. The
qualifiers and array values should come after 'Sector_Name'. The array values should consist
of 0's (no sector membership for that cell) or 1's (indicating sector membership for that cell).
EXPLANATION:
Sectors are collections of grid blocks that the simulator uses to summarize various quantities.
A grid block can belong to different sectors. See the description of *SECTOR above for
further information about sectors.
The *SECTORARRAY keyword provides an array-based alternative to the *SECTOR rangebased input format.
Example:
To set sector membership in a sector named 'Sector-1' for a few cells in the reservoir, use the
following
*SECTORARRAY 'Sector-1' *IJK 1:5 1:5 1:1
4
4
1
2
3
1
0
1
1
where it is assumed the grid is dimensioned 5x5x1. Note that the string "1:5 1:5 1:1 0" is not
actually required as "no membership" is the default state.
User's Guide IMEX
Sector Assignment via Names and an Array (Optional)

*ISECTOR, *SECTORNAMES
PURPOSE:
*SECTORNAMES introduces sectors via a list of sector names and corresponding sector
numbers (see *SECTOR for a discussion of sectors). This keyword should only appear once
in the data set.
*ISECTOR assigns these sector numbers to cells using the standard array concepts.
FORMAT:
*SECTORNAMES 'Sector_Name_1' i1 'Sector_Name_2' i2 'Sector_Name_3' i3
*ISECTOR
DEFINITIONS:
'Sector_Name_1' i1
'Sector_Name_2' i2
Sector names. (16 characters maximum) and their associated numbers

DEFAULTS:
Optional keyword. The full FIELD sector, and MATRIX and FRACTURE sectors if dual
porosity or dual permeability is used, are always defined internally.
CONDITIONS:
These keywords must be in the RESERVOIR DESCRIPTION keyword group.
*SECTORNAMES should appear before *ISECTOR. All array qualifiers and array reading
options are allowed for specifying the required ninjnk values for *ISECTOR. The values
assigned by *ISECTOR should consist of the values that appear in the *SECTORNAMES list.
EXPLANATION:
Sectors are collections of grid blocks that the simulator uses to summarize various quantities.
(See the description of *SECTOR for further information about sectors.) The *ISECTOR
keyword provides an array-based alternative to assigning sector numbers that have been defined
using *SECTORNAMES. Several instances of *ISECTOR can appear in the data set.
Example:
To set sector membership in two sectors named 'LAYER-1' and 'LAYER-2', use the following:
*SECTORNAMES LAYER-1 1 LAYER-2 2
*ISECTOR *IJK 1:5 1:5 1:1 1
1:5 1:5 2:2 2
where it is assumed that the grid is dimensioned 5x5x2. Provided that this not a dual porosity
problem, then three sectors will be defined in the simulation: 'LAYER-1', 'LAYER-2' and
FIELD, where the latter is the default sector consisting of all the active cells.
User's Guide IMEX
Flux Sector (Optional)
*FLUX_SECTORNAMES, *FLUX_ISECTOR
PURPOSE:
The keyword group flux sector introduces special sector definitions which are used to
calculate reservoir flow between these flux sectors. Flow into these sectors and between
these sectors can be reported to the output print file and to the simulation results file as time
series data. Flow can be further broken down into I-direction, J-direction and K-direction
flow components.
*FLUX_SECTORNAMES defines a list of flux sector names and their corresponding flux
sector numbers. *FLUX_ISECTOR assigns these flux sector numbers to cells using standard
array input options. Both keywords can only appear once in the data set.
FORMAT:
*FLUX_SECTORNAMES 'Flux_Sector_Name_1' i1 'Flux_Sector_Name_2' i2
'Flux_Sector_Name_3' i3
*FLUX_ISECTOR
DEFINITIONS:
Flux sector names (16 characters maximum) and their associated numbers
which must be greater than one.
DEFAULTS:
Optional keyword. If a grid block is left undefined via *FLUX_ISECTOR it will be given a
zero value as the sector number.
CONDITIONS:
These keywords must be in the RESERVOIR DESCRIPTION keyword group.
The keyword *FLUX_SECTORNAMES should appear before keyword *FLUX_ISECTOR.
All array qualifiers and array reading options are allowed for specifying the required values for
*FLUX_ISECTOR. The values assigned by *FLUX_ISECTOR should be one of the values
that appear in the *FLUX_SECTORNAMES list.
EXPLANATION:
Flux sectors are collections of grid blocks like normal sectors. Unlike normal sectors, flux
sectors are used to determine flow between these sectors. If two flux sectors share a boundary,
then flow from one flux sector to another will be calculated. In addition, total flow (Influx)
from all surrounding blocks into a flux sector will also be calculated.
User's Guide IMEX
The *FLUX_ISECTOR keyword provides an array-based assignment for flux sector numbers
that have been defined using *FLUX_SECTORNAMES. As the result, each defined grid block
holds a non-zero flux sector number. Undefined grid blocks will hold a zero flux sector number
which will not be included in the flux sector reports. The map of flux sectors thus comes out
without overlaps. This is different from normal sectors which are oriented to inplace
calculations and therefore allow overlaps.
The flux sector does not conduct inplace calculations as the normal SECTOR option does.
However, if necessary, maps of flux sectors can simply be duplicated in SECTOR option via
similar pair of keywords *SECTORNAMES and *ISECTOR.
Example:
To set flux sector membership in four sectors named 'Area-I', 'Area-I', 'Area-III' and 'Area-IV'
use the following:
*FLUX_SECTORNAMES 'Area-I' 1 'Area-II' 2
*FLUX_ISECTOR
*IJK 1:6
1:7
1:6
7:13
1:7
1:6
1:6
8:14
1:6
7:13
8:14
1:6
'Area-III' 3 'Area-IV' 4
1
2
3
4
where it is assumed that the grid is dimensioned 13x14x6.

The output frequency for flux sectors is controlled by *WPRN *SECTOR and *WSRF
*SECTOR. Please also see *OUTPRN *FLUX-SECTOR and *OUTSRF *FLUX-SECTOR for
output options.
User's Guide IMEX
Lease Planes (Optional)
*LEASE
PURPOSE:
*LEASE defines a lease plane which is used to summarize various regional flows.
FORMAT:
*LEASE 'lease'
i1:i2
:
j1:j2
:
k1:k2
:
(*IDIR)
(*JDIR)
(*KDIR)
DEFINITIONS:
'lease'
Lease plane identification name, enclosed in single quotes (16 character
maximum).
i1:i2
Indicates the I beginning and ending indices of the grid region where the
lease plane is to be located.
j1:j2
Indicates the J beginning and ending indices of the grid region where the
k1:k2
Indicates the K beginning and ending indices of the grid region where the
*IDIR
Lease plane is perpendicular to the I direction.
*JDIR
Lease plane is perpendicular to the J direction.
*KDIR
Lease plane is perpendicular to the K direction.
DEFAULTS:
Optional keyword. Default: no lease data.
CONDITIONS:
User's Guide IMEX
EXPLANATION:
Lease planes can be used to summarize and report the flow of fluids across planes or regions
in the reservoir.
Lease planes are composed of collections of subregions of the form "i1:i2 j1:j2 k1:k2" that
are each assigned a flow direction, indicated by one of the *IDIR, *JDIR, *KDIR keywords.
Usually if *IDIR is specified, i1 will be the same i2. The same will hold for the other
directions. If this is not the case, the flow calculation will accumulate the net flow across the
region.
Flows are determined for each subregion by first calculating the flow into each block from its
lower indexed neighbour, as determined from the direction specified for that subregion. These
contributions are then accumulated over the subregion and reported. (Note that flow from a
lower indexed block to its higher indexed neighbour is accumulated with a positive sign.)
Thus, if grid block (I,J,K) lies in a subregion carrying the *IDIR direction indicator, the flow
from block (I-1,J,K) to (I,J,K) is added into the subregion's total. (This calculation is only
done if (I-1,J,K) is a valid block in communication with block (I,J,K).) Similar comments
hold for the other directions.
Example:
*LEASE
'LEASE1'
5:5
1:10
1:3
*IDIR
can be used to calculate the flow across a plane perpendicular to the I direction and separating
all the grid points (5,J,K) from the grid points (4,J,K) where J=1,..,10 and K=1,...,3.
User's Guide IMEX
Special Connections (Optional)
*SCONNECT
PURPOSE:
*SCONNECT allows the definition of flow connections between blocks that would not
otherwise be generated by the simulator. Also, *SCONNECT allows existing flow
connections to be overridden.
FORMAT:
*SCONNECT
1st_cell_desc_1 2nd_cell_desc_1 trms_1

...
DEFINITIONS:
1st_cell_d..., 2nd_cell_d...
Descriptions of active cells. If the cell is on the fundamental grid, then its
cell descriptor consists of 3 integers, i j k, separated by blanks. If the cell is
on a refined grid, then its cell descriptor consists of 3 integers, giving the
location of the fundamental grid parent cell, a slash (/), and 3 integers giving
the location of the cell on the refined grid. An *M or *F can follow
indicating Matrix or Fracture if the simulation is using *DUALPOR or
*DUALPERM. All items should be separate by blanks. Two such
descriptors are required, describing the two cells between which flow is to
occur. The cells cannot be null or pinched out, and they cannot be parent
cells.
trms
The numerical value for the flow transmissibility. The value is expected to
be the result of taking a cross-sectional area for the flow, multiplying it by a
permeability, and dividing it by a centre-to-centre inter-block distance. No
fluid mobility contribution is expected, as it will be supplied internally by the
simulator.
Dimensions are (md-m | md-ft).
DEFAULTS:
Optional keyword. No defaults.
CONDITIONS:
EXPLANATION:
*SCONNECT allows the definition of special flow connections and the alteration of flow
connections generated by the simulator.
If the two descriptors describing a connection in a *SCONNECT line are a pair that the
simulator has NOT already generated connection between, then the connection given by the
User's Guide IMEX
*SCONNECT input is added into the simulator's connection list. These connections are
treated like all the others.
If the two descriptors describing a connection in a *SCONNECT line are a pair that the
simulator has already generated a connection between, then the connection given by the
*SCONNEC input overwrites the simulator-generated connection. This has the effect of
replacing the existing transmissibility value by the "trms" value provided by the
*SCONNECT keyword. This can be used to set transmissibilities to specific values in a way
that avoids printing out lists of inter-block connections and using multipliers to adjust values.
Example:
Connect the first and last blocks in a 1 dimensional problem:
...
*GRID *CARTESIAN 10 1 1
...
*SCONNECT 1 1 1 10 1 1
...
200.
User's Guide IMEX
Fault Array (Optional)
*FAULTARRAY
PURPOSE:
*FAULTARRAY signals the input of an array of binary flags which controls whether
individual block faces are connected using standard connections or fault connections.
ARRAY:
*FAULTARRAY
DEFAULTS:
Standard connections assumed.
CONDITIONS:
This keyword must be in the Reservoir Description keyword group. *FAULTARRAY is not
necessary with corner-point options as the block corners determine connections directly. For
Cartesian grids, GridBuilder will automatically generate this array if faults exist in the
topmost structure map. Care must be taken if the user overrides the automatically generated
values.
All array reading options are valid. The most commonly used array reading subkeyword used
with this option would be *CON.
EXPLANATION:
The *FAULTARRAY values consist of a single integer which defines how all of a grid
blocks connections are made. A standard connection does not account for depth as it
connects two blocks. It only takes layer number into account. In other words, blocks are
connected even if the difference in the two block depths make a physical connection
impossible. A fault connection accounts for depth when creating a connection and would
create connections between blocks which physically touch. This is the default for all cornerpoint options.
The value of *FAULTARRAY controls how each of the four areal connections are made.
The four connections are labeled nilow, nihigh, njlow, njhigh where i refers to the i direction
and j refers to the j direction. Low refers to flow between block i (or j) and i-1 (or j-1). High
refers to flow between block i (or j) and block i+1 (j+1).
The *FAULTARRAY binary integer flag uses the following convention:
nilow, nihigh, njlow, njhigh = 0 if the connection is a standard connection
nilow, nihigh, njlow, njhigh = 1 if the connection is a fault connection
The value of *FAULTARRAY for a block is:
IVAL = nilow + 2*nihigh + 4*njlow + 8*njhigh
Thus if all connections are standard IVAL = 0, and if all connections take into account block
depths (are fault connections), IVAL = 15.
User's Guide IMEX
Use:
**all connections areally are fault connections
*FAULTARRAY *CON 15
**all i connections are fault connections,
**all J connections are standard
*FAULTARRAY *CON
3
Example:
Standard Connections:
i Connections
Fault Connections:
j Connections
1,1 is connected to 2,1
3,2
2,2
3,1
2,1
1,2
1,1
User's Guide IMEX
Compaction/Dilation Rock Type (Optional)
*CROCKTYPE
PURPOSE:
This keyword is used when more than a single rock type is assigned in different regions of a
reservoir or when a table of porosity and permeability multipliers vs. pressure is required.
FORMAT:
*CROCKTYPE
rock_number
DEFINITIONS:
*CROCKTYPE:
Indicates the use of compressibility/compaction/dilation rock type regions in
the reservoir.
rock_number
Rock_number for this compressibility/compaction/dilation rock type.
DEFAULTS:
Optional keyword. No default value.
CONDITIONS:
This keyword must be located in the Reservoir Description section. Its subkeywords have to be
input in formats as explained on consecutive pages. If this keyword does not appear, the
simulator will use a constant rock compressibility and its corresponding reference pressure
which are entered by using the *CPOR and *PRPOR keywords.
EXPLANATION:
This option allows the user to define different rock compressibility options in various parts of
the reservoir. In addition to the standard relation which calculates por(p) as a function of a
constant rock compressibility, a reference pressure and reference porosity (below),
por(p) = por_input*[1+ccpor*(p-cprpor)]
where:
por_input
ccpor
cprpor
p
por(p)
:
:
:
:
:
initial porosity
constant rock compressibility
constant reference pressure
pressure
porosity at pressure p
the user can input a table of porosity and permeability multipliers vs. pressures, which
exhibits reversibly, irreversibly or hysteresis to model compaction processes. When table
options are used, please note that mul_por(p) (below) must be set up to account for both
normal compressibility and compaction/dilation effects.
If the constant rock compressibility and reference pressure option is used, the equation above
is used to calculate the porosity at pressure p. If the table of multipliers vs. pressure option is
used, the relation below is applied to calculate porosity at pressure p:
por(p) = por_input * mul_por(p)
User's Guide IMEX
where:
por_input
mul_por(p)
por(p)
:
:
:
initial porosity
porosity multiplier at a pressure p
It should be noted that the analytical aquifer only uses values of rock compressibility and its
corresponding reference pressure entered with keywords *CPOR and *PRPOR. When the
aquifer option is applied and these values are not given (i.e. omitted from the data file), the
aquifer model will default rock compressibility in all aquifers to 0.0. Hence *CPOR and
*PRPOR must always be included in data sets which use an analytical aquifer, even if the
reservoir compressibility is being obtained from data entered under *CROCKTYPE.
*CROCKTYPE
*CROCKTABE
*CCPOR (*MATRIX)
(*FRACTURE)
*CROCKTAB
*CROCKTABD
*CPRPOR (*MATRIX)
(*FRACTURE)
*CROCKTABU
*CIRREVERS
*CROCKTABH
*CROCKTABC
*CROCKTABR
*CTYPE
Keyword Relation in Rock Compaction/Dilation Model

The figure above shows the relationship among keywords being used in the rock
compaction/dilation model. *CROCKTYPE is the main keyword to signal when the model is
used. Without this keyword, other keywords under it are invalid. Descriptions and usage of
all the compaction/dilation keywords can be seen on consecutive pages. It is also noted that
keywords used for the rock compaction model can not be mixed with those used for the rock
dilation model and vice versa. Compaction and dilation may be used together in different
*CROCKTYPE regions of your reservoir model. However, the two regions must not be in
communication with each other and must be separated by null blocks. Examples of using
those keywords follow the keyword descriptions.
User's Guide IMEX
Compaction/Dilation Rock Compressibility (Optional)

*CCPOR, *CPRPOR
PURPOSE:
*CCPOR signals the input of a rock compressibility value that will be used in a
*CROCKTYPE region of the reservoir defined on the *CTYPE card.
*CPRPOR signals the input of a reference pressure for the corresponding rock
compressibility under the keyword *CCPOR.
These keywords are equivalent to *CPOR and *PRPOR but can be defined in different
regions of the reservoir.
FORMAT:
*CCPOR
*CPRPOR
(*MATRIX)
(*FRACTURE)
(*MATRIX)
(*FRACTURE)
ccpor
cprpor
DEFINITIONS:
ccpor
Pressure dependence of formation porosity; that is, rock compressibility.
(1/kPa | 1/psi | 1/kPa | 1/(kg/cm2))
cprpor
Reference pressure for calculating the effect of rock compressibility
ccpor. (kPa | psi | kPa | kg/cm2)
DEFAULTS:
Optional keywords. No default values.
CONDITIONS:
This keyword must be located in the Reservoir Description section.
The keywords *CCPOR and *CPRPOR must be located under the keyword *CROCKTYPE.
The *MATRIX and *FRACTURE subkeywords are used to indicate the matrix and fracture
dependence.
When using dual porosity models, both matrix and fracture rock compressibilities and
reference pressures should be input. If no fracture rock compressibility and its corresponding
reference pressure are input, their values will be assumed to be the same as those of the
matrix blocks, however; a warning message will be given. If no matrix rock compressibility
and its reference pressure are input, an error message is given and the simulator stops.
EXPLANATION:
These keywords are equivalent to the keywords *CPOR and *PRPOR but can be applied to
individual rock type regions. See definition of *CPOR and *PRPOR in the Reservoir
Description section.
User's Guide IMEX
Compaction Rock Table (Optional)
*CROCKTAB
PURPOSE:
*CROCKTAB indicates a data table for porosity and permeability multipliers versus pressure
lookup. This keyword replaces the keywords *CCPOR and its reference pressure when the
relation between rock compressibility and pressure is not linear or when permeability is
changing with pressure. The porosity and permeability multipliers will be determined by
linear interpolation of the *CROCKTAB table. Please note that *CPOR is still required by
the analytical aquifer model to calculate aquifer compressibilities.
FORMAT:
*CROCKTAB
press por_mult
hor_perm_mult
ver_perm_mult
DEFINITIONS:
press
Minimum pressure is 101 kPa (14.7 psia). (kPa | psi | kPa | kg/cm2).
por_mult
porosity multiplier (dimensionless) = V/Vi
where:
V is the pore volume measured at the pressure press.
Vi is the initial pore volume. It is assumed that the porosity array (*POR) is
defined at the initial pressure.
hor_perm_mult
horizontal permeability multiplier (dimensionless) = permh/permhi
where:
permh is the horizontal permeability measured at the pressure press.
permhi is the input (initial) horizontal permeability. Permhi is *PERMI
when flow in the I direction is considered, and is *PERMJ when flow in the J
direction is considered.
ver_perm_mult
vertical permeability multiplier (dimensionless) = permv/permvi
where:
permv is the vertical permeability measured at the pressure press.
permvi is the input (initial) vertical permeability. Permvi is the array defined
by *PERMK.
User's Guide IMEX
DEFAULTS:
Optional keyword. No default values for press or por_mult.
Default value of ver_perm_mult is hor_perm_mult.
Default value of hor_perm_mult is 1.00.
CONDITIONS:
When the keyword *CROCKTAB appears, some restrictions apply:
1. This keyword must be located in Reservoir Description section.
2. This keyword must be located under the keyword *CROCKTYPE.
3. This keyword can not be used with subdomain, MINC and hybrid grids.
4. When using the keyword *CROCKTAB, keywords *CCPOR and *CPRPOR
should not appear in the same *CROCKTYPE region. *CCPOR and *CPRPOR
are ignored for the *CROCKTYPE which uses *CROCKTAB. Compressibilities
in the reservoir are determined using table lookup. *CPOR may still be required if
an analytical aquifer model is used.
5. The first two columns in the table, pressure and porosity multipliers always exist.
The permeability multipliers (third and fourth columns) are optional.
6. If a current pressure falls outside the range of the table, linear extrapolation is
applied to compute the multipliers at that pressure.
7. Pressures in the table should increase monotonically down the column.
8. The *INT table input option does not apply to this table.
EXPLANATION:
Under the keyword *CROCKTAB, multipliers (porosity and permeability) at one pressure
are computed on the basis of table lookup and linear interpolation methods. Normally the
pressure range of this table should encompass the minimum and maximum pressures
expected during a simulation. The minimum producer bottom hole pressure and maximum
injector injection pressure would normally define this range. If the pressure falls outside the
table range, linear extrapolation is applied to compute values of multipliers outside the range.
When the porosity multiplier is computed, the porosity at that pressure is given by:
por(p) = por_input * mul_por(p)
where:
por_input
mul_por(p)
por(p)
:
:
:
initial input porosity (*PORI)

porosity multiplier at a pressure p
Please note that the compressibility *CPOR is not used to determine por(p), but is required if
the analytical aquifer model is employed.
The use of *CROCKTAB keyword allows the user to input non-linear relationships between
porosity/permeability multipliers and pressures. Without the use of keywords *CROCKTABH
and *IRREVERS (as discussed later), the rock behavior is reversible (a default process in the
simulator) i.e. the relationship between porosity/permeability multipliers vs. pressures always
remains on the main path defined by this single table for this region.
User's Guide IMEX
Compaction Hysteresis Rock Table (Optional)
*CROCKTABH
PURPOSE:
*CROCKTABH signals entry of hysteresis subtables for a rock type.
FORMAT:
*CROCKTABH
press por_mult
hor_perm_mult
ver_perm_mult
DEFINITIONS:
pressure
Minimum pressure is 101 kPa (14.7 psia) (kPa | psi | kPa | kg/cm2).
por_mult
porosity multiplier (dimensionless) = V/Vi
where:
V is the pore volume measured at a the pressure press.
hor_perm_mult
horizontal permeability multiplier (dimensionless) = permh/permhi
where:
Permhi is the input initial horizontal permeability. Permhi is *PERMI when
flow in the I direction is considered, and is *PERMJ when flow in the J
ver_perm_mult
vertical permeability multiplier (dimensionless) = permv/permvi
where:
permvi is the input initial vertical permeability. Permvi is the array defined
by *PERMK.
DEFAULTS:
Optional keyword. No default values for press or por_mult.
Default value of hor_perm_mult is 1.0.
CONDITIONS:
When the keyword *CROCKTABH appears, the following restrictions apply:
User's Guide IMEX
1. This keyword must be located in the Reservoir Description section.

2. This keyword is optional and only used under the keyword *CROCKTAB
3. This keyword can not be combined with the keyword *CIRREVERS.
4. There must be at least two subtables (two hysteresis curves).
5. The first two columns in the table, pressure and porosity multipliers must always
exist. The permeability multipliers (third and fourth columns) are optional.
6. The pressure and multipliers in the first row of each subtable of the corresponding
rock compaction hysteresis curve must lie on the main path curve. Further, to avoid
potential numerical problems, the first row of each subtable must be identical to a
corresponding row on the main path from which the hysteresis path starts.
7. Pressures in each subtable must increase monotonically down the column.
8. First pressures in subtables must increase monotonically from the first subtable to
the last subtable.
9. To do interpolation based on the data subtables of hysteresis curves the pressure on
the main path (called a rebound pressure) at which a hysteresis phenomenon starts
occurring must lie between a pressure range defined by the first pressures in the
first and last subtables. If the rebound pressure is less than the first pressure in the
first subtable or is greater than the first pressure in the last subtable, the rock
behaves as if it was on the main path and there is no hysteresis occurring.
EXPLANATION:
This keyword is designed to model porosity and permeability multipliers versus pressure
behavior that exhibits hysteresis effects. In the explanation below, porosity behavior will be
described but the behavior also applies to permeability multipliers.
While pressure decreases, porosity multiplier vs. pressure behavior is defined by the main
path (non-hysteresis) input on the *CROCKTAB table. When the pressure increases while
on the main path, a rebound pressure is set. The porosity multiplier vs. pressure behavior is
now free to follow a path defined by *CROCKTABH relationships.
If the rebound pressure lies within the range defined by the first pressures on the first and last
*CROCKTABH tables, the porosity multiplier vs. pressure relationship is defined by
interpolating bounding *CROCKTABH curves to obtain a hysteresis porosity multiplier vs.
pressure relationship for the specific rebound pressure. The porosity multiplier is calculated
using the rebound pressure and bounding *CROCKTABH tables until the pressure is reduced
below the rebound pressure. When this occurs, the *CROCKTAB table is used to define the
porosity multipliers vs. pressures relationship.
If the rebound pressure is less than the first pressure in the first *CROCKTABH table or
greater than the last pressure in the first *CROCKTABH table, the *CROCKTAB table is
used to define the porosity multiplier vs. pressure behavior. Hysteresis is only allowed to
occur if the rebound pressure lies between the first pressures in the first and last
*CROCKTABH tables.
When the *CTROCKTAB table is quite nonlinear it is important to define a sufficient
number of hysteresis branches using *CROCKTABH. If the *CROCKTABH data is sparse it
User's Guide IMEX
is possible that there will be a porosity discontinuity at the rebound pressure. This can be
minimized by increasing the number of *CROCKTABH tables and can be eliminated entirely
by defining a *CROCKTABH table for every pressure point in the main table.
Initial
pressure
Multiplier
CROCKTABH Table 2
MAIN PATH
CROCKTABH Table 1
HYSTERESIS PATH
Pressure
Rebound Pressure
Hysteresis Process for a Rock Type with Two *CROCKTABH Tables
User's Guide IMEX
Treat *CCPOR / *CROCKTAB / *CROCKTABH as Depletion

Parameters / Tables (Optional)
*DEPLETION
PURPOSE:
The *DEPLETION option makes the input reference porosity (*POR) the actual input
porosity and makes the porosity a function of pressure depletion rather than pressure.
The *DEPLETION keyword in the data set indicates that all *CROCKTYPES using
*CCPOR (*CPROR), *CROCKTAB and *CROCKTABH use the Depletion Option.
FORMAT:
*DEPLETION
DEFINITIONS:
*DEPLETION
Occurrence of this keyword turns the Depletion option on for all
*CROCKTYPES using *CCPOR (*CPROR), *CROCKTAB and
*CROCKTABH.
DEFAULTS:
CONDITIONS:
This keyword must be located in the Reservoir Description keyword group, just after the first
*CROCKTYPE keyword.
Dilation Tables (*CROCKTABE/D/U/C/R) do not make use of this option.
Any grid blocks not explicitly defined in a *CTYPE region will also use the Depletion option
(using *CPOR and *PRPOR), if (1) *DEPLETION is read in and if (2) at least one block is
assigned to a valid *CROCKTYPE which can use the Depletion option.
The Depletion option will work when a *CROCKTAB table uses the *CIRREVERS option.
EXPLANATION:
The Depletion option is in reality two separate options.
In the Case of models where *CCPOR and *CPRPOR are used, the value of *CPRPOR for
each block is overwritten by the initial reservoir pressure in each block (Prs0(iblk)). This has
the effect of changing the porosity reference pressure of each block to be the initial block
pressure.
Hence *POR input becomes the actual measured porosity of the block, not the porosity at a
reference pressure *CPRORP (actually every block has its own CPRPOR which has been set
equal to Prs0(iblk)).
User's Guide IMEX
In the Case of Models where *CROCKTAB and *CROCKTABH tables are used
(*CROCKTAB and the *CIRREVERS option may also be used), the use of the Depletion
option operates in a completely different manner. When the Depletion option is used with
*CROCKTAB, the compaction tables are interpolated with respect to P* = Pshift(itype) +
Prs(iblk) - Prs0(iblk).
Where Prs(iblk) is the current grid block pressure, Prs0(iblk) is the initial grid block pressure,
itype is the *CROCKTYPE of grid block iblk and Pshift(itype) is the pressure in the
*CROCKTAB table of *CROCKTYPE itype where the Porosity Multiplier (por_mult) is
equal to or closest to 1.00.
IMEX will interpolate within the *CROCKTAB table to find the best value of Pshift.
As P* rather than Prs(iblk) is used in the interpolation procedure to find the value of
por_mult, this procedure initially calculates every grid blocks por_mult to be the value at
Pshift(itype). This causes the input reference porosity (*POR) to be the measured porosity (as
por_mult = 1.00).
The value of por_mult for each grid block calculated as depletion occurs, using this
procedure, is calculated in the tables using a pressure offset from Pshift(itype) by Prs(iblk)
Prs0(iblk).
Therefore, it is how much pressure in a block has depleted from the initial pressure which
determines por_mult and hence compressibility, rather than the actual pressure itself.
When the Depletion option is used, IMEX will printout the value of Pshift(itype) it
determines for each itype containing a *CROCKTAB table as well as diagnostics on how it
arrived at the value chosen. To make the choice (by interpolation) of Pshift as
straightforward as possible, the user should use values of por_mult which increase with
pressure and which start at a por_mult value below 1.0 and end at values either equal to or
greater than 1.0. Por_mult should only increase with pressure.
User's Guide IMEX
Compaction Irreversibility Flag (Optional)
*CIRREVERS
PURPOSE:
This keyword indicates the behavior of a rock type to be irreversible.
FORMAT:
*CIRREVERS
DEFAULTS:
If this keyword does not appear, the rock behavior is reversible.
CONDITIONS:
This keyword must be located in Reservoir Description section.
This keyword is optional and located under the keyword *CROCKTAB (CROCKTAB data
entries follow *CIRREVERS)
This keyword can not be combined with the keyword *CROCKTABH. It should not be input
when the keyword *CROCKTABH appears.
EXPLANATION:
This keyword is designed to allow the rock behavior to be irreversible. As long as pressure
decreases the main path (defined by *CROCKTAB) is followed. When the pressure
increases while on the main path defined by the *CROCKTAB table a rebound pressure is set
at this point. The multiplier versus pressure behavior no longer is defined by the main path
but by a rebound path where multipliers are kept constant with pressure. On the horizontal
path (or rebound path), if the pressure is higher than the rebound pressure, the multipliers are
kept at the values defined by the rebound pressure. When the pressure drops below the
rebound pressure, the rock behavior will again resume the multiplier versus pressure behavior
defined by the main path until the next increase in pressure, at which a new rebound pressure
and path is set.
Multiplier
Initial
pressure
MAIN PATH
REBOUND PATH
Pressure
Rebound Pressure
Irreversible Process
User's Guide IMEX
Dilation Rock Table in Elastic Zone (Optional)
*CROCKTABE
PURPOSE:
This keyword is used for the rock dilation model only. The keyword indicates that rock
behavior is acting within an elastic zone when pressure in the rock is below its yield point
pressure. If an analytical aquifer is applied, *CPOR and *PRPOR are still required.
ARRAY:
*CROCKTABE
press por_mult
hor_perm_mult
ver_perm_mult
DEFINITIONS:
press
Minimum pressure is 101 kPa (14.7 psia). (kPa | psi | kPa | kg/cm2).
Maximum pressure is a yield point pressure of a rock type.
por_mult
Porosity multiplier (dimensionless) = V/Vi
where:
V is the pore volume measured at the pressure press
hor_perm_mult
Horizontal permeability multiplier (dimensionless) = permh/permhi
where:
ver_perm_mult
Vertical permeability multiplier (dimensionless) = permv/permvi
where:
by *PERMK.
User's Guide IMEX
DEFAULTS:
Optional keyword. No default value for press or por_mult .
Default value of hor_perm_mult is one.
If a value of hor_per_mult is less than zero, it is replaced by a value of one.
If a value of ver_per_mult is less than zero, it is replaced by a value of hor_perm_mult on the
same row.
CONDITIONS:
When the keyword *CROCKTABE appears, some conditions apply:
2. This keyword must be located under the keyword *CROCKTYPE.
3. This keyword can not be used with subdomain, MINC and hybrid grids.
4. When keyword *CROCKTABE is used, keywords *CCPOR and *CPRPOR
should not appear under the same *CROCKTYPE. Keywords *CPOR and
PRPOR may still be required if an analytical aquifer model is used.
5. The first two columns in the table such as pressure and porosity multiplier always
6. Pressure in the table should increase monotonically down the column.
7. Minimum pressure in the table must be greater than zero.
8. There is only one keyword *CROCKTABE for one rock type.
9. This keyword can not be combined with any other keywords related to the rock
compaction model.
EXPLANATION:
Under the keyword *CROCKTABE, multipliers (porosity and permeability) at one pressure
are computed on the basis of table lookup and linear interpolation methods. The maximum
pressure in the table must be a yield point pressure of that rock type. When pressure in a rock
is lower than its yield point pressure, the rock behaves elasticity or reversibility on that path.
User's Guide IMEX
Dilation Rock Table in the Plastic Zone (Optional)

*CROCKTABD
PURPOSE:
This keyword is used for the rock dilation model only. It is optionally entered and located
after keyword *CROCKTABE. The keyword indicates that rock behavior is acting in the
plastic region when pressure in the rock is higher than its yield point pressure.
FORMAT:
*CROCKTABD
press por_mult
hor_perm_mult
ver_perm_mult
DEFINITIONS:
press
Minimum pressure is the yield point pressure of a rock type (kPa | psi | kPa |
kg/cm2) which is equal to maximum pressure in the *CROCKTABE table.
por_mult
where:
hor_perm_mult
where:
ver_perm_mult
where:
by *PERMK.
User's Guide IMEX
DEFAULTS:
same row.
CONDITIONS:
When the keyword *CROCKTABD appears, some conditions apply:
2. This keyword is optional and located under the keyword *CROCKTABE.
5. First pressure in the table must be the same as that of the last pressure in the
*CROCKTABE table for the same rock type.
6. There is only one keyword *CROCKTABD for one rock type.
compaction model.
EXPLANATION:
Under the keyword *CROCKTABD, multipliers (porosity and permeability) at one pressure are
computed on the basis of table lookup and linear interpolation methods. The minimum pressure
in the table must be a yield point pressure of that rock type and that pressure must be the same
as that of the maximum pressure in the *CROCKTABE table. When pressure in a rock is
continuously increasing, the rock behaves plasticity on the path given by the *CROCKTABD
table. When pressure in the rock goes beyond the maximum pressure in the table, an
extrapolation method would be applied. However, it is strongly recommended that the
maximum pressure in the table should be large enough so that extrapolation does not occur.
User's Guide IMEX
Dilation Rock Table in Unloading Zone (Optional)
*CROCKTABU
PURPOSE:
This keyword is used for the rock dilation model only. It must be entered after
1. keyword *CROCKTABD, if it is the first *CROCKTABU table in a
*CROCKTYPE, after
2. *CROCKTABU or
3. after *CROCKTABR. The keyword indicates that rock behavior is irreversible in
the plastic region when pressure in the rock is decreasing.
FORMAT:
*CROCKTABU
press por_mult
hor_perm_mult
ver_perm_mult
DEFINITIONS:
press
Maximum pressure is one of the pressures in the *CROCKTABD table (kPa |
psi | kPa | kg/cm2). Minimum pressure must be greater than zero.
por_mult
where:
hor_perm_mult
where:
ver_perm_mult
where:
by *PERMK.
User's Guide IMEX
DEFAULTS:
same row.
CONDITIONS:
This keyword must be entered and located under the keyword *CROCKTABE.
1. When the keyword *CROCKTABU appears, some conditions apply:
4. Pressure in the table should decrease monotonically down the column.
5. First pressure in each table must be the same as one of pressures in the
*CROCKTABD table for the same rock type.
6. There must be at least two *CROCKTABU tables for one rock type.
7. Each subsequent *CROCKTABU table must start at a pressure higher than the last.
8. The minimum pressure in the table must be greater than zero.
compaction model.
EXPLANATION:
Under the keyword *CROCKTABU, multipliers (porosity and permeability) at one pressure
are computed on the basis of table lookup and linear interpolation methods applied to two
unloading paths given by two *CROCKTABU tables for one rock type. On the dilation path,
when pressure in the rock suddenly decreases, the rock behavior would follow an arbitrary
unloading path that lies between two given unloading tables. If the pressure keeps on
decreasing, the rock follows the unloading path until it reaches the minimum pressure in the
table. When pressure in the rock increases, the rock will follow the same path until it reaches
the dilation path. When on the unloading path and the pressure increases above the pressure
at the intersection between unloading path and dilation path the rock will follow the dilation
path. It is valid to model dilation using one *CROCKTABE table, one *CROCKTABD table
and two or more *CROCKTABU tables. Recompaction and reloading may also be modeled
by using the *CROCKTABC and *CROCKTABR tables which will be described on their
own keyword.
User's Guide IMEX
If pressure in the rock decreases but it is less than the first pressure in the first
*CROCKTABU table or greater than the last pressure in the first *CROCKTABU table, the
rock behavior is still following the dilation path and the CROCKTABD table is used to
compute multiplier vs. pressure. The rock only follows an arbitrary unloading path when
pressure in the rock lies between the first pressure in the first and last *CROCKTABU tables.
User's Guide IMEX
Dilation Rock Table in Recompacting Zone (Optional)

*CROCKTABC
PURPOSE:
This keyword is used for the rock dilation model only and is optionally entered after keyword
*CROCKTABU. The keyword indicates that rock behavior is recompaction in the unloading
zone when pressure in the rock is decreasing below the minimum pressure in the unloading
*CROCKTABU table.
FORMAT:
*CROCKTABC
press por_mult
hor_perm_mult
ver_perm_mult
DEFINITIONS:
press
Maximum pressure is equal to minimum pressure in *CROCKTABU table
(kPa | psi | kPa | kg/cm2).
por_mult
where:
hor_perm_mult
where:
ver_perm_mult
where:
by *PERMK.
User's Guide IMEX
DEFAULTS:
same row.
CONDITIONS:
When the keyword *CROCKTABC appears, some conditions apply:
2. The keyword is optional. It should be entered and located under the keyword
*CROCKTABU.
4. Pressure in the table should decrease monotonically down the column.
*CROCKTABU table for the same rock type.
6. There must be at least two *CROCKTABC tables for one rock type corresponding
to at least two *CROCKTABU tables.
compaction model.
EXPLANATION:
Under the keyword *CROCKTABC, multipliers (porosity and permeability) at one pressure
recompaction paths given by two *CROCKTABC tables for one rock type. On the unloading
path, when pressure in the rock is below the minimum pressure in the *CROCKTABU table,
the rock behavior would follow an arbitrary recompaction path that lies between the two
given *CRACKTABC tables. If the pressure keeps on decreasing, the rock follows the path
until it reaches the minimum pressure in the table.
The use of the *CROCKTABU card, *CROCKTABC card and the *CROCKTABR card
defines a complete Unloading Recompaction Reloading path. When *CROCKTABC and
*CROCKTABR are used with *CROCKTABU tables, each *CROCKTABU table
(unloading curve) must be followed by a *CROCKTABC table (recompaction curve) and a
*CROCKTABR table (reloading curve).
User's Guide IMEX
Dilation Rock Table in Reloading Zone (Optional)
*CROCKTABR
PURPOSE:
This keyword is used for the rock dilation model only. It must be entered and located after
existing keyword *CROCKTABC. The keyword indicates that rock behavior is in a
reloading state when pressure in the rock is increasing in the recompaction zone.
FORMAT:
*CROCKTABR
press por_mult
hor_perm_mult
ver_perm_mult
DEFINITIONS:
press
Minimum pressure in this table is equal to minimum pressure in the
*CROCKTABC table (kPa | psi | kPa | kg/cm2). Maximum pressure in the
table is pressure of the intersection point between this reloading path and the
dilation path.
por_mult
where:
hor_perm_mult
where:
ver_perm_mult
where:
by *PERMK.
DEFAULTS:
User's Guide IMEX

same row.
CONDITIONS:
This keyword must be entered and located under the keyword *CROCKTABC. This
keyword exists only when *CROCKTABC exists.
When the keyword *CROCKTABR appears, some conditions apply:
*CROCKTABC table for the same rock type.
5. The pressure in the last row of the table must lie on the dilation path and must be
lower than the first pressure on the *CROCKTABU in its Unloading
Recompaction Reloading path.
6. There must be at least two *CROCKTABR tables for one rock type.
compaction model.
EXPLANATION:
Under the keyword *CROCKTABR, multipliers (porosity and permeability) at one pressure
unloading paths given by two *CROCKTABR tables for one rock type. On the recompaction
path, when pressure in the rock increases, the rock behavior would follow an arbitrary
reloading path that lies between the two given *CROCKTABR tables. If the pressure keeps
on increasing, the rock follows the path until it reaches the dilation path. The rock behavior
will jump over the dilation plasticity path if pressure in the rock is larger than pressure at the
intersection between the reloading path and the dilation path. When pressure in the rock
decreases, the rock will follow the reloading path until it meets the recompaction path. If
pressure in the rock is less than pressure at the intersection point between the recompaction
path and the reloading path, the rock will follow the recompaction path.
User's Guide IMEX
GENERAL RESTRICTIONS
1. For one rock type, it is noted that if a closed loop contains keywords
*CROCKTABU, *CROCKTABC and *CROCKTABR, other loops in that rock
type must also include those keywords (see Model 1). If an open loop contains one
keyword *CROCKTABU, other loops in that rock type also contain that keyword
only (see Model 2). A mix of model 1 and model 2 in one rock type is not
allowed.
2. In one reservoir, either the compaction model is used or the dilation model is used
but both models can not be applied simultaneously.
3. It is strongly recommended that the compaction model is used when initial pressure
in a reservoir is high and decreasing due to production. Whereas the dilation model
is used when initial pressure in the reservoir is low and increasing due to injection.
An improper selection of the models for a project may affect results.
MODEL 1: A Closed Loop Dilation Model
User's Guide IMEX
MODEL 2: An Open Loop Dilation Model

Example1: A sample data of closed loop dilation model is
given as below:
*CROCKTYPE 1
*CROCKTABE
** pressure
14.7
100
200
300
400
*CROCKTABD
400
500
600
800
1000
1200
1400
1600
2000
por.mult
1
1.008566484
1.018702746
1.028940879
1.039281906
1.039281906
1.114638347
1.195458746
1.375104874
1.581747107
1.819442253
2.092856753
2.407358287
3.185247519
hor.trans.mult
1
1
1
1
1
1
1
1
1
1
1
1
1
1
ver.trans.mult
1
1
1
1
1
1
1
1
1
1
1
1
1
1
User's Guide IMEX
*CROCKTABU
** pressure
1000
900
800
700
600
500
400
300
200
*CROCKTABC
200
100
90
50
0
*CROCKTABR
0
100
200
300
400
488.2
*CROCKTABU
1600
1400
1200
1000
800
600
400
200
*CROCKTABC
200
100
90
50
0
*CROCKTABR
0
100
200
300
400
500
614
User's Guide IMEX
por.mult
1.581747107
1.56600846
1.550426416
1.534999415
1.519725916
1.50460439
1.489633326
1.474811227
1.46013661
hor.trans.mult
1
1
1
1
1
1
1
1
1
ver.trans.mult
1
1
1
1
1
1
1
1
1
1.46013661
1.239838343
1.219725828
1.142485901
1.0527776
1
1
1
1
1
1
1
1
1
1
1.0527776
1.063358191
1.074045118
1.084839451
1.095742269
1.105446145
1
1
1
1
1
1
1
1
1
1
1
1
2.407358287
2.3596894
2.312964419
2.267164655
2.222271787
2.178267857
2.135135264
2.092856753
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2.092856753
1.541460769
1.495036423
1.322905928
1.135338718
1
1
1
1
1
1
1
1
1
1
1.135338718
1.146749062
1.158274082
1.16991493
1.18167277
1.193548779
1.207233087
1
1
1
1
1
1
1
1
1
1
1
1
1
1
3.5
Porosity Multiplier
3
2.5
2
1.5
1
0
500
1000
1500
2000
Pressure (psi)
Example2: A sample data of open loop dilation model is

given as below:
*CROCKTYPE 1
*CROCKTABE
** pressure
14.7
100
150
200
250
300
350
*CROCKTABD
400
500
600
800
1000
1200
1400
1600
2000
por.mult
1
1.00856648
1.01362194
1.01870275
1.02380901
1.02894088
1.03409847
1.03928191
hor.trans.mult
ver.trans.mult
1.03928191
1.11463835
1.19545875
1.37510487
1.58174711
1.81944225
2.09285675
2.40735829
3.18524752
User's Guide IMEX
*CROCKTABU
** pressure
1000
800
700
600
500
400
300
200
100
50
1
0.5
por.mult
1.58174711
1.55042642
1.53499942
1.51972592
1.50460439
1.48963333
1.47481123
1.46013661
1.44560801
1.43839801
1.4313671
1.43129553
*CROCKTABU
1200
1000
800
600
500
400
300
200
100
50
1
0.5
1.81944225
1.78341488
1.7481009
1.71348619
1.69643671
1.67955689
1.66284502
1.64629943
1.62991848
1.62178923
1.6138619
1.61378121
*CROCKTABU
1400
1200
1000
800
600
400
300
200
100
50
1
0.5
2.09285675
2.05141541
2.01079467
1.97097826
1.93195028
1.8936951
1.87485252
1.85619743
1.83772795
1.82856225
1.81962421
1.81953323
User's Guide IMEX
hor.trans.mult
ver.trans.mult
*CROCKTABU
1600
1400
1200
1000
800
600
300
200
100
50
1
0.5
2.40735829
2.3596894
2.31296442
2.26716465
2.22227179
2.17826786
2.11389031
2.09285675
2.07203248
2.06169818
2.05162056
2.05151799
3.5
Porosity Multiplier
3
2.5
2
1.5
1
0.5
0
0
500
1000
1500
2000
2500
Pressure (psi)
User's Guide IMEX
Compaction/Dilation Rock Region (Optional)
*CTYPE
PURPOSE:
*CTYPE assigns compressibility/compaction/dilation rock types to reservoir grid blocks.
ARRAY:
*CTYPE
DEFAULTS:
Optional keyword. The default is *CTYPE *CON 1 (all blocks use
compressibility/compaction/dilation rock type 1).
CONDITIONS:
This keyword must be located in the Reservoir Description section.
Number of *CTYPE regions must be less than or equal to number of *CROCKTYPE.
EXPLANATION:
This keyword assigns rock properties defined by *CROCKTYPE keywords to specific grid
block ranges.
*CTYPE indicates the input of an array of grid values. All array reading subkeywords are
valid. See the section on Array Reading Options for more details.
EXAMPLES
The following are examples of a reservoir consisting of 50X50X4 grid blocks.
Example 1:
*CROCKTYPE
1
*CROCKTAB: table of rock compaction data for rock type 1
**pressure
porosity
perm.
perm.
multiplier
multiplier (hoziz.)
multiplier (vert.)
1000
0.96
0.98
1.2
2000
0.99
0.99
1.3
3000
1.0
1.0
1.4
4000
1.01
1.0
1.5
*CROCKTYPE
2
*CROCKTAB
*CIRREVERS
**pressure
porosity multiplier
1000
0.96
2000
0.97
3000
0.98
4000
0.99
*CROCKTYPE
3
*CROCKTAB
User's Guide IMEX
**pressure
porosity multiplier
perm. multiplier
1000
0.96
0.98
2000
0.99
0.99
3000
1.0
1.0
4000
1.01
1.0
*CROCKTABH
1000
0.96
0.98
2000
0.97
0.985
3000
0.98
0.99
4000
0.99
0.995
*CROCKTABH
2000
0.99
0.99
3000
0.995
0.995
4000
0.998
0.998
*CROCKTABH
3000
1.0
1.0
4000
1.01
1.0
*CROCKTYPE
4
*CCPOR
3.0e-6
*CPRPOR
14.7
*CTYPE
*KVAR 1 2 3 4
** assign different rock type
** to each k layer
In the above example, columns 3 and 4 do not appear in the data table of rock type 2. In this
case the permeability multipliers in vertical and horizontal directions are default to be one.
Moreover, column 4 does not appear in rock type 3, in such a case, values of permeability
multipliers are assumed to be the same in both vertical and horizontal directions.
Example 2 (using the same model 50X50X4):
*CPOR
3.0e-6
** This
*PRPOR 14.7
** This
*CROCKTYPE
1
*CCPOR
*MATRIX
*CCPOR
*FRACTURE
*CPRPOR
*MATRIX
*CPRPOR
*FRACTURE
*CTYPE
*MATRIX
*IJK
*CTYPE
*FRACTURE *IJK
keyword was discussed in the manual.

keyword was discussed in the manual.
2.5e-6
2.0e-5
14.7
14.7
1:50
1:50
1:50
1:50
3:4
3:4
1
1
In example 2 above, k layers 1 and 2 were not explicitly defined to belong in *CTYPE
region 1. Thus they will use the global values of *CPOR and *PRPOR in their porosity
calculation for both matrix and fracture. Layers 3 and 4 will use the separate matrix and
fracture values of *CCPOR and *CPRPOR defined in *CTYPE 1.
The *CPOR and *PRPOR values are used for layer 1 and 2 because *CTYPE was found, and
layers 1 and 2 were omitted from any *CTYPE region.
If *CTYPE would have been defaulted in example 2 (i.e. *CTYPE keyword omitted), all
blocks, including those in layer 1 and 2, would have been assigned to *CTYPE 1. In this case
*CPOR and *PRPOR would only be used in the analytical aquifer calculation (if required).
User's Guide IMEX
Example 3 (using two rock models)

This example shows how two rock models can be used in one simulation.
Data for rock dilation model:
*CROCKTYPE 1
*CROCKTABE
** pressure
14.7
100
150
200
250
300
350
400
*CROCKTABD
400
500
600
800
1000
1200
1300
1400
1500
*CROCKTABU
600
550
500
450
400
300
250
200
*CROCKTABC
200
150.0003
100.0005
50.0008
0.001
*CROCKTABR
0.001
85.1742
170.3474
255.5205
340.6937
425.8669
User's Guide IMEX
por.mult
1
1.0085665
1.0136219
1.0187027
1.023809
1.0289409
1.0340985
1.0392819
hor.trans.mult
1
1.0160131
1.0255614
1.035232
1.0450268
1.0549481
1.064998
1.0751787
ver.trans.mult
1
1
1
1
1
1
1
1
1.0392819
1.1146383
1.1954587
1.3751049
1.5817471
1.8194423
1.9513667
2.0928568
2.244606
1.0751787
1.2327992
1.424045
1.9523725
2.8069041
4.3221774
5.56285
7.4048444
10.3099231
1
1
1
1
1
1
1
1
1
1.1954587
1.1894964
1.1835637
1.1776607
1.1717871
1.1601276
1.1543414
1.1485841
1.424045
1.409072
1.3943166
1.3797745
1.3654418
1.3373887
1.3236606
1.3101264
1
1
1
1
1
1
1
1
1.1485841
1.1133907
1.0792756
1.0462059
1.0141494
1.3101264
1.2300356
1.1565273
1.088905
1.0265618
1
1
1
1
1
1.0141494
1.0228241
1.031573
1.0403968
1.049296
1.0582714
1.0265618
1.0431316
1.0600671
1.077379
1.0950784
1.1131769
1
1
1
1
1
1
*CROCKTABU
1000
800
700
600
500
400
300
200
*CROCKTABC
200
150.0003
100.0005
50.0008
0.001
*CROCKTABR
0.001
97.6353
195.2696
292.9039
390.5383
488.1726
*CROCKTABU
1200
1000
800
600
500
400
300
200
*CROCKTABC
200
150.0003
100.0005
50.0008
0.001
*CROCKTABR
0.001
105.0777
210.1545
315.2312
420.308
525.3847
1.5817471
1.5504264
1.5349994
1.5197259
1.5046044
1.4896333
1.4748112
1.4601366
2.8069041
2.6555066
2.5841702
2.5155461
2.4494987
2.3859011
2.3246338
2.2655846
1
1
1
1
1
1
1
1
1.4601366
1.3454868
1.2398394
1.1424873
1.0527793
2.2655846
1.853819
1.5401945
1.2959291
1.1020726
1
1
1
1
1
1.0527793
1.0631084
1.0735388
1.0840716
1.0947077
1.1054481
1.1020726
1.1230351
1.1445465
1.1666257
1.1892928
1.2125686
1
1
1
1
1
1
1.8194423
1.7834149
1.7481009
1.7134862
1.6964367
1.6795569
1.662845
1.6462994
4.3221774
4.0419251
3.7873389
3.5553807
3.4470591
3.3434488
3.244281
3.1493054
1
1
1
1
1
1
1
1
1.6462994
1.4804376
1.3312861
1.1971613
1.0765494
3.1493054
2.3476953
1.8083243
1.4283473
1.1508205
1
1
1
1
1
.0765494
1.0879211
1.0994129
1.111026
1.1227619
1.1346217
1.1508205
1.174786
1.1994412
1.2248124
1.2509275
1.2778154
1
1
1
1
1
1
User's Guide IMEX
*CROCKTABU
1400
1300
1200
1000
800
600
300
200
*CROCKTABC
200
150.0003
100.0005
50.0008
0.001
*CROCKTABR
0.001
113.438
226.8749
340.3119
453.7489
567.1858
2.0928568
2.0720325
2.0514154
2.0107947
1.9709783
1.9319503
1.8748525
1.8561974
7.4048444
7.091086
6.7965223
6.258994
5.7816897
5.3559411
4.7986518
4.6316616
1
1
1
1
1
1
1
1
1.8561974
1.6300455
1.431447
1.257045
1.1038914
4.6316616
3.0589838
2.1545655
1.587556
1.2091707
1
1
1
1
1
1.1038914
1.1164849
1.1292221
1.1421046
1.155134
1.1683121
1.2091707
1.2368996
1.265511
1.2950425
1.3255336
1.3570259
1
1
1
1
1
1
Data for rock compaction model:

*CROCKTYPE 2
*CROCKTAB
** press
por.mult
14.7
1.0
100.0
1.000256
200.0
1.000556
500.0
1.001456
1000.
1.002956
2000.
1.005956
5000.
1.014956
7000.
1.020956
10000.
1.029956
20000.
1.059956
*CROCKTABH
1000.0
1.002956
2000.0
1.004956
3000.0
1.006956
4000.0
1.008956
5000.0
1.010956
6000.0
1.012956
7000.0
1.014956
8000.0
1.016956
User's Guide IMEX
**
**
perm.mult
1.0
**
1.001 **
1.002 **
1.005 **
1.01
**
1.02
**
1.05
1.07
1.1
1.2
rock type number 2

data table for main path
(hor)
press : pressure
por.mult : porosity multiplier
perm.mult (hor): hor. perm. multiplier.
Vertical permeability multiplier is the
same as horizontal perm. multiplier when
column 4 does not appear.
1.01
1.02
1.03
1.04
1.05
1.06
1.07
1.08
*CROCKTABH
2000.0
1.005956
3000.0
1.007956
4000.0
1.009956
5000.0
1.011956
6000.0
1.013956
7000.0
1.015956
8000.0
1.017956
*CROCKTABH
5000.0
1.014956
6000.0
1.016956
7000.0
1.018956
8000.0
1.020956
1.02
1.03
1.04
1.05
1.06
1.07
1.08
1.05
1.06
1.07
1.08
Two reservoirs connected by null blocks are shown in the figure below:
4
null
blocks
1
1
Assume there is only one layer K=1, the above two rock models are assigned in the field as
follows:
*CTYPE
or
*CTYPE
*IJK
1:3
5:8
1:4
1:4
1
1
1
2
*IJK
1:4
4:8
1:4
1:4
1
1
1
2
The above application is valid since each reservoir will used one rock model only.
However, invalid cases can be seen as follows:
*CTYPE
or
*CTYPE
*IJK
1:3
3:8
1:4
1:4
1
1
1
2
*IJK
1:8
1:8
1:2
3:4
1
1
1
2
As a general rule, if two adjacent active blocks have different rock models to each other, the
simulator will stop running.
User's Guide IMEX
Dispersion Coefficients (Conditional)
*DISPI, *DISPJ, *DISPK
PURPOSE:
*DISPI indicates input of polymer dispersion coefficients for the I direction (m2/day | ft2/day
| cm2/min).
*DISPJ indicates input of polymer dispersion coefficients for the J direction (m2/day | ft2/day
| cm2/min).
*DISPK indicates input of polymer dispersion coefficients for the K direction (m2/day | ft2/day
| cm2/min).
ARRAY:
*DISPI
*DISPJ
*DISPK
Please note in the 2002.10 release the time unit of the dispersion coefficient changed from
seconds to the time unit which is consistent with the value of *INUNIT. The user must alter
data accordingly.
DEFINITIONS:
*DISPI
Indicates these are dispersion coefficients for the I direction.
*DISPJ
Indicates these are dispersion coefficients for the J direction.
*DISPK
Indicates these are dispersion coefficients for the K direction.
DEFAULTS:
Conditional keyword. By default, all dispersion coefficients are 0.0
CONDITIONS:
This keyword must be in the Reservoir Description keyword group. This keyword is only
used with *MODEL *POLY or *MODEL *POLYOW.
EXPLANATION:
The dispersion coefficient keywords are arrays. All array reading option subkeywords are
valid. *EQUALSI can be used to set *DISPJ and *DISPK after *DISPI has been input. See
the section on Array Reading Options for more details.
User's Guide IMEX
Example:
Enter the dispersion coefficients for the polymer, when it is a constant value for all directions
and all fracture blocks.
*DISPI *FRACTURE *CON 0.25
*DISPJ *EQUALSI
*DISPK *EQUALSI
Physical dispersion may be calculated using the equation:

disp =
v
molard
+ alpha *
L
theta
where disp is the dispersion coefficient in the i, j, or k directions, molard is the molar
diffusion coefficient, L is the tortuosity, alpha is the dispersion parameter in the i, j, and k
direction, and v is the velocity of the water phase.
Please refer to Appendix E for more information about dispersion coefficients.
The acceptable range of values for the dispersion coefficients is:
min
SI
m2/day
0.0
1.0E+20
Field
ft2/day
0.0
1.076E+21
Lab
cm2/min
0.0
1.0E+24
User's Guide IMEX
Fault Transmissibilities (Optional)
*TRANSF
PURPOSE:
Adjusts transmissibilities on a fault basis.
FORMAT:
*TRANSF
<pair or single>
Fault_Name
<pair or single>
fault_trans_mult
DEFINITIONS:
*TRANS
Keyword introducing the fault name, multiplier and fault description.
Fault_Name
A name for this fault (which must be contained within single quotes).
fault_trans_mult
Transmissibility multiplier for the connections across this fault. Multipliers
of this type apply cumulatively to previously applied multipliers.
<pair or single>
Identifiers for the connections that are to make up this fault. The identifier
can either be of the pair or single variety. A pair identifier looks like:
i1 j1 k1 [*IDIR or *JDIR or *KDIR] i2 j2 k2
which refers to an existing connection between cells, while a single
identifier looks like:
i1 j1 k1 [*IDIR- or *IDIR+ or *JDIR- or *JDIR+ or *KDIR- or *KDIR+]
which refers to all connections on a certain cell face. Multiple mixed pair
or single identifiers can follow a *TRANSF line.
DEFAULTS:
Optional keyword.
CONDITIONS:
This keyword, if present, must be in the RESERVOIR DESCRIPTION keyword group.
Pair-type identifiers should refer to pairs of cells that actually are connected in the grid
direction specified by the given direction identifier (*IDIR, *JDIR or *KDIR). Refined grid
cells cannot be referenced in the description of a pair or single identifier.
User's Guide IMEX
EXPLANATION:
This keyword allows the imposition of a single multiplier to a group of connections. The
group of connections can be thought of as corresponding to a single fault, and the multiplier
as a way of adjusting the sealing properties of this fault. If the descriptor is of pair type, the
multiplier will be applied to an (existing) connection. If the descriptor is of single type, the
multiplier will be applied to all connections that the cell has to other cells on a specified face.
The face for the latter is identified using *IDIR-/+, *JDIR-/+ or *KDIR-/+ descriptors. The
- identifier refers to the face crossed by moving from the cell to its neighbour along the grid
direction corresponding to decreasing the appropriate I, J or K index, and the + identifier to
increasing the appropriate I, J or K index. The multiplier can be 0 if desired, which will
eliminate connections.
For non-isothermal simulations, this multiplier applies only to fluid flow, not to heat flow.
User's Guide IMEX
Fluid Model (Required)
*MODEL
PURPOSE:
*MODEL signals the input of the fluid component model to use for the simulation.
FORMAT:
*MODEL
(*BLACKOIL)
(*OILWATER)
(*MISCG)
(*MISNCG)
(*POLY)
(*POLYOW)
(*API-INT)
(*API-INTOW)
(*GASWATER)
(*GASWATER_WITH_CONDENSATE)
(*BLACKOIL_SEAWATER)
(*OILWATER_SEAWATER)
(*VOLATILE_OIL)
DEFINITIONS:
*BLACKOIL
Use a black-oil model, modelling flow of oil, water, and gas.
*OILWATER
Use a two phase, oil and water model, with no modelling of free gas, or
variation in solution gas.
*MISCG
Use a pseudo-miscible model, with chase gas. Assumes solution gas always
remains in solution.
*MISNCG
Use a pseudo-miscible model with no chase gas. Assumes gas injected (if
any) has the same composition as solution gas.
User's Guide IMEX
Component Properties 427
*POLY
Use a polymer model, modelling the flow of oil, water, gas, and polymer.
*POLYOW
Use a polymer model, with no gas flow or variation in solution gas.
*API-INT
Use an API tracking model, handling three phases (water, oil and gas) and
four components (water, gas and two oil components). The two oil
components flow only in the oil phase while the gas component can be
present both in the oil and gas phases. See: *PVTAPI, *APIGRAD, *API,
*APIT, *APIGRAV.
*API-INTOW
Use an API tracking model, with no gas flow or variation in solution gas.
See: *PVTAPI, *APIGRAD, *API, *APIT, *APIGRAV.
*GASWATER
Use a two phase gas and water model, with no oil phase modelling. Only two
equations are solved for simultaneously. Gas PVT properties are entered using
the *PVTG option. Krow, Krog, So, Pb, Co and Cvo are not input. PCGW, if
required, is input on the *SWT table and is properly handled in both
*BLOCK_CENTER and *DEPTH_AVE *TRANZONE initialization methods.
*GASWATER_WITH_CONDENSATE
Uses an extension of the *GASWATER option. Oil (as condensate) can
initially exist in the gas phase either in a saturated or undersaturated state.
Condensate can be produced at surface or can drop out in the reservoir (and
possibly be produced as a liquid). Condensate, Gas, and Water equations are
solved for simultaneously.
Saturated gas PVT properties are entered using the PVTCOND table.
Undersaturated GAS PVT properties are entered using the
*EGUST/*BGUST/*ZGUST and *VGUST keywords. Condensate
properties are entered using oil phase keywords (e.g. *DENSITY *OIL).
The keywords associated with the use of this option are: *PVTCOND,
*EGUST, *VGUST, *PDEW, *PDEWT, *PSPLIT, *OUTPRN/*OUTSRF
GRID (*DPP | *RV), *NORM *PDW and *MAXCHANGE *PDW
*BLACKOIL_SEAWATER
Use a seawater injection model, modelling the flow of oil, water, gas, and
seawater. This model tracks seawater production and reduces well
productivity accordingly if used with the Scale Buildup/Well Damage model.
428 Component Properties
User's Guide IMEX
*OILWATER_SEAWATER
Use a seawater injection model, with no gas flow or variation in solution gas.
This model tracks seawater production and reduces well productivity
accordingly if used with the Scale Buildup/Well Damage model.
The *BLACKOIL_SEAWATER and OILWATER_SEAWATER models simply track
seawater flow in the reservoir. Seawater viscosity is modeled separately from formation
water viscosity (See *SVISC). The Seawater models are designed to be coupled to the
Scale Deposition/Damage model defined using the *SCLDPS, *SCLDMG, *SCLTBLWELL, *SCLTBL-LAYER, *SCLRMV-WELL and *SCLRMV-LAYER keywords. This
allows scale buildup and well damage to be both a function of water produced and seawater
volume fraction in the produced water.
The Scale Deposition/Damage model can be used without the Seawater model, however;
when this is done well damage is only a function of formation water produced.
*VOLATILE_OIL
Uses an extension of the *BLACKOIL option. Gas can exist in solution in
the oil phase and oil can exist in the gas phase. This option is similar to the
*GASWATER_WITH_CONDENSATE. It is more flexible in that the gas
phase need not always exist. This allows the model to be used in situations
where oil can be above its bubble point.
The model initializes like a *BLACKOIL model and uses either PVTCOND or PVTVO in
place of the *BLACKOIL models PVT table. Both *PB and *PDEW need to be defined
with this model. In addition *DWOC and *DGOC need to be defined. It is possible to use
*BGUST and *VGUST to describe undersaturated gas conditions and *BOT and *VOT
tables (or *CO, *COT and *CVO) to describe undersaturated oil conditions when using the
*PVTCOND saturated table. When using the *PVTVO table *BOT and *VOT tables can be
used to define undersaturated oil conditions.
DEFAULTS:
Required keyword. No default.
CONDITIONS:
This keyword must be at the start of the Component Property keyword group, following
immediately after the Reservoir Description keyword group.
EXPLANATION:
Thirteen different fluid models are available for use in IMEX. Only one option may be
entered. When the *OILWATER keyword is used, the model solves only two equations per
grid block. The gas phase is absent and consequently is not modelled. When the
*GASWATER keyword is used, the model solves only two equations per grid block. The oil
phase is absent and consequently is not modelled.
When the chase gas option, *MISCG, is used the solution gas always remains in solution.
The solution GOR is fixed and is input by the user using the *GORINT keyword. The bubble
point pressure vs. Rs curve in the PVT table belongs to the chase gas. So the total amount of
dissolved gas in oil consists of two parts. The solution gas given by *GORINT and the
dissolved chase gas given by the Rs vs. bubble point pressure curve, or
User's Guide IMEX
Total_Rs = GORINT + Chase_gas_Rs

PVT DATA:
PVT tables may be entered directly using the *PVT keyword. Alternatively, a differential
liberation table may be entered using the *DIFLIB keyword. If the *DIFLIB option is used
then the *BUBBLE, *BOBF and *RSIF keywords must also be present in this data section.
One of *PVT or *DIFLIB must be present in the data set.
If the *API-INT or *API-INTOW options are used then the PVT properties MUST be
specified using the keywords *PVTAPI and *APIGRAD.
If the *GASWATER option is used then the PVT properties MUST be specified using the
*PVTG keyword.
If the *GASWATER_WITH_CONDENSATE option is used then the PVT properties MUST
be specified using the *PVTCOND keyword.
If the *VOLATILE_OIL option is used then the PVT properties MUST be specified using
either the *PVTCOND or *PVTVO keyword. For multiple PVT region models, all regions
must use one type of table (*PVTCOND or *PVTVO).
User's Guide IMEX
Reservoir Temperature (Optional)
*TRES
PURPOSE:
*TRES indicates the input of reservoir temperature.
FORMAT:
*TRES value
DEFINITIONS:
value
Value of reservoir temperature. (F | C | C)
DEFAULTS:
Required keyword. No default values.
CONDITIONS:
These keywords must be in the Component Property keyword group. This keyword is
required if the *ZG option will be used to input the gas compressibility in the PVT table.
The value of *TRES entered acts on PVT tables which follow its occurrence. Multiple
*TRES keywords may be entered. PVT table temperature is determined from the last defined
value of *TRES.
EXPLANATION:
The reservoir temperature specified here is used to convert the Z-factor of gas Zg to the gas
expansion factor Eg as follows:
It is assumed that the stock tank conditions are 14.7 psia, (101.325 kPa | 1.03 kg/cm2) and 60
F, (15.56C).
Then in field units:
Eg = 198.588 * P/ZT (SCF/Res.BBL)
with P in psia
and T in R
and in SI and Lab units:
Eg = 2.84935 * P/ZT (Std.m3/Res.m3)
with P in kPa
and T in K
Example:
*TRES
180.0
The acceptable range of values for reservoir temperature is:
min
max
User's Guide IMEX
SI
C
-100
1000
Field
F
-100
1000
Lab
C
-100
1000
Oil and Gas PVT Table (Optional)
*PVT
PURPOSE:
*PVT indicates start of the oil and gas PVT table.
TABLE:
*PVT *EG set_number
p
:
rs
:
bo
:
eg
:
viso
:
visg
:
(co)
:
(srftn)
:
bg
:
viso
:
visg
:
(co)
:
(srftn)
:
zg
:
viso
:
visg
:
(co)
:
(srftn)
:
-or*PVT *BG set_number

p
:
rs
:
bo
:
-or*PVT *ZG set_number

p
:
rs
:
bo
:
DEFINITIONS:
set_number
Set_number for this particular PVT region. This is the number used with
*PTYPE to assign PVT tables to grid blocks.
*EG
Keyword indicating that the gas expansion factor will be used. If all *EG,
*BG and *ZG are not present in the data set then this is the default.
*BG
Keyword indicating that the gas formation volume factor will be used instead
of the gas expansion factor.
*ZG
Keyword indicating that the gas compressibility factor will be used instead of
the gas expansion factor. The reservoir temperature TRES must be specified
before the *PVT keyword in order to use this option.
p
Pressure (kPa | psi | kPa | kg/cm2).
User's Guide IMEX
rs
Solution gas-oil ratio (m3/m3 | scf/STB | cm3/cm3) for saturated oil at pressure
p. When the chase gas option is used this is the Rs of the dissolved chase gas.
bo
Formation volume factor (m3/m3 | RB/STB | cm3/cm3) for saturated oil at
pressure p.
eg
Gas expansion factor (m3/m3 | scf/RB | cm3/cm3) at pressure p. When the
chase gas option is used this is the chase gas eg.
bg
Gas formation volume factor (m3/m3 | RB/scf | cm3/cm3) at pressure p.
When the chase gas option is used this is the chase gas bg.
zg
Gas compressibility factor (dimensionless) at pressure p. When the chase gas
option is used this is the chase gas zg.
viso
Viscosity (mPa-s | cp | cp) of saturated oil at pressure.
visg
Gas viscosity (mPa-s | cp | cp) at pressure p. When the chase gas option is
used this is the chase gas viscosity.
srftn
Gas-oil interfacial tension. It must be monotonically decreasing and greater
than zero (dyne/cm).
co
Oil compressibility. It must remain constant or be monotonically increasing.
Oil compressibility can alternatively be entered using the *CO or *COT
keywords (1/kPa | 1/psi | 1/kPa | 1/kg/cm2).
DEFAULTS:
Optional keyword. No default values. One of *PVT, *DIFLIB *PVTAPI *PVTG,
*PVTCOND or *PVTVO must be present in the data set.
CONDITIONS:
This keyword must be in the Component Property keyword group.
Table entries must be in order of increasing pressure and should extend to above the initial
bubble point pressure. Surface tension is optional.
User's Guide IMEX
EXPLANATION:
In general, this table must contain at least two entries, and may contain more entries if desired.
When the two phase (oil and water) option is used, there are two ways of specifying the
initial solution gas-oil ratio. In the first method the first entry of the table must be at the
saturation pressure and one entry will suffice. The Rs value from the first entry will be set as
the initial solution GOR. In the second method the solution gas-oil ratio may be specified
using the *PB or *PBT keywords. The producing GOR from each well layer is obtained
from the input bubble point pressure (from *PB or *PBT input for each block) If the *PB or
*PBT keywords are not found then the first method is used.
Typical PVT data curves used by this simulator are given by the solid lines in Figure 7(a),
(b), (c), (d), and (e) in Appendix D. Note that they are the curves that would be obtained if
there were unlimited gas present (saturated conditions).
The dotted lines in Figure 7(a), (b), and (d) indicate the usual curves obtained when the
pressure exceeds the bubble point pressure (PB). The latter curves are assumed to be linear.
The gradient of the Rs curve above the bubble point is equal to zero. IMEX automatically uses
a variable substitution technique to model undersaturated conditions (see theoretical outline in
Appendix A).
The minimum pressure in the PVT table should be below any pressure constraint specified on
a production well in the Well Data section. Normally a value of zero would be chosen. The
maximum pressure in the table should be comfortably above any well injection pressure
specified in the Well Data section.
The formation volume factor and viscosity of water are calculated from functional
relationships.
STC means stock tank conditions. (14.7 psia and 60 F | 101.3 kPa and 288.15 K | 1.03
kg/cm2 and 288.15 K)
rs =
(volume of dissolved gas at STC )

(volume of oil at STC )
bo =
(volume of oil at reservoir conditions )

eg =
(volume of
gas at STC )
(volume of gas at reservoir conditions )
bg =

(volume of gas at STC )
The default is *EG. Use *BG to enter values for bg instead of eg.
Set_numbers are used to assign different PVT tables to grid blocks in the reservoir. If only
one PVT region is used, the set_number may be omitted since the default is 1.
User's Guide IMEX
If there is more than one PVT region, the first PVT table must have a set_number of one and
set_numbers must increase consecutively for subsequent table sets. Set_numbers are
assigned to grid blocks with the *PTYPE keyword.
Example:
*PVT 1
** p
14.7
264.7
514.7
1014.7
2014.7
2514.7
3014.7
4014.7
5014.7
9014.7
** field units
rs
bo
eg
1.0
1.062
6.0
90.5
1.15
82.692
180.0
1.207
159.388
371.0
1.295
312.793
636.0
1.435
619.579
775.0
1.5
772.798
930.0
1.565
925.926
1270.0
1.695 1233.046
1618.0
1.827 1540.832
2984.0
2.36
2590.674
viso
1.04
0.975
0.91
0.83
0.69
0.64
0.594
0.51
0.449
0.203
visg
0.0080
0.0096
0.0112
0.0140
0.0189
0.0208
0.0228
0.0268
0.0309
0.0470
The acceptable range of values for pressure is:
min
max
SI
kPa
0.0
1.0E+10
Field
psi
0.0
1.45E+9
Lab
kPa
0.0
1.0E+10
Mod. SI
kg/cm2
0.0
1.01E+8
The acceptable range of values for the solution gas-oil ratios is:
min
max
SI
m3/m3
0.0
1.0E+3
Field
scf/STB
0.0
5,617
Lab
cm3/cm3
0.0
1.0E+3
The acceptable range of values for the formation volume factors is:
min
max
SI
m3/m3
0.5
5.0
Field
scf/STB
0.5
5.0
Lab
cm3/cm3
0.5
5.0
The acceptable range of values for the gas expansion factors is:
min
max
SI
m3/m3
1.0E-20
1.0E+3
Field
scf/STB
5.617E-20
5617
Lab
cm3/cm3
1.0E-20
1.0E+3
The acceptable range of values for the oil viscosity is:
min
max
User's Guide IMEX
SI
mPa*sec
.0095
86,400.
Field
cp
.0095
86,400.
Lab
mPa*sec
.0095
86,400.
The acceptable range of values for the gas viscosity is:
min
max
SI
mPa*sec
9.5E-5
86,400.
Field
cp
9.5E-5
86,400.
Lab
mPa*sec
9.5E-5
86,400.
The acceptable range of values for the surface tension is:
min
SI
dyne/cm
-1.1E-20
Field
lbf/ft
-6.852E-25
Lab
dyne/cm
-1.1E-20
The acceptable range of values for the oil compressibility is:
min
max
SI
1/kPa
0.0
1.0E-3
Field
1/psia
0.0
6.89E-3
Lab
1/Pa
0.0
1.0E-3
Mod. SI
1/(kg/cm2)
0.0
1.0E-2
A positive total hydrocarbon compressibility check is incorporated in IMEX. This check

ensures physically meaningful results when gas comes out of solution. Fundamentally the
check ensures that when gas comes out of solution, the resulting oil-gas mixture takes up less
volume, or:
Bg*dRS/dP - dBo/dP > 0.0 must be true,
or in terms of table entries i and i+1 we check that both:
Bg(i)*(Rs(i+1)-Rs(i))/(P(i+1)-P(i))
and
Bg(i+1)*(Rs(i+1)-Rs(i))/(P(i+1)-P(i))
are greater than or equal to
(Bo(i+1)-Bo(i))/(P(i+1)-P(i))
User's Guide IMEX
Oil and Gas Differential Liberation Table (Optional)
*DIFLIB
PURPOSE:
*DIFLIB indicates start of the oil and gas differential liberation table.
TABLE:
*DIFLIB
p
:
*EG
*BUBBLE
*BOBF
*RSIF
rsd
:
-or*DIFLIB
p
:
eg
:
*BG
*BUBBLE
*BOBF
*RSIF
rsd
:
-or*DIFLIB
p
:
bod
:
set_number
pressure
bobf
rsif
bod
:
bod
:
visg
:
(co)
:
(srftn)
:
visg
:
(co)
:
(srftn)
:
visg
:
(co)
:
(srftn)
:
set_number
pressure
bobf
rsif
bg
:
*ZG
*BUBBLE
*BOBF
*RSIF
rsd
:
viso
:
viso
:
set_number
pressure
bobf
rsif
zg
:
viso
:
DEFINITIONS:
set_number
*EG
*BG
User's Guide IMEX
*ZG
before the *PVT keyword in order to use this option.
*BUBBLE
Indicates the bubble point pressure in the *DIFLIB table used for converting
*DIFLIB data to *PVT data.
pressure
This is the bubble point pressure which will be used to convert the differential
liberation table to the standard PVT table (kPa | psi | kPa | kg/cm2).
*BOBF
Is the oil formation volume factor of the bubble point oil determined by
flashing the oil through the appropriate surface separators and is measured
relative to the stock tank oil volume.
bobf
This is the oil formation volume factor of the bubble point oil determined by
flashing the oil through the appropriate surface separators and is measured
relative to the stock tank oil volume (m3/m3 | RB/STB | cm3/cm3).
*RSIF
Is the solution gas-oil-ratio of the bubble- point oil, determined by flashing
the oil through the appropriate surface separators, and is measured relative to
the oil volume at standard conditions.
rsif
This is the solution gas-oil-ratio of the bubble- point oil, determined by
flashing the oil through the appropriate surface separators, and is measured
relative to the oil volume at standard conditions (m3/m3 | scf/STB | cm3/cm3).
p
rsd
Differential solution gas-oil-ratio (m3/m3 | scf/STB | cm3/cm3) for saturated
oil at pressure p. When the chase gas option is used this is the Rs of the
dissolved chase gas.
bod
Differential formation volume factor (m3/m3 | RB/STB | cm3/cm3) for
saturated oil at pressure p.
User's Guide IMEX
eg
Gas expansion factor (m3/m3 | scf/RB | cm3/cm3) at pressure p. When the
chase gas option is used this is the chase gas eg.
bg
When the chase gas option is used this is the chase gas bg.
zg
Gas compressibility factor (dimensionless) at pressure p. When the chase
gas option is used this is the chase gas zg.
viso
visg
srftn
co
Oil compressibility can alternatively be entered using the *CO or *COT
keywords (1/kPa | 1/PSI | 1/kPa | 1/kg/cm2).
DEFAULTS:
Optional keyword. No default values. One of *PVT, *DIFLIB *PVTAPI *PVTG,

CONDITIONS:

EXPLANATION:
initial solution gas-oil ratio. In the first method the first entry of the table must be at the
saturation pressure and one entry will suffice. The Rs value from the first entry will be set as
the initial solution GOR. In the second method the solution gas-oil ratio may be specified
using the *PB or *PBT keywords. The producing GOR from each well layer is obtained
from the input bubble point pressure (from *PB or *PBT input for each block) If the *PB or
*PBT keywords are not found then the first method is used.
User's Guide IMEX
The gradient of the Rs curve above the bubble point is equal to zero. IMEX automatically
uses a variable substitution technique to model undersaturated conditions (see theoretical
outline in Appendix A).
The formation volume factor and viscosity of water are calculated from functional relationships.
volume of dissolved gas at STC determined
by the differential liberation experiment
rsd =
volume of residual oil in the
differential liberation experiment
bod =
(volume of
oil at differential liberation conditions )

volume of residual oil in the
differential liberation experiment
eg =
(volume of
gas at STC )
bg =

CONVERSION OF *DIFLIB DATA TO *PVT DATA
Differential liberation data must be converted to PVT data. This done by following the
method of L.P. Dake, "Fundamental of Reservoir Engineering", pp. 66-69, First Edition,
Elsevier Scientific Publishing Company.
User's Guide IMEX
bo
rs
=
=
bod * (bobf/bobd)
rsif - (rsid - rsd)*(bobf/bobd)
bo
the corrected PVT Bo;
bod
bod entry in the above *DIFLIB table;
bobf =
bobf entry using the *BOBF keyword;
bobd =
bod entry in the above *DIFLIB table at the pressure specified by the
*BUBBLE keyword;
rs
the corrected PVT Rs;
rsif
rsif entry using the *RSIF keyword;
rsid
rsd entry in the above *DIFLIB table at the pressure specified by the
*BUBBLE keyword; and
rsd
rsd entry in the above *DIFLIB table.
where:
Set_numbers are used to assign different *DIFLIB tables to grid blocks in the reservoir. If
only one PVT region is used, the set_number may be omitted since the default is 1.
If there is more than one PVT region, the first DIFLIB table must have a set_number of one
and set_numbers must increase consecutively for subsequent table sets. Set_numbers are
Example:
*DIFLIB 1
** field units
*BUBBLE 2014.7
*RSIF
600.0
*BOBF
1.35
** p
rsd
bod
eg
14.7
1.0
1.062
6.0
264.7
90.5
1.15
82.692
514.7
180.0
1.207
159.388
1014.7
371.0
1.295
312.793
2014.7
636.0
1.435
619.579
2514.7
775.0
1.5
772.798
3014.7
930.0
1.565
925.926
4014.7 1270.0
1.695 1233.046
5014.7 1618.0
1.827 1540.832
9014.7 2984.0
2.36
2590.674
viso
1.04
0.975
0.91
0.83
0.69
0.64
0.594
0.51
0.449
0.203
visg
0.0080
0.0096
0.0112
0.0140
0.0189
0.0208
0.0228
0.0268
0.0309
0.0470
min
max
User's Guide IMEX
SI
kPa
0.0
1.0E+10
Field
psi
0.0
1.45E+9
Lab
kPa
0.0
1.0E+10
Mod. SI
kg/cm2
0.0
1.01E+8
min
max
SI
m3/m3
0.0
1.0E+3
Field
scf/STB
0.0
5,617
Lab
cm3/cm3
0.0
1.0E+3
min
max
SI
m3/m3
0.5
5.0
Field
Rb/STB
0.5
5.0
Lab
cm3/cm3
0.5
5.0
min
max
SI
m3/m3
1.0E-20
1.0E+3
Field
scf/STB
5.617E-20
5617
Lab
cm3/cm3
1.0E-20
1.0E+3
min
max
SI
mPa*sec
.0095
86,400.
Field
cp
.0095
86,400.
Lab
mPa*sec
.0095
86,400.
min
max
SI
mPa*sec
9.5E-5
86,400.
Field
cp
9.5E-5
86,400.
Lab
mPa*sec
9.5E-5
86,400.
min
SI
dyne/cm
-1.1E-20
Field
lbf/ft
-6.852E-25
Lab
dyne/cm
-1.1E-20
min
max
SI
1/kPa
0.0
1.0E-3
Field
1/psia
0.0
6.89E-3
Lab
1/Pa
0.0
1.0E-3
Mod. SI
1/(kg/cm2)
0.0
1.0E-2
User's Guide IMEX

volume, or:
or in terms of table entries i and i+1 we check that both:
Bg(i)*(Rs(i+1)-Rs(i))/(P(i+1)-P(i))
and
Bg(i+1)*(Rs(i+1)-Rs(i))/(P(i+1)-P(i))
are greater than or equal to
(Bo(i+1)-Bo(i))/(P(i+1)-P(i))
User's Guide IMEX
Gas PVT Table for API Model (Optional)
*PVTAPI
PURPOSE:
*PVTAPI indicates start of PVT information for a PVT region and also indicates start of gas
PVT table. The *APIGRAD Oil PVT Tables must follow the *PVTAPI gas table for each
PVT region.
TABLE:
*PVTAPI *EG set_number

p
:
eg
:
visg
:
-or*PVTAPI *BG set_number

p
:
bg
:
visg
:
-or*PVTAPI *ZG set_number

p
:
zg
:
visg
:
DEFINITIONS:
set_number
*PTYPE to assign PVT tables to grid blocks
*EG
*BG
*ZG
before the *PVTAPI keyword in order to use this option.
User's Guide IMEX
p
eg
Gas expansion factor (m3/m3 | scf/RB | cm3/cm3) at pressure p.
bg
zg
Gas compressibility factor (dimensionless) at pressure p.
visg
DEFAULTS:
Optional keyword. No default values. *PVTAPI must be present in the data set when the
*API-INT model is used.
CONDITIONS:

bubble point pressure.
EXPLANATION:
In general, this table must contain at least two entries, and may contain more entries if
desired. Typical gas PVT data curves used by this simulator are given by the Figure 7(c) and
(e) in Appendix D.
relationships. STC means stock tank conditions. (14.7 psia and 60 F | 101.3 kPa and 288.15
K | 1.03 kg/cm2 and 288.15 K)
eg =
(volume of
gas at STC )
bg =

User's Guide IMEX
Example: (SI units)
*PVTAPI 1
** gas table
** p
eg
100.
1.000
1000.
5.000
1500.
300.000
2000.
600.000
2500.
800.000
3000.
1000.000
3500.
1100.000
4000.
1200.000
4500.
1500.000
visg
0.020
0.010
0.015
0.019
0.021
0.023
0.026
0.028
0.031
min
max
SI
kPa
0.0
1.0E+10
Field
psi
0.0
1.45E+9
Lab
kPa
0.0
1.0E+10
Mod. SI
kg/cm2
0.0
1.01E+8
min
max
SI
m3/m3
1.0E-20
1.0E+3
Field
scf/STB
5.617
5617
Lab
cm3/cm3
1.0E-20
1.0E+3
min
max
SI
mPa*sec
9.5E-5
86,400.
Field
cp
9.5E-5
86,400.
Lab
mPa*sec
9.5E-5
86,400.
User's Guide IMEX
Oil PVT Tables for API Model (Optional)
*APIGRAD
PURPOSE:
*APIGRAD indicates the start of an oil PVT table. The first *APIGRAD card for a PVT
region must follow immediately after the *PVTAPI table for that region.
The stc_oil_dens of the first *APIGRAD table in a PVT region is the density of the heavy oil
component. The stc_oil_dens of the last *APIGRAD table in a PVT region is the density of
the light component.
TABLE:
*APIGRAD stc_oil_dens
p
:
rs
:
bo
:
viso
:
co
:
(srftn)
:
DEFINITIONS:
stc_oil_dens
oil phase density at STC (kg/m3 | lbm/ft3 | g/cm3) for which the PVT table is
valid.
p
rs
Solution gas-oil ratio (m3/m3 | scf/STB | cm3/cm3) for saturated oil at
pressure p.
bo
pressure p.
viso
srftn
co
Oil compressibility must be entered in the *APIGRAD table when using API
tracking option. Other entry methods (*CO *COT) are disabled.
DEFAULTS:
Optional keyword. No default values. *APIGRAD must be present in the data set when the
*API-INT model is used.
User's Guide IMEX
CONDITIONS:
This keyword must be in the Component Property keyword group, following the *PVTAPI
keyword. At least two *APIGRAD must be specified for each *PVTAPI keyword. The
*APIGRAD tables must be entered in order of decreasing STC oil density (increasing API
gravity).
Even though the flexibility of multiple PVT regions is allowed in the API tracking option, the
user should model density changes across connected portions of the reservoir using the API
tracking option explicitly (i.e. all APIGRAD tables the same in each PVT region).
The multiple PVT regions should be used to model multiple equilibrium regions or multiple
PVT regions which are not connected and so cannot communicate.
EXPLANATION:
A minimum of two *APIGRAD tables have to be specified for each PVT region, although the
user can specify as many as he or she wants to better describe properties dependencies with
the oil composition. The stc_oil_dens for the first table corresponds to STC oil density for
the heavy component, while stc_oil_dens for the last table corresponds to STC oil density for
the light component. When using the API tracking option, the only way to input the oil
compressibility, co, is through the *APIGRAD keyword.
Also, the user is not allowed to use the *DENSITY *OIL keyword with the API tracking
option, the input of STC oil density is only allowed through *APIGRAD, since this quantity
can vary within the reservoir according to oil composition.
When using the API tracking option, the oil is assumed to be comprised of two components, a
light component and a heavy component.
The volume fraction of light oil at STC is defined as Vl, the volume fraction of heavy oil is
Vh = (1-Vl). The density of pure light oil at STC is denol(STC), and the density of pure
heavy oil is denoh(STC). The density denoh(STC) is the density of the oil mixture at Vl=0.0.
This is the density (stc_oil_dens) associated with the first *APIGRAD table. The last
*APIGRAD table must be at defined for pure light oil (Vl=1.0), The value of stc_oil_dens for
the final *APIGRAD table is therefore denol(STC).
The density of the mixture deno(STC) is:
deno(STC) = denoh(STC)*Vh + denol(STC)*Vl
The solution gas ratio of the mixture Rs is:
Rs = Rs(mix),
where Rs(mix) is obtained by interpolation of *APIGRAD tables whose oil mixture densities
are just lower and just higher than the density defined by the equation for deno(STC) above.
User's Guide IMEX
If only two APIGRAD tables are used for a PVT region, this is equivalent to:
Rs(mix) = Rsh*Vh + Rsl*Vl
where Rsh is the Rs of the heavy component and Rsl is the Rs of the light component.
Oil mixture formation volume factor oil compressibility (co) and viscosity are derived in the
same manner.
Using the *API keyword or the *APIT keyword light oil volume fraction (at STC) is
initialized. Using the *PVTAPI keyword a gas PVT property table is input for each PVT
region (note: separate gas and liquid tables are used in this option).
The *APIGRAD tables follow the *PVTAPI gas table for each PVT region. Each APIGRAD
table is associated with a Oil mixture density for a PVT region. Oil mixture densities must
entered from heaviest (Vl=0) to lightest (Vl=1.0). Multiple PVT Regions may be used with
this option.
initial solution gas-oil ratio. In the first method, *PB is not specified, then the first entry of
the table must be at the saturation pressure and one entry will suffice. The Rs value from the
first entry will be set as the solution GOR.
In the second method for specifying initial gas-oil ratio in two phase problems (oil and
water), the solution gas-oil ratio is specified using the *PB keyword. The solution gas ratio
for producing well layers will be the Rs value of the block the layer is in. *PB will not vary
as it is not being solved for. In addition, the user must ensure the reservoir pressure never
drops below *PB, as gas is assumed never to come out of solution.
Care must be taken that the desired method is employed. A *PB card in the data will override
using the RS from the first PVT table entry.
Typical oil PVT data curves used by this simulator are given by the solid lines in Figure 7(a),
(b), and (d) in Appendix D. Note that they are the curves that would be obtained if there were
unlimited gas present (saturated conditions).
The gradient of the Rs curve above the bubble point is equal to zero. IMEX automatically uses
Appendix A).
User's Guide IMEX
rs =

bo =

Example: (Field units)

*APIGRAD 60.0
** oil pvt table for
** p
rs
bo
1000.
70.
1.03
1500. 110.
1.04
2000. 150.
1.05
2500. 190.
1.06
3000. 230.
1.07
3500. 260.
1.08
4000. 310.
1.09
4500. 350.
1.10
5000. 390.
1.11
*APIGRAD 45.0
** oil pvt table for
** p
rs
bo
1000.
200.
1.10
1500.
400.
1.15
2000.
500.
1.20
2500.
700.
1.25
3000.
800.
1.30
3500.
900.
1.35
4000. 1000.
1.40
4500. 1200.
1.45
5000. 1300.
1.50
the heavy component

viso
co
160.0 1.0e-5
125.0 1.0e-5
110.0 1.0e-5
100.0 1.0e-5
90.0 1.0e-5
70.0 1.0e-5
60.0 1.0e-5
50.0 1.0e-5
40.0 1.0e-5
the light component
viso
co
3.0
5.0e-6
2.0
5.0e-6
1.0
5.0e-6
.90
5.0e-6
.85
5.0e-6
.80
5.0e-6
.78
5.0e-6
.75
5.0e-6
.70
5.0e-6
min
max
SI
kPa
0.0
1.0E+10
Field
psi
0.0
1.45E+9
Lab
kPa
0.0
1.0E+10
Mod. SI
kg/cm2
0.0
1.01E+8
min
max
SI
m3/m3
0.0
1.0E+3
Field
scf/STB
0.0
5,617
Lab
cm3/cm3
0.0
1.0E+3
User's Guide IMEX
The acceptable range of values for the formation volume factor is:
min
max
SI
m3/m3
0.5
5.0
Field
Rb/STB
0.5
5.0
Lab
cm3/cm3
0.5
5.0
min
max
SI
mPa*sec
.0095
86,400.
Field
cp
.0095
86,400.
Lab
mPa*sec
.0095
86,400.
min
SI
dyne/cm
-1.1E-20
Field
lbf/ft
-6.852E-25
Lab
dyne/cm
-1.1E-20
min
max
SI
1/kPa
0.0
1.0E-3
Field
1/psia
0.0
6.89E-3
Lab
1/Pa
0.0
1.0E-3
Mod. SI
1/(kg/cm2)
0.0
1.0E-2

volume, or:
or in terms of table entries i and i+1:
Bgint*(Rs(i+1)-Rs(i))/(P(i+1)-P(i)),
is greater than or equal to
(Bo(i+1)-Bo(i))/(P(i+1)-P(i))
where Bgint is obtained from the interpolation of the gas table entered using *PVTAPI
keyword at the pressure P(i).
User's Guide IMEX
Gas PVT Table for GASWATER Model (Optional)
*PVTG
PURPOSE:
*PVTG indicates start of PVT information for a PVT region and also indicates start of gas
PVTG table.
TABLE:
*PVTG *EG set_number

p
:
eg
:
visg
:
(srftn)
:
visg
:
(srftn)
:
visg
:
(srftn)
:
-or*PVTG *BG set_number

p
:
bg
:
-or*PVTG *ZG set_number

p
:
zg
:
DEFINITIONS:
set_number
*EG
*BG
*ZG
before the *PVTG keyword in order to use this option.
User's Guide IMEX
eg
Gas expansion factor (m3/m3 | scf/RB | cm3/cm3) at pressure p.
bg
zg
Gas compressibility factor (dimensionless) at pressure p.
visg
Gas viscosity (mPa-s | cp | cp) at pressure p.
srftn
Water-gas surface tension. It must be monotonically decreasing and greater
DEFAULTS:
Optional keyword. No default values. *PVTG must be present in the data set when the
*GASWATER model is used.
CONDITIONS:

EXPLANATION:
In general, this table must contain at least two entries, and may contain more entries if
desired. Typical gas PVT data curves used by this simulator are given by the Figure 7(c) and
(e) in Appendix D.
relationships. STC means stock tank conditions. (14.7 psia and 60 F | 101.3 kPa and 288.15
K | 1.03 kg/cm2 and 288.15 K)
eg =
(volume of
gas at STC )
bg =

User's Guide IMEX
Example: (SI units)
*PVTG 1
** gas table
** p
eg
100.
1.000
1000.
5.000
1500.
300.000
2000.
600.000
2500.
800.000
3000.
1000.000
3500.
1100.000
4000.
1200.000
4500.
1500.000
visg
0.020
0.010
0.015
0.019
0.021
0.023
0.026
0.028
0.031
min
max
SI
kPa
0.0
1.0E+10
Field
psi
0.0
1.45E+9
Lab
kPa
0.0
1.0E+10
Mod. SI
kg/cm2
0.0
1.01E+8
min
max
SI
m3/m3
1.0E-20
1.0E+3
Field
scf/STB
5.617
5617
Lab
cm3/cm3
1.0E-20
1.0E+3
min
max
SI
mPa*sec
9.5E-5
86,400.
Field
cp
9.5E-5
86,400.
Lab
mPa*sec
9.5E-5
86,400.
User's Guide IMEX
Condensate - Oil and Gas PVT Table (Optional)
*PVTCOND
PURPOSE:
*PVTCOND indicates start of the oil and gas PVTCOND table.

TABLE:
*PVTCOND *EG set_number

p
:
rs
:
rv
:
bo
:
eg
:
viso
:
visg
:
(srftn)
:
bo
:
bg
:
viso
:
visg
:
(srftn)
:
bo
:
zg
:
viso
:
visg
:
(srftn)
:
-or*PVTCOND *BG set_number

p
:
rs
:
rv
:
-or*PVTCOND *ZG set_number

p
:
rs
:
rv
:
NOTE:
We will define the following terms:

1. Black Oil - Liquid at surface, Liquid at reservoir conditions
2. Condensate - Liquid at surface, Gas at reservoir conditions
3. Free Gas - Gas at surface, Gas at reservoir conditions
4. Solution Gas - Gas at surface, Liquid at reservoir conditions
The Gas phase in the reservoir is made up of Condensate and Free Gas. The Oil phase in the
reservoir is made up of Black Oil and Solution Gas. The Gas phase at the surface is made up
of Free Gas and Solution Gas. The Oil phase at the surface is made up of Black Oil and
Condensate. Thus Oil Gas ratio would be (Condensate + Black Oil) / (Free gas + Soln Gas).
By default IMEX will output Oil and Gas based production. Condensate, Black Oil, Soln
Gas and Free Gas production splits may be obtained by using the *PSPLIT keyword in the IO
Control section.
DEFINITIONS:
set_number
*EG
Keyword indicating that the gas expansion factor will be used. If *EG, *BG
and *ZG are not present in the data set then this is the default.
User's Guide IMEX
*BG
*ZG
before the *PVTCOND keyword in order to use this option.
p
rs
pressure p.
rv
Oil Content (m3/m3 | STB/scf | cm3/cm3) for condensate saturated gas at
pressure p.
bo
pressure p.
eg
Gas expansion factor (m3/m3 | scf/RB | cm3/cm3) for condensate saturated
gas at pressure p.
bg
Gas formation volume factor (m3/m3 | RB/scf | cm3/cm3) for condensate
saturated gas at pressure p.
zg
Gas compressibility factor (dimensionless) for condensate saturated gas at
pressure p.
viso
visg
Gas viscosity (mPa-s | cp | cp) for condensate saturated gas at pressure p.
srftn
Gas-liquid interfacial tension. It must be monotonically decreasing and
greater than zero (dyne/cm).
User's Guide IMEX
DEFAULTS:
Optional keyword. No default values. . One of *PVT, *DIFLIB, *PVTAPI, *PVTG,

CONDITIONS:

Interpolation of gas and oil properties in PVTCOND tables occurs between the first table
value and ten times the last table value (extrapolated). During dry gas injection it is
important that the dry gas pressure (where Rv = 0.0) is included in the table. In addition the
largest table entry should be large enough to avoid a large amount of extrapolation (i.e. ten
times the maximum entry).
When a detailed EOS representation is used to generate PVT properties for condensate (such
as the table below) it is possible that many properties such as Rs, Rv, Bo or Gas viscosity
may no longer be monotonic functions of pressure. In most cases this does not present a
problem to IMEX, but under some circumstances the existence of minima or maxima in PVT
can cause convergence difficulties.
EXPLANATION:
The minimum pressure in the PVTCOND table should be below any pressure constraint
specified on a production well in the Well Data section. Normally a value of atmospheric
would be chosen as this would represent a dry gas value. The maximum pressure in the table
should be comfortably above any well injection pressure specified in the Well Data section.
rs =

rv =
(volume of
(volume
bo =

eg =
condensate at STC
of gas at STC )
gas at STC )
User's Guide IMEX
(volume of
bg =

Set_numbers are used to assign different PVTCOND tables to grid blocks in the reservoir. If
If there is more than one PVT region, the first PVTCOND table must have a set_number of
one and set_numbers must increase consecutively for subsequent table sets. Set_numbers are
Example:
*MODEL *GASWATER_WITH_CONDENSATE
*PVTCOND *BG
** pressure solution solution
oil
gas
oil
gas
**
psia
GOR
CGR
FVF
FVF
vis,cp
vis,cp
** -----------------------------------------------------------------------14.70
0.0
0.0
1.10
0.20534
0.24
0.012
515.00
288.7
0.025060e-3 1.2644 5.86124e-3 0.19215 0.01240
1015.00
618.7
0.021148e-3 1.4766 2.72302e-3 0.14588 0.01394
1515.00
980.1
0.026445e-3 1.6987 1.71098e-3 0.11643 0.01652
2015.00
1377.7
0.039146e-3 1.9345 1.23923e-3 0.09735 0.02097
2515.00
1781.6
0.062337e-3 2.1601 1.00106e-3 0.08666 0.02779
3015.00
1996.6
0.100128e-3 2.2398 0.89373e-3 0.08849 0.03686
3265.00
1854.7
0.122098e-3 2.1129 0.87063e-3 0.10269 0.04188
3515.00
1501.4
0.138469e-3 1.8633 0.85536e-3 0.14397 0.04624
3765.00
1155.2
0.146526e-3 1.6384 0.84058e-3 0.22699 0.04948
4006.47
912.6
0.149747e-3 1.4891 0.82634e-3 0.34108 0.05189
min
max
SI
kPa
0.0
1.0E+10
Field
psi
0.0
1.45E+9
Lab
kPa
0.0
1.0E+10
Mod. SI
kg/cm2
0.0
1.01E+8
min
max
SI
m3/m3
0.0
1.0E+3
Field
scf/STB
0.0
5,617
Lab
cm3/cm3
0.0
1.0E+3
User's Guide IMEX
The acceptable range of values for the oil content is:
min
max
SI
m3/m3
0.0
1.0E-1
Field
STB/scf
0.0
0.0178
Lab
cm3/cm3
0.0
1.0E-1
min
max
SI
m3/m3
0.5
5.0
Field
Rb/STB
0.5
5.0
Lab
cm3/cm3
0.5
5.0
min
max
SI
m3/m3
1.0E-20
1.0E+3
Field
scf/STB
5.617E-20
5617
Lab
cm3/cm3
1.0E-20
1.0E+3
min
max
SI
mPa*sec
.0095
86,400.
Field
cp
.0095
86,400.
Lab
mPa*sec
.0095
86,400.
min
max
SI
mPa*sec
9.5E-5
86,400.
Field
cp
9.5E-5
86,400.
Lab
mPa*sec
9.5E-5
86,400.
min
SI
dyne/cm
-1.1E-20
Field
lbf/ft
-6.852E-25
Lab
dyne/cm
-1.1E-20
min
max
SI
1/kPa
0.0
1.0E-3
Field
1/psia
0.0
6.89E-3
Lab
1/Pa
0.0
1.0E-3
Mod. SI
1/(kg/cm2)
0.0
1.0E-2
A positive total hydrocarbon compressibility check is incorporated in IMEX for the

condensate model. Its derivation by K.H. Coats can be found in A Note on IMPES and
some IMPES-based Simulation Models SPEJ Journal 5 (3), Sept 2000.
User's Guide IMEX
We check that:
Rs*Rv < 1
Bg > Rv*Bo
Bo > Rs*Bg
and
(Bg-Rv*Bo)dRs/dP > (1-Rs*Rv)*dBo/dP
or in terms of table entries i and i+1 we check that:
Rs(i)*Rv(i) < 1
Bg(i) > Rv(i)*Bo(i)
Bo(i) > Rs(i)*Bg(i)
(Bg(i)-Rv(i)*Bo(i))*(Rs(i+1)-Rs(i))/(P(i+1)-P(i)) >
(1- Rs(i)*Rv(i))* ))*(Bo(i+1)-Bo(i))/(P(i+1)-P(i))
and
(Bg(i+i)-Rv(i+1)*Bo(i+1))*(Rs(i+1)-Rs(i))/(P(i+1)-P(i)) >
(1- Rs(i+1)*Rv(i+1))* ))*(Bo(i+1)-Bo(i))/(P(i+1)-P(i))
User's Guide IMEX
Volatile - Oil and Gas PVT Table (Optional)
*PVTVO
PURPOSE:
*PVTVO indicates start of the oil and gas PVTVO table.

TABLE:
*PVTVO *EG set_number

p
:
rs
:
rv
:
bo
:
eg
:
viso
:
visg
:
egd
:
visgd
:
(srftn)
:
bg
:
viso
:
visg
:
bgd
:
visgd
:
(srftn)
:
zg
:
viso
:
visg
:
zgd
:
visgd
:
(srftn)
:
-or*PVTVO *BG set_number

p
:
rs
:
rv
:
bo
:
-or*PVTVO *ZG set_number

p
:
rs
:
rv
:
bo
:
NOTE:
We will define the following terms:

1. Black Oil - Liquid at surface, Liquid at reservoir conditions
2. Condensate - Liquid at surface, Gas at reservoir conditions
3. Free Gas - Gas at surface, Gas at reservoir conditions
4. Solution Gas - Gas at surface, Liquid at reservoir conditions
The Gas phase in the reservoir is made up of Condensate and Free Gas. The Oil phase in the
reservoir is made up of Black Oil and Solution Gas. The Gas phase at the surface is made up
of Free Gas and Solution Gas. The Oil phase at the surface is made up of Black Oil and
Condensate. Thus Oil Gas ratio would be (Condensate + Black Oil) / (Free gas + Soln Gas).
By default IMEX will output Oil and Gas based production. Condensate, Black Oil, Soln
Gas and Free Gas production splits may be obtained by using the *PSPLIT keyword in the IO
Control section.
DEFINITIONS:
set_number
*EG
Keyword indicating that the gas expansion factor will be used. If *EG, *BG
and *ZG are not present in the data set then this is the default.
User's Guide IMEX
*BG
*ZG
before the *PVTVO keyword in order to use this option.
p
rs
pressure p.
rv
Oil Content (m3/m3 | STB/scf | cm3/cm3) for condensate saturated gas at
pressure p.
bo
pressure p.
eg
Gas expansion factor (m3/m3 | scf/RB | cm3/cm3) for condensate saturated
gas at pressure p.
bg
Gas formation volume factor (m3/m3 | RB/scf | cm3/cm3) for condensate
saturated gas at pressure p.
zg
Gas compressibility factor (dimensionless) for condensate saturated gas at
pressure p.
viso
visg
Gas viscosity (mPa-s | cp | cp) for condensate saturated gas at pressure p.
egd
Gas expansion factor (m3/m3 | scf/RB | cm3/cm3) for dry gas at pressure p.
User's Guide IMEX
bgd
Gas formation volume factor (m3/m3 | RB/scf | cm3/cm3) for dry gas at
pressure p.
zgd
Gas compressibility factor (dimensionless) for dry gas at pressure p.
visgd
Gas viscosity (mPa-s | cp | cp) for dry gas at pressure p.
srftn
Gas-liquid interfacial tension. It must be monotonically decreasing and
greater than zero (dyne/cm).
DEFAULTS:
Optional keyword. No default values. One of *PVT, *DIFLIB, *PVTAPI, *PVTG,

CONDITIONS:

During dry gas injection it is important that the dry gas pressure (where Rv = 0.0) is included
in the table. When a detailed EOS representation is used to generate PVT properties for
condensate (such as the table below) it is possible that many properties such as Rs, Rv, Bo or
Gas viscosity may no longer be monotonic functions of pressure. In most cases this does not
present a problem to IMEX, but under some circumstances the existence of minima or
maxima in PVT can cause convergence difficulties.
EXPLANATION:
The minimum pressure in the PVTVO table should be below any pressure constraint
specified on a production well in the Well Data section. Normally a value of atmospheric
would be chosen as this would represent a dry gas value. The maximum pressure in the table
should be comfortably above any well injection pressure specified in the Well Data section.
rs =

rv =
(volume of
(volume
User's Guide IMEX
condensate at STC
of gas at STC )
bo =

eg =
(volume of
gas at STC )
bg =

Set_numbers are used to assign different PVTVO tables to grid blocks in the reservoir. If
If there is more than one PVT region, the first PVTVO table must have a set_number of one
and set_numbers must increase consecutively for subsequent table sets. Set_numbers are
The *PVTVO table includes dry gas as well as condensate saturated gas properties. At
pressures other than the dew point pressure, gas properties are obtained by interpolating
between saturated and dry conditions. Thus *EGUST and *VGUST should not be used with
the *PVTVO option as they are not required (*EGUST and *VGUST also define gas
properties at pressures other than the dew point pressure).
Example:
*MODEL *VOLATILE_OIL
*PVTVO *BG
**pressure solution solution
oil
gas
oil
gas
dry gas dry gas
** kpa
GOR
CGR
FVF
FVF
vis,cp
vis,cp
FVF
vis,cp
** --------------------------------------------------------------------------------101.32
0.0
0.00000
1.0
1.35118
0.180
0.01300 1.35118 0.01300
4926.30 21.8867
0.000027
1.1436 0.02779
0.17945
0.01470 0.02777 0.01473
8372.75 41.5325
0.000033
1.2175 0.01597
0.16769
0.01553 0.01598 0.01553
14576.30 83.2489
0.000059
1.3447 0.00899
0.14546
0.01773 0.00902 0.01756
21469.18 143.3570 0.000119
1.5225 0.00617
0.12204
0.02159 0.00620 0.02068
29740.60 254.5648 0.000264
1.8504 0.00472
0.09578
0.02856 0.00467 0.02507
38012.04 530.0487 0.000656
2.7030 0.00420
0.06721
0.04125 0.00387 0.02973
39816.28 801.7297 0.001026
3.6174 0.004144 0.06200
0.05026 0.00374 0.03097
min
max
SI
kPa
0.0
1.0E+10
Field
psi
0.0
1.45E+9
Lab
kPa
0.0
1.0E+10
Mod. SI
kg/cm2
0.0
1.01E+8
User's Guide IMEX
min
max
SI
m3/m3
0.0
1.0E+3
Field
scf/STB
0.0
5,617
Lab
cm3/cm3
0.0
1.0E+3
The acceptable range of values for the oil content is:
min
max
SI
m3/m3
0.0
1.0E-1
Field
STB/scf
0.0
0.0178
Lab
cm3/cm3
0.0
1.0E-1
min
max
SI
m3/m3
0.5
5.0
Field
Rb/STB
0.5
5.0
Lab
cm3/cm3
0.5
5.0
min
max
SI
m3/m3
1.0E-20
1.0E+3
Field
scf/STB
5.617E-20
5617
Lab
cm3/cm3
1.0E-20
1.0E+3
min
max
SI
mPa*sec
.0095
86,400.
Field
cp
.0095
86,400.
Lab
mPa*sec
.0095
86,400.
min
max
SI
mPa*sec
9.5E-5
86,400.
Field
cp
9.5E-5
86,400.
Lab
mPa*sec
9.5E-5
86,400.
min
SI
dyne/cm
-1.1E-20
Field
lbf/ft
-6.852E-25
Lab
dyne/cm
-1.1E-20
min
max
User's Guide IMEX
SI
1/kPa
0.0
1.0E-3
Field
1/psia
0.0
6.89E-3
Lab
1/Pa
0.0
1.0E-3
Mod. SI
1/(kg/cm2)
0.0
1.0E-2
A positive total hydrocarbon compressibility check is incorporated in IMEX for the

condensate model. Its derivation by K.H. Coats can be found in A Note on IMPES and
some IMPES-based Simulation Models SPEJ Journal 5 (3), Sept 2000.
We check that:
Rs*Rv < 1
Bg > Rv*Bo
Bo > Rs*Bg
and
(Bg-Rv*Bo)dRs/dP > (1-Rs*Rv)*dBo/dP
or in terms of table entries i and i+1 we check that:
Rs(i)*Rv(i) < 1
Bg(i) > Rv(i)*Bo(i)
Bo(i) > Rs(i)*Bg(i)
(Bg(i)-Rv(i)*Bo(i))*(Rs(i+1)-Rs(i))/(P(i+1)-P(i)) >
(1- Rs(i)*Rv(i))* ))*(Bo(i+1)-Bo(i))/(P(i+1)-P(i))
and
(Bg(i+i)-Rv(i+1)*Bo(i+1))*(Rs(i+1)-Rs(i))/(P(i+1)-P(i)) >
(1- Rs(i+1)*Rv(i+1))* ))*(Bo(i+1)-Bo(i))/(P(i+1)-P(i))
User's Guide IMEX
Oil Compressibility as a Function of Pressure (Optional)

*COT
PURPOSE:
*COT indicates the input of an oil compressibility table that is a function of both pressure, P,
and bubble point pressure, Pb.
FORMAT:
*COT set_number
p
:
co
:
DEFINITION:
set_number
PVT region that corresponds to the co table.
p
Pressure above the corresponding bubble point pressure in the PVT table. (kPa
| psi | kPa | kg/cm2). The first entry of this table is the oil compressibility value
at the bubble point pressure. All subsequent pressure entries must be greater
than this bubble point pressure and must increase in a monotonic fashion.
co
Oil Compressibility entry at the corresponding pressure. (1/kPa | 1/psi | 1/kPa
| 1/(kg/cm2)).
DEFAULT:

CONDITIONS:
This keyword must be in the Component Properties section. *COT may be used when the oil
compressibility is a function of both pressure and bubble point pressure. This corresponds to
an oil formation volume factor (Bo) which has a nonlinear functional dependence on
pressures above the bubble point pressure. The *COT keyword cannot be used with the *CO
keyword, the *BOT keyword, or when co is entered in the PVT table. The *COT keyword
cannot be used with the *GASWATER option (The oil phase is not modelled). The *COT
keyword cannot be used with the *GASWATER_WITH_CONDENSATE option (The oil
phase is always saturated with gas).
EXPLANATION:
The oil compressibility co, gives the slope of the bo curve divided by bo(Pb) when the oil is
above the bubble point pressure. For a typical black-oil system, co assumes a constant value for
P > Pb. A weak compositional dependence can be exhibited by allowing co to vary with Pb. co
is further allowed to vary in a nonlinear manner with P (P > Pb). If co is taken as constant, then
bo(P) = bo(Pb) * [1 - co * (P-Pb)]
User's Guide IMEX
If co is a function of bubble point pressure only, then a different value of co must be used for
each bubble point pressure. If co is also a function of pressure, then co is integrated over the
range of pressures between Pb and P and used in the equation above. This integration is
performed numerically within IMEX.
Because of the integration step, it is not trivial to calculate co's obtained from measured
values of Bo above the bubble point into IMEX by hand. However, ModelBuilder can easily
perform the calculations required to convert measured Bo above the bubble point into co
above the bubble point for *COT table input.
Many *COT keywords, with each *COT keyword followed two columns of numbers (p vs.
co entries), may appear in the Component Properties Data section. If more than one PVT
region is specified, then each *COT keyword must be followed by an integer specifying the
PVT region that the p vs. co table represents. If the number of PVT regions is only one, then
it is not necessary to specify the set_number.
The minimum number of entries specified for each table is one (co at the bubble point). If the
calculated block pressure is outside the range of data specified, then the last table entry value
of co is used for the calculated block pressure.
The minimum number of *COT keywords required per PVT region is one. If more than one
*COT keyword is defined per PVT region, then the bubble point pressure (the first pressure
entry in the table) must be greater than the previous tables value and less than the next tables
value. If more than one table is specified, then a linear interpolation is performed to
determine the variation of co with p for the calculated block bubble point pressure. If the
calculated block bubble point pressure falls outside the tables range, then the table which is
closest to the calculated value is used. In other words, values are not extrapolated beyond the
range of data specified.
Example:
*PVT 1
** field units
** p
rs
bo
eg
14.7
1.0
1.062
6.0
264.7
90.5
1.15
82.692
514.7
180.0
1.207
159.388
1014.7
371.0
1.295
312.793
2014.7
636.0
1.435
619.579
2514.7
775.0
1.5
772.798
3014.7
930.0
1.565
925.926
4014.7 1270.0
1.695 1233.046
5014.7 1618.0
1.827 1540.832
9014.7 2984.0
2.36
2590.674
*COT 1
** p
co
14.7
2.00e-06
1000.0
1.90e-06
2000.0
1.75e-06
5000.0
1.70e-06
9000.0
1.68e-06
viso
1.04
0.975
0.91
0.83
0.69
0.64
0.594
0.51
0.449
0.203
visg
0.0080
0.0096
0.0112
0.0140
0.0189
0.0208
0.0228
0.0268
0.0309
0.0470
User's Guide IMEX
*COT 1
** p
3014.7
4000.0
5000.0
9000.0
*COT 1
** p
7000.0
8000.0
9000.0
co
1.80e-06
1.75e-06
1.73e-06
1.70e-06
co
1.10e-06
1.09e-06
1.09e-06
The acceptable range of values for oil compressibility is:
min
max
User's Guide IMEX
SI
1/kPa
0.0
1.0E-3
Field
1/psia
0.0
6.89E-3
Lab
1/Pa
0.0
1.0E-3
Mod. SI
1/(kg/cm2)
0.0
1.0E-2
Bo Above the Bubble Point as a Function of Pressure

(Optional)
*BOT
PURPOSE:
*BOT indicates the input of an Bo table that is a function of both pressure, P, and bubble
point pressure, Pb.
FORMAT:
*BOT set_number
p
:
Bo
:
DEFINITION:
set_number
PVT region that corresponds to the bo table.
p
Pressure above the corresponding bubble point pressure in the PVT table. (kPa |
psi | kPa | kg/cm2). The first entry of this table is the oil formation volume factor
value at the bubble point pressure. All subsequent pressure entries must be
greater than this bubble point pressure and must increase in a monotonic fashion.
Bo
Oil formation volume factor entry at the corresponding pressure. (1/kPa |
1/psi | 1/kPa | 1/(kg/cm2)).
DEFAULT:

CONDITIONS:
This keyword must be in the Component Properties section. *BOT may be entered when the
oil formation volume factor is a function of both pressure and bubble point pressure. This
corresponds to an oil formation volume factor (Bo) which has a nonlinear functional
dependence on pressures above the bubble point pressure. The *BOT keyword cannot be used
with the *COT keyword, *CO keyword or when co is entered in the PVT table. The *BOT
keyword cannot be used with the *GASWATER option (The oil phase is not modelled). The
*BOT keyword cannot be used with the *GASWATER_WITH_CONDENSATE option (The
oil phase is always saturated with gas).
EXPLANATION:
The oil formation volume factor can exhibit a weak compositional dependence when the oil is
above the bubble point. The *BOT table allows us to model this dependence varying with both
bubble point pressure and block pressure.
User's Guide IMEX
Many *BOT keywords, with each *BOT keyword followed two columns of numbers (p vs.
bo entries), may appear in the Component Properties Data section. If more than one PVT
region is specified, then each *BOT keyword must be followed by an integer specifying the
PVT region that the p vs. Bo table represents. If the number of PVT regions is only one, then
The minimum number of entries specified for each table is two (Bo at the bubble point and Bo
at a pressure above the bubble point). If the calculated block pressure is outside the range of
data specified, then the last table entry value of Bo is used for the calculated block pressure.
The minimum number of *BOT keywords per PVT region required is one. If more than one
*BOT keyword is defined per PVT region, then the bubble point pressure (the first pressure entry
in the table) must be greater than the previous tables value and less than the next tables value. If
more than one table is specified, then a linear interpolation is performed to determine the variation
of Bo with p for the calculated block bubble point pressure. If the calculated block bubble point
pressure falls outside the tables range, then the table which is closest to the calculated value is
used. In other words, values are not extrapolated beyond the range of data specified.
Example:
*PVT 1
** p
14.7
264.7
514.7
1014.7
2014.7
2514.7
3014.7
4014.7
5014.7
9014.7
** field units
rs
bo
eg
1.0
1.062
6.0
90.5
1.15
82.692
180.0
1.207
159.388
371.0
1.295
312.793
636.0
1.435
619.579
775.0
1.5
772.798
930.0
1.565
925.926
1270.0
1.695 1233.046
1618.0
1.827 1540.832
2984.0
2.36
2590.674
*BOT 1
** p
14.7
1000.0
2000.0
5000.0
9000.0
bo
1.062
1.051
1.041
1.009
0.966
*BOT 1
** p
3014.7
4000.0
5000.0
9000.0
bo
1.565
1.549
1.534
1.471
User's Guide IMEX
viso
1.04
0.975
0.91
0.83
0.69
0.64
0.594
0.51
0.449
0.203
visg
0.0080
0.0096
0.0112
0.0140
0.0189
0.0208
0.0228
0.0268
0.0309
0.0470
*BOT 1
** p
7000.0
8000.0
9000.0
bo
2.000
1.980
1.960
User's Guide IMEX
Oil Viscosity Above the Bubble Point as a Function of

Pressure (Optional)
*VOT
PURPOSE:
*VOT indicates the input of an oil viscosity (Vo) table that is a function of both pressure, P,
and bubble point pressure, Pb.
FORMAT:
*VOT set_number
p
:
Vo
:
DEFINITION:
set_number
PVT region that corresponds to the bo table.
p
Pressure above the corresponding bubble point pressure in the PVT table. (kPa
| psi | kPa | kg/cm2). The first entry of this table is the oil viscosity value at the
bubble point pressure. All subsequent pressure entries must be greater than
this bubble point pressure and must increase in a monotonic fashion.
Vo
Oil viscosity entry at the corresponding pressure. (mPa-s | cp | cp | mPa-s).
DEFAULT:

CONDITIONS:
This keyword must be in the Component Properties section. *VOT may be entered when the
oil viscosity is a function of both pressure and bubble point pressure. This corresponds to an
oil viscosity (Vo) which has a nonlinear functional dependence on pressures above the bubble
point pressure. The *VOT keyword cannot be used with the *CVO keyword. The *VOT
keyword cannot be used with the *GASWATER option (The oil phase is not modelled). The
*VOT keyword cannot be used with the *GASWATER_WITH_CONDENSATE option (The
oil phase is always saturated with gas).
EXPLANATION:
The oil viscosity can exhibit a weak compositional dependence when the oil is above the bubble
point. The *VOT table allows us to model this dependence varying with both bubble point
pressure and block pressure.
User's Guide IMEX
Many *VOT keywords, with each *VOT keyword followed two columns of numbers (p vs.
Vo entries), may appear in the Component Properties Data section. If more than one PVT
region is specified, then each *VOT keyword must be followed by an integer specifying the
PVT region that the p vs. Vo table represents. If the number of PVT regions is only one, then
The minimum number of entries specified for each table is two (oil viscosity at the bubble
point and oil viscosity at a pressure above the bubble point). If the calculated block pressure
is outside the range of data specified, then the last table entry value of Vo is used for the
calculated block pressure.
The minimum number of *VOT keywords required per PVT region is one. If more than one
*VOT keyword is defined per PVT region, then the bubble point pressure (the first pressure
entry in the table) must be greater than the previous tables value and less than the next tables
value. If more than one table is specified, then a linear interpolation is performed to
determine the variation of Vo with p for the calculated block bubble point pressure. If the
calculated block bubble point pressure falls outside the tables range, then the table which is
closest to the calculated value is used. In other words, values are not extrapolated beyond the
range of data specified.
Example:
*PVT 1
** p
14.7
264.7
514.7
1014.7
2014.7
2514.7
3014.7
4014.7
5014.7
9014.7
** field units
rs
bo
eg
1.0
1.062
6.0
90.5
1.15
82.692
180.0
1.207
159.388
371.0
1.295
312.793
636.0
1.435
619.579
775.0
1.5
772.798
930.0
1.565
925.926
1270.0
1.695 1233.046
1618.0
1.827 1540.832
2984.0
2.36
2590.674
*VOT 1
** p
14.7
1000.0
2000.0
5000.0
9000.0
vo
1.040
1.064
1.088
1.161
1.259
*VOT 1
** p
3014.7
4000.0
5000.0
9000.0
vo
0.594
0.618
0.642
0.764
viso
1.04
0.975
0.91
0.83
0.69
0.64
0.594
0.51
0.449
0.203
visg
0.0080
0.0096
0.0112
0.0140
0.0189
0.0208
0.0228
0.0268
0.0309
0.0470
User's Guide IMEX
*VOT 1
** p
7000.0
8000.0
9000.0
co
0.326
0.350
0.375
The acceptable range of values for oil viscosity is:
min
max
User's Guide IMEX
SI
mPa-s
0.095
8.64E+4
Field
cp
0.095
8.64E+4
Lab
mPa-s
0.095
8.64E+4
Mod. SI
mPa-s
0.095
8.64E+4
Condensate Model - Undersaturated Eg, Bg, Zg Table

*EGUST, *BGUST, *ZGUST
PURPOSE:
*EGUST, *BGUST or *ZGUST indicates the input of a gas expansion factor, gas formation
volume factor or gas super compressibility table for condensate undersaturated gas.
FORMAT:
*EGUST set_number
p
:
:
Eg/Bg/Zg
:
:
DEFINITION:
set_number
PVT region that corresponds to the table.
p
Dew point pressure of a new saturated gas mixture formed by reducing the
oil content of the original saturated gas. The first entry is the driest gas dew
point pressure, the last entry is the original (unreduced) saturated gas dew
point pressure (kPa | psi | kPa | kg/cm2).
Eg/Bg/Zg
Gas expansion factor (formation volume factor or super compressibility) of a
new saturated gas mixture formed by reducing the oil content of the original
saturated gas. The first entry is the Eg/Bg/Zg at the driest gas dew point
pressure, the last entry is the Eg/Bg/Zg at the original (unreduced) saturated
gas dew point pressure.
DEFAULT:
Optional keyword. No default. No default. May be used with the

*GASWATER_WITH_CONDENSATE or *VOLATILE_OIL option. If not present, values
of Eg/Bg/Zg from the *PVTCOND table are used.
CONDITIONS:
This keyword must be in the Component Properties section. *EGUST/*BGUST/*ZGUST

may be entered when the gas density is a function of both pressure and dew point pressure.
This corresponds to an undersaturated gas density that has a nonlinear functional dependence
on amount of condensate within the gas phase. This keyword can only be used with the
*GASWATER_WITH_CONDENSATE or *VOLATILE_OIL option.
User's Guide IMEX
Many *EGUST (refers to BGUST and ZGUST as well) keywords, with each *EGUST
keyword followed two columns of numbers (pdew vs. Eg entries), may appear in the
Component Properties Data section for each PVT region. If more than one *EGUST keyword
is defined per PVT region, then the saturated gas pressure (the last pressure entry in the table)
must be greater than the previous tables value and less than the next tables value. It would be
normal for the first entry in each table to be atmospheric pressure. This would allow the user to
input a dry gas value.
EXPLANATION:
The gas expansion factor Eg, is ideally a near linear function of pressure describing how gas
density varies with gas pressure. The use of the condensate model may require the modelling
of the condensate compositional dependence when condensate in gas is modelled at pressures
above the dew point pressure (undersaturated conditions). The *EGUST tables allows the user
to input Eg values at dew point pressures below the saturated gas pressure
If more than one PVT region is specified, then each *EGUST keyword must be followed by
an integer specifying the PVT region that the dew pressure vs. Eg table represents. If the
number of PVT regions is only one, then it is not necessary to specify the set_number.
The minimum number of entries specified for each table is two (Eg at a dew pressure below
the saturated condition and Eg at the dew pressure of the saturated condition). If the
calculated dew point pressure is outside the range of data specified, then the last table entry
value of Eg is used for the calculated dew pressure.
The minimum number of *EGUST keywords required per PVT region is one. If more than
one *EGUST keyword is defined per PVT region, then the saturated dew point pressure (the
last pressure entry in the table) must be greater than the previous tables last value and less than
the next tables last value. If more than one table is specified, then a linear interpolation is
performed to determine the variation of Eg with dew point pressure for the calculated block
saturated gas pressure. If the calculated block saturated gas pressure falls outside the tables
range, then the table which is closest to the calculated value is used. In other words, values
are not extrapolated beyond the range of data specified.
Example 1: *BGUST (using all saturation pressures available from example in *PVTCOND
keyword)
*BGUST 1
**press
gas
**psia
FVF
** -----------------------14.70
5.83237e-3
** Dry Gas
515.00
5.86124e-3
** Saturated
*BGUST 1
14.70
515.00
1015.00
User's Guide IMEX
2.74702e-3
2.71717e-3
2.72302e-3
** Dry Gas
** Undersaturated
** Saturated
*BGUST 1
14.70
515.00
1015.00
1515.00
1.74664e-3
1.71313e-3
1.71908e-3
1.71098e-3
**
**
**
**
Dry Gas
Undersaturated
Undersaturated
Saturated
*BGUST 1
14.70
515.00
1015.00
1515.00
2015.00
1.27707e-3
1.25368e-3
1.25760e-3
1.25228e-3
1.23923e-3
**
**
**
**
**
Dry Gas
Undersaturated
Undersaturated
Undersaturated
Saturated
*BGUST 1
14.70
515.00
1015.00
1515.00
2015.00
2515.00
1.02302e-3
1.01330e-3
1.01483e-3
1.01276e-3
1.00808e-3
1.00106e-3
**
**
**
**
**
**
Dry Gas
Undersaturated
Undersaturated
Undersaturated
Undersaturated
Saturated
*BGUST 1
14.70
515.00
1015.00
1515.00
2015.00
2515.00
3015.00
0.87555e-3
0.87713e-3
0.87682e-3
0.87724e-3
0.87860e-3
0.88262e-3
0.89373e-3
**
**
**
**
**
**
**
Dry Gas
Undersaturated
Undersaturated
Undersaturated
Undersaturated
Undersaturated
Saturated
*BGUST 1
14.70
515.00
1015.00
1515.00
2015.00
2515.00
3015.00
3265.00
0.82551e-3
0.83126e-3
0.83031e-3
0.83161e-3
0.83505e-3
0.84268e-3
0.85903e-3
0.87063e-3
**
**
**
**
**
**
**
**
Dry Gas
Undersaturated
Undersaturated
Undersaturated
Undersaturated
Undersaturated
Undersaturated
Saturated
User's Guide IMEX
*BGUST 1
14.70
515.00
1015.00
1515.00
2015.00
2515.00
3015.00
3265.00
3515.00
0.78619e-3
0.79518e-3
0.79373e-3
0.79570e-3
0.80070e-3
0.81100e-3
0.83113e-3
0.84458e-3
0.85536e-3
**
**
**
**
**
**
**
**
**
Dry Gas
Undersaturated
Undersaturated
Undersaturated
Undersaturated
Undersaturated
Undersaturated
Undersaturated
Saturated
*BGUST 1
14.70
515.00
1015.00
1515.00
2015.00
2515.00
3015.00
3265.00
3515.00
3765.00
0.75438e-3
0.76592e-3
0.76408e-3
0.76657e-3
0.77278e-3
0.78513e-3
0.80809e-3
0.82294e-3
0.83464e-3
0.84058e-3
**
**
**
**
**
**
**
**
**
**
Dry Gas
Undersaturated
Undersaturated
Undersaturated
Undersaturated
Undersaturated
Undersaturated
Undersaturated
Undersaturated
Saturated
*BGUST 1
14.70
515.00
1015.00
1515.00
2015.00
2515.00
3015.00
3265.00
3515.00
3765.00
4006.47
0.72856e-3
0.74214e-3
0.73999e-3
0.74290e-3
0.75007e-3
0.76402e-3
0.78920e-3
0.80512e-3
0.81754e-3
0.82380e-3
0.82634e-3
**
**
**
**
**
**
**
**
**
**
**
Dry Gas
Undersaturated
Undersaturated
Undersaturated
Undersaturated
Undersaturated
Undersaturated
Undersaturated
Undersaturated
Undersaturated
Saturated
The acceptable range of values for Gas Expansion Factor is:
min
max
User's Guide IMEX
SI
m3/m3
0.0
1.0E+3
Field
scf/STB
0.0
5,615.0
Lab
cm/3cm3
0.0
1.0E+3
Mod. SI
m3/m3
0.0
1.0E+3
Condensate Model - Undersaturated Gas Viscosity Table

*VGUST
PURPOSE:
*VGUST indicates the input of a gas viscosity table for condensate undersaturated gas
FORMAT:
*VGUST set_number
p
:
:
Vg
:
:
DEFINITION:
set_number
PVT region that corresponds to the co table.
p
Dew point pressure of a new saturated gas mixture formed by reducing the
oil content of the original saturated gas. The first entry is the driest gas dew
point pressure, the last entry is the original (unreduced) saturated gas dew
point pressure (kPa | psi | kPa | kg/cm2).
Vg
Gas viscosity of a new saturated gas mixture formed by reducing the oil
content of the original saturated gas. The first entry is gas viscosity at the
driest gas dew point pressure, the last entry is the gas viscosity at the original
(unreduced) saturated gas dew point pressure.
DEFAULT:
Optional keyword. No default. May be used with the *GASWATER_WITH_CONDENSATE or

*VOLATILE_OIL option. If not present, values of Vg from the *PVTCOND table are used.
CONDITIONS:
This keyword must be in the Component Properties section. *VGUST may be entered when
the gas viscosity is a function of both pressure and dew point pressure. This corresponds to
an undersaturated gas viscosity that has a nonlinear functional dependence on amount of
condensate within the gas phase. This keyword can only be used with the
*GASWATER_WITH_CONDENSATE or *VOLATILE_OIL option.
Many *VGUST keywords, with each *VGUST keyword followed two columns of numbers
(pdew vs. Vg entries), may appear in the Component Properties Data section for each PVT
region. If more than one *VGUST keyword is defined per PVT region, then the saturated gas
pressure (the last pressure entry in the table) must be greater than the previous tables value and
less than the next tables value. It would be normal for the first entry in each table to be
atmospheric pressure. This would allow the user to input a dry gas value.
User's Guide IMEX
EXPLANATION:
The gas Viscosity, Vg, is ideally a function of pressure describing how gas viscosity varies with
gas pressure. The use of the condensate model may require the modelling of the condensate
compositional dependence when condensate in gas is modelled at pressures above the dew
point pressure (undersaturated conditions). The *VGUST tables allows the user to input Vg
values at dew pressures below the saturated gas pressure
If more than one PVT region is specified, then each *VGUST keyword must be followed by
an integer specifying the PVT region that the dew pressure vs. Vg table represents. If the
number of PVT regions is only one, then it is not necessary to specify the set_number.
The minimum number of entries specified for each table is two (Vg at a dew pressure below
the saturated condition and Vg at the dew pressure of the saturated condition). If the
calculated dew pressure is outside the range of data specified, then the last table entry value
of Vg is used for the calculated dew pressure.
The minimum number of *VGUST keywords required per PVT region is one. If more than
one *VGUST keyword is defined per PVT region, then the saturated dew point pressure (the
last pressure entry in the table) must be greater than the previous tables last value and less than
the next tables last value. If more than one table is specified, then a linear interpolation is
performed to determine the variation of Vg with dew point pressure for the calculated block
saturated gas pressure. If the calculated block saturated gas pressure falls outside the tables
range, then the table which is closest to the calculated value is used. In other words, values
are not extrapolated beyond the range of data specified.
Example: VGUST (using all saturation pressures available from example in *PVTCOND
keyword)
*VGUST 1
**press gas
**psia
vis,cp
** -----------------------14.70
0.01251
** Dry Gas
515.00
0.012403
** Saturated
*VGUST 1
14.70
0.01384
515.00
0.01396
1015.00 0.013943
** Dry Gas
** Undersaturated
** Saturated
*VGUST 1
14.70
515.00
1015.00
1515.00
**
**
**
**
User's Guide IMEX
0.01590
0.01648
0.01638
0.016523
Dry Gas
Undersaturated
Undersaturated
Saturated
*VGUST 1
14.70
515.00
1015.00
1515.00
2015.00
0.01894
0.02018
0.01997
0.02025
0.020973
**
**
**
**
**
Dry Gas
Undersaturated
Undersaturated
Undersaturated
Saturated
*VGUST 1
14.70
515.00
1015.00
1515.00
2015.00
2515.00
0.02287
0.02473
0.02442
0.02483
0.02585
0.027793
**
**
**
**
**
**
Dry Gas
Undersaturated
Undersaturated
Undersaturated
Undersaturated
Saturated
*VGUST 1
14.70
515.00
1015.00
1515.00
2015.00
2515.00
3015.00
3265.00
0.02918
0.03169
0.03129
0.03183
0.03314
0.03555
0.03954
0.041883
**
**
**
**
**
**
**
**
Dry Gas
Undersaturated
Undersaturated
Undersaturated
Undersaturated
Undersaturated
Undersaturated
Saturated
*VGUST 1
14.70
515.00
1015.00
1515.00
2015.00
2515.00
3015.00
3265.00
3515.00
0.03094
0.03361
0.03319
0.03376
0.03514
0.03769
0.04191
0.04438
0.046243
**
**
**
**
**
**
**
**
**
Dry Gas
Undersaturated
Undersaturated
Undersaturated
Undersaturated
Undersaturated
Undersaturated
Undersaturated
Saturated
*VGUST 1
14.70
515.00
1015.00
1515.00
2015.00
2515.00
3015.00
3265.00
3515.00
3765.00
0.03248
0.03527
0.03483
0.03543
0.03687
0.03954
0.04395
0.04655
0.04851
0.049483
**
**
**
**
**
**
**
**
**
**
Dry Gas
Undersaturated
Undersaturated
Undersaturated
Undersaturated
Undersaturated
Undersaturated
Undersaturated
Undersaturated
Saturated
User's Guide IMEX
*VGUST 1
14.70
515.00
1015.00
1515.00
2015.00
2515.00
3015.00
3265.00
3515.00
3765.00
4006.47
0.03386
0.03675
0.03629
0.03691
0.03840
0.04116
0.04573
0.04844
0.05048
0.05148
0.051893
**
**
**
**
**
**
**
**
**
**
**
Dry Gas
Undersaturated
Undersaturated
Undersaturated
Undersaturated
Undersaturated
Undersaturated
Undersaturated
Undersaturated
Undersaturated
Saturated
The acceptable range of values for Gas viscosity is:
min
max
User's Guide IMEX
SI
MPa-s
0.095
8.64E+4
Field
cp
0.095
8.64E+4
Lab
MPa-s
0.095
8.64E+4
Mod. SI
MPa-s
0.095
8.64E+4
Reference Pressure for Gas-Oil Capillary Pressure (Optional)

*REFPST
PURPOSE:
*REFPST indicates the input of a reference pressure for the gas-oil (liquid) capillary pressure
input.
FORMAT:
*REFPST pst
DEFINITION:
pst
Value for reference pressure for gas-oil (liquid) capillary pressure (kPa | psi |
kPa | kg/cm2).
DEFAULT:

CONDITIONS:
*REFPST needs to be entered when the surface tension is entered with the gas-oil PVT tables.
With the surface tension, this reference pressure determines the gas-oil (liquid) capillary
pressure.
EXPLANATION:
The calculation for interfacial effects on gas-oil capillary pressure is:

Pcog =
int t
* PcogI
int tI
where:
intt
inttI
PcogI
interfacial tension at current block pressure

interfacial tension at reference pressure, pst
gas-oil capillary pressure at reference pressure, pst
For the *GASWATER model, the capillary pressure modified is Pcwg entered in the *SWT
table, for the *GASWATER_WITH_CONDENSATE model, the capillary pressure modified
is Pclg entered in the *SWT table.
The acceptable range of values for reference pressure is:
min
max
SI
kPa
0.0
1.0E+6
Field
psi
0.0
145,039
Lab
kPa
0.0
1.0E+6
Mod. SI
kg/cm2
0.0
1.01E+4
User's Guide IMEX
Solvent PVT Table (Conditional)
*PVTS
PURPOSE:
*PVTS indicates the start of the solvent PVT table.

FORMAT:
*PVTS *ES
p
:
rss
:
es
:
viss
:
omega_os
:
bs
:
viss
:
omega_os
:
zs
:
viss
:
omega_os
:
-or*PVTS *BS
p
:
rss
:
-or*PVTS *ZS
p
:
rss
:
DEFINITIONS:
*ES
Keyword indicating that the solvent expansion factor will be used. If all
*ES, *BS and *ZS are not present in the data set then this is the default.
*BS
Keyword indicating the solvent formation volume factor will be used instead
of the solvent expansion factor.
*ZS
Keyword indicating that the solvent compressibility factor will be used
instead of the solvent expansion factor. The reservoir temperature TRES
must be specified before the *PVTS keyword in order to use this option.
p
Pressure (kPa | psi | kPa | kg/cm2) in oil phase.
rss
Solvent-water ratio (m3/m3 | scf/STB | cm3/cm3) at the given pressure p.
es
Solvent expansion factor (m3/m3 | scf/RB | cm3/cm3) at pressure p.
User's Guide IMEX
bs
Solvent formation volume factor (m3/m3 | RB/scf | cm3/cm3).
zs
Solvent compressibility factor (dimensionless) at pressure p.
viss
Solvent phase viscosity (mPa-s | cp | cp) at pressure p.
omega_os
Mixing parameter between oil and solvent. Its value is bounded between 0 and
1 representing the immiscible case and the completely miscible case. It is a
dimensionless fraction. The last entry for omega_os must be greater than zero.
DEFAULTS:
Conditional keyword. No default values.

CONDITIONS:
This keyword must be in the Component Property keyword group. Required with *MISCG
or *MISNCG.
EXPLANATION:
The gradient of the Rss curve above the bubble point is equal to zero. IMEX automatically uses
Appendix A).
specified in the Well Data section. The formation volume factor and viscosity of water are
calculated from functional relationships.
When the pseudo-miscible option is enabled a fourth mass conservation equation is used to
simulate the solvent component. Solvent can dissolve in water but not in oil.
The gas mass conservation equation can be used to simulate chase gas injection. Obviously,
the reservoir must be undersaturated and the solution gas should not vaporize during the
simulation. If chase gas is not injected, the gas properties entered are for the gas-in-place.
For more information about the pseudo-miscible formulation, consult Todd and Longstaff,
1972, "The Development, Testing, and Application of a Numerical Simulator for Predicting
Miscible Fluid Performance", Trans. AIME 253, p874. For complete details of the actual
formulation, consult the IMEX Technical Manual.
User's Guide IMEX
Entries must be in order of increasing pressure.

rss =
es =
bs =
(volume of solvent at STC )

(volume of water at STC )
(volume of solvent at reservoir conditions )

(volume of solvent at reservoir conditions )
The default is *PVTS *EG. Use *BS to enter values for bs instead of es and *ZS to enter
values of zs instead of es.
*PVTS
** p
14.7
500.0
1000.0
1200.0
1500.0
1800.0
2000.0
2302.3
2500.0
3000.0
3500.0
4000.0
4500.0
4800.0
rss
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
es
4.480
178.501
395.100
491.304
641.313
790.077
885.269
1020.096
1100.715
1280.902
1429.797
1555.210
1661.958
1719.100
viss
0.011
0.012
0.013
0.014
0.016
0.018
0.019
0.022
0.023
0.027
0.031
0.034
0.037
0.038
omega_os
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.283
1.0
1.0
1.0
1.0
1.0
min
max
SI
kPa
0.0
1.0E+10
Field
psi
0.0
1.45E+9
Lab
kPa
0.0
1.0E+10
Mod. SI
kg/cm2
0.0
1.01E+8
The acceptable range of values for the solvent- water ratio is:
min
max
SI
m3/m3
0.0
1.0E+3
Field
scf/STB
0.0
5,617
Lab
cm3/cm3
0.0
1.0E+3
The acceptable range of values for the solvent expansion factors is:
min
max
User's Guide IMEX
SI
m3/m3
1.0E-20
1.0E+3
Field
Rb/STB
5.671E-20
5617
Lab
cm3/cm3
1.0E-20
1.0E+3
The acceptable range of values for the formation volume factor is:
min
max
SI
m3/m3
0.5
5.0
Field
Rb/STB
0.5
5.0
Lab
cm3/cm3
0.5
5.0
The acceptable range of values for the solvent viscosity is:
min
max
SI
mPa*sec
.0095
86,400.
Field
cp
.0095
86,400.
Lab
mPa*sec
.0095
86,400.
The acceptable range of values for the omega factor (a fraction) is:
min
max
SI
Field
Lab
0.0
1.0
0.0
1.0
0.0
1.0
User's Guide IMEX
Densities (Required)
*DENSITY, *GRAVITY
PURPOSE:
*DENSITY indicates the input of a density (for a PVT region).

*GRAVITY indicates the input of a gravity (for a PVT region).
FORMAT:
*DENSITY
*DENSITY
*DENSITY
*OIL
*GAS
*WATER
value
value
value
(*DENSITY *SOLVENT value)

*GRAVITY *GAS gas_gravity
DEFINITIONS:
*OIL
Keyword signaling that this is the oil phase density at STC.
*GAS
Keyword signaling that this is the gas phase density or gas phase gravity at
STC.
*WATER
Keyword signaling that this is the water phase density at STC.
*SOLVENT
Keyword signaling that this is the solvent phase density at STC. (Required
for *MISCG and *MISNCG). A single solvent density is defined for the
entire reservoir. It is not entered on a PVT region basis.
value
Value of density (kg/m3 | lbm/ft3 | g/cm3).
gas_gravity
Value of gas specific gravity (Air = 1).
DEFAULTS:
Required keyword. No default values. If the API or GASWATER option is used,

*DENSITY *OIL is not required.
CONDITIONS:
These keywords must be in the Component Property keyword group. *SOLVENT is

required for *MISCG and *MISNCG. One of *DENSITY *GAS or *GRAVITY *GAS must
be present in the data set. The *DENSITY *OIL keyword cannot be used with the
*GASWATER option (The oil phase is not modelled).
User's Guide IMEX
EXPLANATION:
Note that if multiple PVT regions are used a *DENSITY keyword for oil and gas must be
specified corresponding to each PVT region.
If a single *DENSITY keyword for water is encountered it will be applied to all PVT regions.
If multiple *DENSITY keywords for water are encountered, one must be defined for each
PVT region.
The densities will be read in the same order that the PVT tables were specified i.e., the first
*DENSITY keyword will correspond to the first PVT table and so on. STC means stock tank
conditions (usually 14.7 psia, 60 F). If gas gravity is entered it is converted internally to
kg/m3:
gas_density (kg/m3) = gas_gravity * 1.2222
Each of the oil, gas, and water densities must be input. Solvent density must be input when
using the pseudo-miscible option.
Example:
*DENSITY
*DENSITY
*DENSITY
*DENSITY
*OIL
38.53
*GAS
0.06864
*WATER
62.4
*SOLVENT 0.06248
The acceptable range of values for oil density is:
min
max
SI
kg/m3
250.0
1500.0
Field
lbm/ft3
15.60
93.64
Lab
g/cm3
0.25
1.5
The acceptable range of values for gas density is:
min
max
SI
kg/m3
0.05
50.0
Field
lbm/ft3
3.12E-3
3.12
Lab
g/cm3
5.0E-5
0.05
The acceptable range of values for gas gravity is:
min
max
SI
Air=1
0.05
50.0
Field
Air=1
0.05
50.0
Lab
Air=1
0.05
50.0
The acceptable range of values for water density is:
min
max
SI
kg/m3
500.0
1500.0
Field
lbm/ft3
31.21
93.64
Lab
g/cm3
0.5
1.5
User's Guide IMEX
The acceptable range of values for solvent density is:
min
max
User's Guide IMEX
SI
kg/m3
0.001
100.0
Field
lbm/ft3
6.24E-5
6.24
Lab
g/cm3
1.0E-6
0.01
Under-Saturated Oil Compressibility (Optional)
*CO
PURPOSE:
*CO indicates input of oil compressibility (for a PVT region). (Alternate method for
inputting oil compressibility)
FORMAT:
*CO oil_compress
DEFINITIONS:
oil_compress
Compressibility of the oil phase when above the bubble point pressure
(inverse pressure) (1/kPa | 1/psi | 1/kPa | 1/(kg/cm2)).
DEFAULTS:
Optional keyword. No default value.

CONDITIONS:
This keyword must be in the Component Property keyword group. This keyword must be
omitted if oil compressibility is input in the PVT table as a function of pressure, or if the
*COT or *BOT options are used.
If a single *CO keyword is encountered it will be applied to all PVT regions.
If multiple *CO keywords are encountered, one must be defined for each PVT region.
The *CO keyword cannot be used with the *GASWATER option (The oil phase is not
modelled) or the *GASWATER_WITH_CONDENSATE option (The oil phase is always
saturated with gas).
EXPLANATION:
oil_compress gives the slope of the bo curve divided by bo(Pb) when the oil is above the
bubble point pressure. In IMEX this is taken as a constant value, independent of the bubble
point pressure. Refer to Figure 7(b) in Appendix D.
bo(p) = bo(pb) * [1 - co * (p - pb)]
where pb is the bubble point pressure. The acceptable range of values for oil compressibility
is:
min
max
SI
1/kPa
0.0
1.0E-3
Field
1/psia
0.0
6.89E-3
Lab
1/Pa
0.0
1.0E-3
Mod. SI
1/(kg/cm2)
0.0
1.0E-2
User's Guide IMEX
Oil Phase Viscosity Pressure Dependence (Optional)
*CVO
PURPOSE:
*CVO signals the input of the pressure dependence of oil viscosity (for a PVT region).
FORMAT:
*CVO cviso
DEFINITIONS:
cviso
Pressure dependence of the viscosity curve for oil above the bubble point
pressure (mPa-s/kPa | cp/psi | cp/(kg/cm2).
DEFAULTS:
Optional. If *CVO is not present the default is zero.

CONDITIONS:
This keyword must be in the Component Property keyword group. The *CVO keyword
cannot be used with the *GASWATER option (The oil phase is not modelled) or the
*GASWATER_WITH_CONDENSATE option (The oil phase is always saturated with gas).
The *CVO option cannot be used if the *VOT option is used.
If a single *CVO keyword is encountered it will be applied to all PVT regions.
If multiple *CVO keywords are encountered, one must be defined for each PVT region.
EXPLANATION:
In IMEX, cviso is assumed to be a constant value, independent of the bubble point pressure.
Refer to Figure 7(b) in Appendix D. It is calculated by the equation:
viso(p) = viso + cviso * (p - pb)
where pb is the bubble point pressure.
The acceptable range of values for oil viscosity pressure dependence is:
min
max
User's Guide IMEX
SI
mPa*s/kPa
0.0
10.0
Field
cp/psi
0.0
68.9
Lab
cp/(kg/cm2)
0.0
0.102
Water Formation Volume Factor (Required)
*BWI, *CW, *REFPW
PURPOSE:
*BWI indicates the input of the water formation volume factor (for a PVT region).
*CW indicates the input of water compressibility (for a PVT region).
*REFPW indicates the input of reference pressure (for a PVT region).
FORMAT:
*BWI
*CW
*REFPW
bwi
cw
prw
DEFINITIONS:
bwi
Water formation volume factor (m3/m3 | RB/STB | cm3/cm3) at reference
pressure prw.
cw
Water compressibility (1/kPa | 1/psi | 1/kPa | 1/(kg/cm2).
prw
Reference pressure (kPa | psi | kPa | kg/cm2) at which water formation
volume factor bwi is calculated.
DEFAULTS:
Required keywords. No default values.

CONDITIONS:
These keywords must be in the Component Property keyword group.

If a single *BWI, *CW, or *REFPW keyword is encountered it will be applied to all PVT
regions.
If multiple *BWI, *CW, or *REFPW keywords are encountered, one must be defined for
each PVT region.
EXPLANATION:
The water phase formation volume factor, bw, is calculated according to the formula
bw = bwi * (1.0 - cw * (p - prw))
where p is the oil-phase pressure.
bw =
(volume of
(volume of
water at reservoir conditions )

water at standard conditions )
User's Guide IMEX
The acceptable range of values for water formation volume factor is:
min
max
SI
m3/m3
0.5
1.5
Field
RB/STB
0.5
1.5
Lab
cm3/cm3
0.5
1.5
The acceptable range of values for water compressibility is:
min
max
SI
1/kPa
0.0
1.0E-3
Field
1/psia
0.0
6.89E-3
Lab
1/Pa
0.0
1.0E-3
Mod. SI
1/(kg/cm2)
0.0
1.0E-2
Lab
1/Pa
0.0
1.0E+6
Mod. SI
1/(kg/cm2)
0.0
1.0E+4
min
max
User's Guide IMEX
SI
1/kPa
0.0
1.0E+6
Field
1/psia
0.0
145,038
Water Phase Viscosity (Required)
*VWI, *CVW
PURPOSE:
*VWI signals the input of water viscosity (for a PVT region).

*CVW signals the input of cvw (for a PVT region).
FORMAT:
*VWI
*CVW
vwi
cvw
DEFINITIONS:
vwi
Viscosity of water phase at the reference pressure prw (mPa-s | cp | cp).
cvw
Pressure dependence of water viscosity (viscosity units/pressure units).
DEFAULTS:
Required keywords. Default for *VWI is 1.0 cp. Default for *CVW is 0.0.
CONDITIONS:
These keywords must be in the Component Property keyword group. If a single *VWI or
*CVW keyword is encountered it will be applied to all PVT regions. If multiple *VWI or
*CVW keywords are encountered, one must be defined for each PVT region.
Care must be made in the choice of CVW as it is constant over pressure. Values that are too
large may cause the water viscosity to become unphysical when *REFPW is large compared
to block or bottom-hole pressure.
Typical *CVW values for pure water :
@ 77 deg. F
@ 122 deg. F
@ 167 deg. F
= 0.0 cp/psi
= 1.3e-6 cp/psi
= 1.7e-6 cp/psi
EXPLANATION:
The water viscosity, vw, is calculated according to the formula:

vw = vwi + cvw * (p - prw)
where p is the oil-phase pressure. prw is a reference pressure input in *REFPW.
The acceptable range of values for water viscosity is:
min
max
SI
mPa*s
0.1
2.0
Field
cp
0.1
2.0
Lab
cp
0.1
2.0
User's Guide IMEX
The acceptable range of values for the change in water viscosity is:
min
max
User's Guide IMEX
SI
mPa*s
mPa*s/kPa
0.0
1.0E-3
Field
cp
cp/psi
0.0
6.89E-3
Lab
cp
cp/kPa
0.0
1.0E-3
Mod. SI
mPa*s
mPa*s/(kg/cm2)
0.0
1.0E-1
PVT Type (Optional)
*PTYPE
PURPOSE:
*PTYPE indicates the start of input of PVT region types. A PVT table set and if specified, a
bubble point pressure vs. depth table is assigned to each grid block (see *PVT and *PBT).
ARRAY:
*PTYPE
DEFAULTS:
Optional keyword. The default is *PTYPE *CON 1, where 1 is a value for set_number. This
corresponds to one PVT region only.
CONDITIONS:
This keyword must be in the Component Property keyword group, after all the PVT data
tables have been entered.
EXPLANATION:
If the reservoir PVT properties differ from region to region, the reservoir can be subdivided
further by assigning a PVT table to each subregion. If more than one set of PVT tables is
input, then a set of tables must be assigned to each grid block.
*PTYPE indicates the input of an array of grid values. All array reading option subkeywords
are valid. See the section on Array Reading Options for more details.
User's Guide IMEX
Maximum Rate of Increase of Solution Gas Ratio or Bubble

Point Pressure (Optional)
*DRSDT, *DPBDT
PURPOSE:
*DRSDT signals the input of the maximum rate of increase of solution gas ratio (for a PVT
region). *DPBDT signals the input of the maximum rate of increase of bubble point pressure
(for a PVT region).
*DRSDT and *DPBDT limits are not applied to reductions of solution gas ratio or bubble
point pressure (respectively).
FORMAT:
*DRSDT (*FREEGAS) value_drsdt

or
*DPBDT (*FREEGAS) value_dpbdt
DEFINITIONS:
value_drsdt
Maximum rate of increase of solution gas ratio. (m3/m3/day | scf/STB/day |
cm3/cm3/min | m3/m3/day).
value_dpbdt
Maximum rate of increase of bubble point pressure. (kPa/day | Psi/day |
kPa/minute | kg/cm2/day).
*FREEGAS
If the *FREEGAS sub-keyword is present, the DRSDT or *DPBDT limit is
only applied to blocks which contain free gas (gas saturation > 0.0).
If *FREEGAS is not present, the *DRSDT or *DPBDT limit is applied to all
blocks, whether gas exists in the block or not, and if the *DRSDT or
*DPBDT limit is exceeded in a block without free gas, free gas will
immediately appear in that block.
DEFAULTS:
Optional. If *DRSDT or *DPBDT is not present in a data set, the maximum rate of solution
of gas in oil is infinite.
CONDITIONS:

*DRSDT or *DPBDT has no effect on reductions of solution gas ratio or bubble point
pressure (respectively).
The *DRSDT or *DPBDT keywords cannot be used with the API tracking options.
If a single *DRSDT or *DPBDT keyword is encountered, it will be applied to all PVT
regions.
User's Guide IMEX
If multiple *DRSDT keywords are encountered, one must be defined for each PVT region.
If multiple *DPBDT keywords are encountered, one must be defined for each PVT region.
*DRSDT and *DPBDT cannot be used in different PVT regions. All PVT regions must
either use *DRSDT or *DPBDT.
Then using *DRSDT or *DPBDT, the *FREEGAS sub-keyword must appear in either all or
none of the PVT regions. If this is not true, the last PVT region defined will be used to
determine whether or not the *FREEGAS option is used in all PVT regions.
Value_drsdt or value_dpbdt cannot be less than zero. If these values are equal to zero or
small, a gas phase can occur when undersaturated oil moves into cells of even more
undersaturated oil (lower Rs). The *FREEGAS option may be used to prevent this from
occurring.
*DPBDT can be used with the volatile oil option. *DRSDT can be used with the volatile oil
option provided that dRs/dPb in the *PVTCOND or *PVTVO table is always positive.
EXPLANATION:
In IMEX, by default, when bubble point pressure or solution gas ratio rises, gas is allowed to
dissolve in oil in an unrestricted manner. *DRSDT/*DPBDT can be used to limit the rate at
which gas can dissolve in the oil.
In blocks with free gas, if *DRSDT or *DPBDT is exceeded, this will result in what is
effectively a bubble point pressure lower than the block pressure.
In blocks without free gas, if *DRSDT or *DPBDT is exceeded, this will result in the
appearance of a gas phase at the current bubble point pressure/solution gas ratio present in the
block.
It is possible when using *DRSDT or *DPBDT that the material balance deteriorates over
time when the timestep converges in a single Newton iteration. The material balance
equations may have just converged when a *DRSDT limit has been applied after the first
iteration. Rather than force a tighter MAXRES convergence tolerance to control this, the
keyword MINCYC has been added in the NUMERICAL section to give the user control of
the minimum number of Newton iterations per timestep. MINCYC 2 will overcome this
issue.
Example:
*DRSDT
0.45 ** scf/stb/day
User's Guide IMEX
Initial Gas-Oil Ratio (Conditional)
*GORINT
PURPOSE:
*GORINT signals the input of initial gas-oil ratio.

FORMAT:
*GORINT gor
DEFINITIONS:
gor
Initial gas-oil ratio used with *MISCG model (m3/m3 | scf/STB | cm3/cm3).
DEFAULTS:
Conditional keyword. No default value.

CONDITIONS:
This keyword must be in the Component Property keyword group. Required with *MISCG.
EXPLANATION:
This keyword is used to account for the solution gas dissolved in the oil phase when running
with chase- gas injection (*MISCG).
Note that mass conservation of initial solution gas is not accounted for in such simulation,
hence it is assumed that pressure will not fall below saturation pressure of the initial solution
gas.
The acceptable range of values for initial gas-oil ratio is:
min
max
User's Guide IMEX
SI
m3/m3
0.0
1.0E+3
Field
scf/STB
0.0
5,617
Lab
cm3/cm3
0.0
1.0E+3
Gas and Solvent Mixing Parameter (Conditional)
*OMEGASG
PURPOSE:
*OMEGASG signals the input of gas and solvent mixing parameter.

FORMAT:
*OMEGASG omega_sg
DEFINITIONS:
omega_sg
Gas and solvent mixing parameter (fraction).
DEFAULTS:

CONDITIONS:
or *MISNCG.
EXPLANATION:
When there is free gas (chase gas or original gas- in-place) this parameter determines the
mixing of free gas with solvent. The value of omega_sg is bounded by 0 and 1 where 0 is the
immiscible case and 1 is the completely miscible case as for omega_os. Typically omega_sg
should be greater than or equal to the maximum value of omega_os that is entered with *PVT.
Recall, for *MISCG, there is chase gas and the original gas-in-place can only be in solution.
If *MISNCG, then there is no chase gas and the original gas-in-place can be in the form of
free gas as well as solution gas.
The acceptable range of values for the gas and oil mixing parameter is:
min
max
SI
Field
Lab
0.0
1.0
0.0
1.0
0.0
1.0
User's Guide IMEX
Minimum Solvent Saturation (Conditional)
*MINSS
PURPOSE:
*MINSS signals input of minimum solvent saturation.

FORMAT:
*MINSS min_sol_sat
(SGTHRESH sgval)
(*SMOOTHEND (*ON|*OFF))
DEFINITIONS:
min_sol_sat
Minimum solvent saturation (fraction).
sgval
Gas saturation threshold. Loss of miscibility if Sg > Sgthresh and Ss < Minss.
*SMOOTHEND
Enable or disable (*ON|*OFF) end point smoothing for loss of miscibility
due to MINSS, SGTHRESH and/or SORM. Note SORM, is an optional part
of the *SWT table.
DEFAULTS:
Conditional keyword. If *MINSS/*SGTHRESH loss of miscibility smoothing is on, then

1. *MINSS has no default.
2. *MINSS is reset to 0.001 if a value less than 0.001 is entered.
3. *SGTHRESH is reset to 0.001 if a value less than 0.001 is entered.
4. If *SGTHRESH is not found in the data, a default value of 0.01 is applied.
If *MINSS/*SGTHRESH loss of miscibility smoothing is off, then
1. *MINSS has no default.
2. If *SGTHRESH is not found in the data, a default value of 0.01 is applied.
By Default, loss of miscibility end point smoothing (*SMOOTHEND *ON or
*SMOOTHEND) is enabled.
CONDITIONS:
or *MISNCG.
EXPLANATION:
The minimum solvent saturation is the solvent saturation in the presence of gas below which
mixing is not possible.
Sorm (part of the *SWT table) is the saturation of oil where mixing is not possible due to
water blocking.
User's Guide IMEX
The acceptable range of values for the minimum solvent saturation and also for the gas
saturation threshold is:
min
max
SI
Field
Lab
0.0
0.5
0.0
0.5
0.0
0.5
User's Guide IMEX
Polymer Adsorption Table (Conditional)
*PADSORP
PURPOSE:
*PADSORP signals start of polymer adsorption table.

TABLE:
*PADSORP
p_con
:
adsorp_level
:
DEFINITIONS:
p_con
Polymer concentration (kg/m3 | lb/STB | g/cm3).
adsorp_level
Polymer adsorption level (kg/m3 | lb/STB | g/cm3).
DEFAULTS:

CONDITIONS:
This keyword must be in the Component Property keyword group. Required for *POLY.
EXPLANATION:
This is the table input of polymer adsorption as a function of concentration subject to the
permeability dependent maximums and residual levels *PPERM.
For dual porosity systems, only one table is required.
Example:
*PADSORP
** p_con
0.0
0.125
0.25
0.37
0.50
adsorp_level
0.0
0.035
0.075
0.09
0.10
NOTE:
Normally lab data for adsorption are expressed as mass of polymer adsorbed per mass of
reservoir rock. To convert the lab values of absorption to those required by IMEX multiply
lab values by the density of the rock and by the factor
(1 ) /
User's Guide IMEX
The factor
(1 ) /
converts from rock volume to pore volume. Typical values of rock density vary between
2000 kg/m3 and 2500 kg/m3.
User's Guide IMEX
Polymer Viscosity Mixing (Conditional)
*PMIX
PURPOSE:
*PMIX indicates the viscosity mixing model for the polymer solution is being defined.
FORMAT:
*PMIX
(*LINEAR)
(*NONLINEAR)
(*TABLE)
(*VELTABLE)
DEFINITIONS:
*LINEAR
Use linear mixing.
*NONLINEAR
Use nonlinear mixing
*TABLE
Use input table to calculate the polymer-water mixture viscosity. A
dimensionless, two column table must follow:
relative concentration
viscosity ratio
:
:
The first column is the relative concentration which is the ratio of local
polymer concentration to the reference polymer concentration (given by
*PREFCONC). The second column is the ratio of polymer viscosity to water
viscosity.
*VELTABLE
Use an input table to calculate the polymer-water mixture viscosity as a
function of both polymer concentration and mixture velocity. The keyword
*VWT followed by a velocity value (*VWT vel) must follow on a separate
line. A dimensionless, two column table must follow the velocity value on a
separate line. A new occurrence of *VWT vel signals the start of another
dimensionless table at a new velocity:
*VWT vel1
:
*VWT vel2
:
User's Guide IMEX
viscosity ratio
:
viscosity ratio
:
*VWT veln
viscosity ratio
:
:
The first column is the relative concentration which is the ratio of local
polymer concentration to the reference polymer concentration (given by
*PREFCONC). The second column is the ratio of polymer-water mixture
viscosity to pure water viscosity.
Vel1 Veln
Value of reservoir condition velocity. (m/day | ft/day | cm/min | m/day)
DEFAULTS:
Conditional keyword. No default.

CONDITIONS:
This keyword must be in the Component Property keyword group. Required with *POLY. If
*PMIX *TABLE is used, the reference polymer viscosity (*PVISC) becomes unnecessary
and will be ignored if it is given. The relative concentration must start with zero and values
can be no larger than one.
EXPLANATION:
When *PMIX *LINEAR is chosen, the polymer solution is calculated by the equation:
vsw(p) = alfa * visrp + (1 - alfa) * vsw(p)
When *PMIX *NONLINEAR is entered, the polymer solution is calculated by the equation:
vsw(p) = visrp**alfa * vsw(p)**(1 - alfa)
The variable visrp is the reference polymer solution viscosity (*PVISC) and alfa is the
relative concentration:
alfa = cp/crp
Where cp is the local polymer concentration and crp is the reference polymer concentration
(*PREFCONC).
When *PMIX *TABLE is used, the viscosity of polymer solution is given by:
vsw(p) = visra(alfa) * vsw(p)
Corresponding to alfa, the visra is the viscosity ratio obtained from linear table interpolation
or extrapolation.
When *PMIX *VELTABLE is encountered, a two dimensional interpolation is used. Waterpolymer mixture velocity is calculated for each gridblock. The velocity is divided by
gridblock porosity and multiplied by a correction factor (*FRWIDTHCOR), or:
Velcor = FRWIDTHCOR * Reservoir Condition Velocity/
Where Velcor is the corrected velocity used in the VELTABLE table, FRWIDTHCOR is the
fracture width correction term entered on a block by block basis, Velocity is the average
velocity in a block and is the blocks porosity
User's Guide IMEX
This reservoir condition velocity (Velcor) is used to interpolate between the different *VWT
tables. The viscosity ratios are then determined by additionally interpolating with respect to
the relative polymer concentration.
If a single *VWT table is encountered, the single table is used for all values of velocity. If a
velocity is less than the velocity on the first *VWT keyword, the first table will be used. If a
velocity greater than the velocity on the last *VWT table, the last table will be used.
If a relative concentration is less than the lowest relative concentration found in a *VWT
table, the lowest tabular value will be used. If a relative concentration is greater than the
largest relative concentration found in a *VWT table, the largest tabular value will be used.
The relative polymer concentration cannot be less than zero or greater than 1.0.
Example:
** Velocity Dependent Table at Velocity 0.10, 10.0 and
40.0 ft/day
*PMIX *VELTABLE
*VWT 0.10
0.0
1.00
0.2
1.50
0.6
2.00
1.0
5.00
*VWT 10.0
0.0
1.00
0.2
1.20
0.6
1.75
1.0
3.00
*VWT 40.0
0.0
1.00
0.2
1.10
0.6
1.40
1.0
2.00
User's Guide IMEX
Reference Polymer Viscosity (Conditional)
*PVISC
PURPOSE:
*PVISC signals the input of reference polymer viscosity.

FORMAT:
*PVISC value
DEFINITIONS:
value
Reference polymer solution viscosity (mPa-s | cp | cp).
DEFAULTS:

CONDITIONS:
This keyword must be in the Component Property keyword group. Required with *POLY or
*POLYOW.
EXPLANATION:
The reference polymer viscosity is the polymer viscosity at the reference concentration
(*PREFCONC).
The acceptable range of values for the polymer viscosity is:
min
max
SI
mPa*s
0.5
10.0
Field
cp
0.5
10.0
Lab
cp
0.5
10.0
User's Guide IMEX
Reference Polymer Concentration (Conditional)
*PREFCONC
PURPOSE:
*PREFCONC signals the input of reference polymer concentration.

FORMAT:
*PREFCONC value
DEFINITIONS:
value
Reference polymer concentration. It must be greater than or equal to the
polymer injection concentration (kg/m3 | lb/STB | g/cm3)
DEFAULTS:

CONDITIONS:
*POLYOW.
EXPLANATION:
The reference concentration is used to weight the polymer concentration for the viscosity
mixing rule. (See equations E.6 and E.7 in Appendix E).
The value for *PREFCONC must be greater than zero since it is used to calculate the
numerical shift, the tolerance (for Newtonian iteration), and the threshold for the threshold
switching criteria.
The acceptable range of values for the polymer concentration is:
min
max
User's Guide IMEX
SI
kg/m3
0.0
29.0
Field
lb/STB
0.0
10.0
Lab
g/cm3
0.0
0.029
Polymer Permeability Table (Conditional)
*PPERM
PURPOSE:
*PPERM signals the start of table of absolute permeability dependent polymer properties.
TABLE:
*PPERM
perm
:
max_ad
:
res_ad
:
p_pore
:
rrf
:
DEFINITIONS:
perm
Absolute permeability (md | md | md)
max_ad
Maximum adsorption capacity (kg/m3 | lb/STB | g/cm3).
res_ad
Residual sorption level (kg/m3 | lb/STB) | g/cm3).
p_pore
Polymer accessible pore volume (fraction).
rrf
Residual resistance factor, >= 1.0.
DEFAULTS:

CONDITIONS:
*POLYOW.
EXPLANATION:
The maximum adsorption capacity is the capacity of the rock to absorb the polymer. The
residual sorption level is the amount of polymer left in the rock after injected water washes
some of the polymer away from the rock.
Example:
*PPERM
**
perm
10.0
1000.0
max_ad
0.3
0.2
res_ad
0.15
0.10
p_pore
0.95
1.0
rrf
1.2
1.2
User's Guide IMEX
Reference Seawater Viscosity (Conditional)
*SVISC
PURPOSE:
*SVISC signals the input of the reference Seawater viscosity.

FORMAT:
*SVISC value
DEFINITIONS:
value
Reference Seawater viscosity (mPa-s | cp | cp).
DEFAULTS:
Conditional keyword. Default is formation water viscosity of PVT region 1.

CONDITIONS:
This keyword must be in the Component Property keyword group. Only used with
*BLACKOIL_SEAWATER or *OILWATER_SEAWATER Model.
EXPLANATION:
The reference seawater viscosity is the viscosity of 100% seawater at pressure *REFPW.
In the Seawater model, both the reference pressure *REFPW and the pressure dependence of
Seawater viscosity *CVW are the values entered in PVT region 1.
Pure seawater viscosity (Vsw) is calculated as:
Vsw = vswi + cvw*(p - prw)
Where vswi is the reference seawater viscosity enters via *SVISC, cvw is the pressure
dependence of water viscosity with respect to pressure entered via *CVW (for PVT region 1)
and prw is the reference pressure of water entered via *REFPW (for PVT region 1)
The mixed seawater/formation water viscosity (Vw) is calculated as:
Vw = alfa*vsw + (1-alfa)*vfw
Where alfa is the seawater volume fraction and vfw is the formation water viscosity
The acceptable range of values for the seawater viscosity is:
min
max
User's Guide IMEX
SI
mPa*s
0.5
10.0
Field
cp
0.5
10.0
Lab
cp
0.5
10.0
Rock-Fluid Data
Notes on Rock-Fluid Data

1. Multiple rock types are allowed in IMEX. Rock types are assigned to grid blocks
using *RTYPE. Relative permeabilities of each rock type can be entered using tables.
The following keywords are used:
*RPT rock_type_number
followed by the tabular values for that rock type introduced by
*SWT and *SGT or *SLT.
2. Hysteresis effects are allowed on the oil-water capillary pressure, Pcow, the oil-gas
capillary pressure, Pcog, the gas phase relative permeability, krg and the oil phase
relative permeability, krow.
Capillary pressure hysteresis is modelled using an approach similar to that of
Killough, 1976. To invoke Pcow hysteresis enter both the drainage and imbibition
Pcow table columns and optionally epspc using *EPSPC. To invoke Pcog
hysteresis enter both the drainage and imbibition Pcog table columns and
optionally epspcg using *EPSPCG. To invoke krg hysteresis enter the keyword
*HYSKRG. To invoke krow hysteresis enter the keyword *HYSKRO.
Trapped oil hysteresis is also available for Pcow and Krow. To invoke the trapped
oil hysteresis, it is only necessary to give the imbibition Pcow in SWT table with
constant Pcow for water saturation points equal to and great than 1-Sormax. Once
activated, trapped oil hysteresis couples Pcow and Krow hysteresis. Keyword
*HYSKRO is not necessary and will be ignored. The *EPSPC is still used when
creating Pcow scanning curves.
3. Relative permeability tables may be scaled to different end points for each grid block.
This is done by entering one or more of the following end points for each grid block.
SWCON - The connate water saturation
SWCRIT - The critical water saturation
SGCON The connate gas saturation
SGCRIT - The critical gas saturation
SOIRW Irreducible oil saturation (oil-water table)
SLCON The connate liquid saturation (gas-liquid table)
SORW The residual oil saturation (water-oil table)
User's Guide IMEX
Rock-Fluid Data 515
SORG - The residual oil saturation (gas-liquid table)

SORMAX - The maximum residual oil saturation (water-oil table, imbibition)
KRWIRO - Water rel. perm. at irreducible oil (oil-water table)
KROCW Oil rel. perm. at connate water (oil-water table)
KRGCL Gas rel. perm at connate liquid (liq.-gas table)
KROGCG Oil rel. perm at connate gas (liq.-gas table)
PCWMAX Oil water capillary pressure at connate water
PGGMAX Gas oil capillary pressure at connate liquid
JFWMAX Oil water J function at connate water
JFGMAX Gas-oil J function at connate liquid
4. Rock wettability may be changed by using the *OILWET keyword. The
*OILWET keyword makes the rock oil wet and water becomes the phase with
intermediate wettability.
5. When three phases are present the default in IMEX is to evaluate Kro using Stone's
second model as normalized by Aziz and Settari. The Krow and Krog values
which enter the Stone's second model formula are evaluated as functions of Sw and
Sg, respectively. There are several other methods for computing Kro which can be
invoked. The other models available are Stone's first model, the Linear Isoperm
Method and a segregated model. For both Stone's first and second models, Krow
and Krog can be looked up as functions of So rather than Sw and Sg. Please see
the entry under the *KROIL keyword for more information.
6.
For relative permeabilities, smoothing is generally done for kr close to zero.

Various smoothing options are available for these cases. The options may be
specified using *SMOOTHEND and related keywords. The default is quadratic
smoothing between kr=0 and the first nonzero table entry for kr.
7. Options are available to ensure smoothness and remove inflection points in the
relative permeabilities and capillary pressures. These options are defaulted off. If
poor timestep behavior occurs with consistently many Newton iterations per
timestep and/or many timestep cuts, these options may improve performance. Use
of these options may alter the shape of the relative permeability and capillary
pressure curves.
8. The smoothing option discussed above can be also used in situations where actual
rock data is sparse to generate curves which produce a best fit to analytical power
law functions.
9. J Functions can be input in place of capillary pressures. See the keywords
*JFUNC, *SRFTNW and *SRFTNG for details.
10. Non-Darcy flow in the reservoir can be modelled by using the keywords
*NONDARCY and *NDARCYCOR. See these keywords for details. Non-Darcy
flow in the well is modelled using the *TURB option on the *PERF (*PERFV,
*PERVHY, *PERFRG) keyword.
516 Rock-Fluid Data
User's Guide IMEX
Start of Rock-Fluid Property Input (Required)
*ROCKFLUID
PURPOSE:
*ROCKFLUID indicates the start of the rock-fluid data.
FORMAT:
*ROCKFLUID
DEFAULTS:
CONDITIONS:
This keyword must be the first keyword in the Rock-Fluid Data keyword group. ROCKFLUID DATA must follow immediately after component properties.
EXPLANATION:
Three phase relative permeabilities are computed using modified Stone's Model II (by
default).
The relative permeability of water in a three phase system is the same as that in a two phase
water-oil system, and is a function of Sw (water saturation) only. The relative permeability
of gas in the three phase system is equal to the gas relative permeability in the two phase
liquid-gas system, and is a function of Sg (gas saturation) only. Four methods are available
for calculating the three phase oil relative permeability, kro. These are: Stone's method 1;
Stone's method 2 as modified by Settari and Aziz (Aziz, K., and Settari, A., "Petroleum
Reservoir Simulation", Applied Science Publishers Ltd., London, 1979); the Linear Isoperm
method and the Segregated method.
The two phase relative permeability curves are entered in tabular form. The various saturation
end points, such as connate water, are determined internally by scanning the krw table entries.
Hysteresis effects upon water-oil capillary pressure, oil-gas capillary pressure, gas relative
permeability and oil (krow) relative permeability can be included.
User's Guide IMEX
Rock-Fluid Data 517
Relative Permeability Curves by Table (Required)
*RPT
PURPOSE:
*RPT indicates that this set of relative permeability curves will be defined by table entries.
FORMAT:
*RPT set_number
(*OILWET)
(*DRAINAGE)
(*IMBIBITION) (*PCOW | *PCOG | *BOTH)
(*SCALING-OLD)
DEFINITIONS:
set_number
Set_number for this set of water-oil and liquid- gas relative permeability
tables. This is the number used with *RTYPE to assign relative permeability
curves to grid blocks.
*OILWET
Keyword indicating that the wetting phase is oil. If not present, assumed
water wet. Hysteresis Please refer to the tutorial section for a further
discussion of the oil wet option used in IMEX.
*DRAINAGE, *IMBIBITION
Hysteresis subkeywords for capillary pressure -- if *DRAINAGE is specified
the drainage capillary pressure curve for both Pcow and Pcog is used
initially; if *IMBIBITION is specified the imbibition curve for both Pcow
and Pcog is used initially. At most one of these subkeywords may be
entered; if neither is entered *DRAINAGE is assumed. Hysteresis effects
may also be included for gas and oil relative permeability. For the gas and
oil relative permeability the drainage curve is always used initially.
*PCOW
Sub-keyword of the *IMBIBITION keyword. If present, the imbibition oilwater capillary pressure curve and the drainage oil-gas capillary pressure
curve are used initially.
*PCOG
Sub-keyword of the *IMBIBITION keyword. If present, the drainage oilwater capillary pressure curve and the imbibition oil-gas capillary pressure
curve are used initially.
*BOTH
Sub-keyword of the *IMBIBITION keyword. If present, the imbibition oilwater capillary pressure curve and the imbibition oil-gas capillary pressure
curve are used initially. This is the same behavior as when *IMBIBITION is
not followed by a sub-keyword
518 Rock-Fluid Data
User's Guide IMEX
*SCALING-OLD
This keyword allows the user to revert back to the older end point scaling
option employed in IMEX prior to the 2000 release. Currently the model
uses eight end points, the old model used four end points. The new model is
a more consistent representation of end point scaling.
DEFAULTS:
Required keyword. No default. If not otherwise specified, the *RPT table is assumed to be a
drainage table for water wet rock.
CONDITIONS:
This keyword must be in the Rock-Fluid Data keyword group. It should be immediately after
the *ROCKFLUID keyword before the specification of relative permeabilities and capillary
pressure data. One *RPT keyword is necessary for each rock type whose properties are
entered in tabular form.
EXPLANATION:
When three phases are present the default in IMEX is to evaluate Kro using Stone's second
model as normalized by Aziz and Settari. The other models available are Stone's first model,
the Linear Isoperm Method and a segregated model. Please see the entry under the *KROIL
keyword for more information. These models calculate the three phase relative permeability
curves from the two phase curves that are input.
There must be both a water-oil relative permeability table and a liquid-gas relative
permeability table immediately following this card except when using the OILWATER
option, the OILWATER option does not require the input of a gas-liquid table.
Set_numbers are used to assign different relative permeability curve sets to different rock
types in the reservoir. If only one set of relative permeability curves is used, the set_number
may be omitted since the default is 1.
If there is more than one rock type, the first relative permeability set must have a set_number
of one and set_numbers must increase consecutively for subsequent table sets. Set_numbers
are assigned to grid blocks with the *RTYPE keyword.
Each *RPT keyword must be immediately followed by a *SWT keyword and table, then a
*SLT or *SGT keyword and table. Rock type dependent keywords are input after the SWT
and SLT (*SGT) tables, they apply to the rock type defined above. After a new *RPT
keyword is input a new set of tables and rock type dependent keywords can be read in.
The *OILWET option specifies oil as the wetting phase and water as the intermediate wetting
phase. Thus for this option the water phase pressure is higher than the oil phase pressure.
User's Guide IMEX
Rock-Fluid Data 519
Water-Oil Relative Permeability Table (Conditional)
*SWT
PURPOSE:
*SWT indicates the start of the water-oil relative permeability table. In addition *SWT
controls the use of a wide variety of relative permeability regression and optimization
functions.
In its most basic form (to define the beginning of a table):
TABLE:
*SWT (*SMOOTHEND Subkeywords) (*SWTKRTHR krthr_value)
Sw
krw
krow
(Pcow)
(Pcowi or Sorm)
:
:
:
:
:
*SWT subkeywords provide the user with a number of unique features which can:
1. Change the shape of the curves as they go to zero.
2. Create new curves obtained from nonlinear regression of available data.
3. Allow the user to interact with the regression procedure to specify the range over
which regression is performed.
4. Regenerate equally spaced tables using a user defined Interpolation interval.
See below for its most complete form:
DEFINITIONS:
(Basic Form)
TABLE:
*SWT (*SMOOTHEND Subkeywords) (*SWTKRTHR krthr_value)
Sw
krw
krow
(Pcow)
(Pcowi or Sorm)
:
:
:
:
:
End-Point Smoothing Options:
*SMOOTHEND
Optional keyword indicating what type of smoothing is to be used for the
interval with kr=0 and kr>0 for krw and krow. These are the table intervals
where water or oil has just become mobile. The keyword may be followed
by a subkeyword as indicated below.
Subkeywords of the *SMOOTHEND Keyword:
*OFF *ON *QUAD *CUBIC
If *SMOOTHEND is not specified or it has no subkeyword, the default is
*SMOOTHEND *ON. This is equivalent to *SMOOTHEND *QUAD.
cubic smoothing is used. For *OFF, no smoothing (that is, linear interpolation)
is used.
520 Rock-Fluid Data
User's Guide IMEX
Critical End-Point Threshold Option:

*SWTKRTHR
Optional keyword to use krthr_value as a threshold value for krw and krow
for end-point determination.
krthr_value
If a tabular value for krw or krow is less than or equal to krthr_value then it
is set to zero. If *SWTKRTHR is not specified, then a default value of 5.0E07 is used for krthr_value. Entered values of krthr_value are reset to lie
within the range 5.0E-16 to 5.0E-07.
Sw
Water saturation (fraction). There is no default value.
krw
Relative permeability to water (fraction) at the given water saturation. There
is no default value.
krow
Relative permeability to oil (fraction) in the presence of the given water
saturation. There is no default value.
Pcow
Water-oil capillary pressure (kPa | psi | kPa | kg/cm2) on the drainage curve.
This column is optional; when omitted it is defaulted to zero and no capillary
pressure effects are included in the simulation. This column is required if Pcowi
or Sorm are to be entered i.e. the fourth column is always assumed to be Pcow.
Pcowi
Water-oil capillary pressure (kPa | psi | kPa | kg/cm2) on the imbibition
curve. This column is optional; if Pcow is not entered then Pcowi must also
not be entered (the fourth column is always assumed to be Pcow). If Pcowi
is not entered it is set equal to Pcow by default. Hence if the user does not
enter Pcowi, no hysteresis effects are included in capillary pressure; to
include hysteresis effects, the user enters values in this column. Hysteresis
effects CANNOT be modelled when using the pseudo-miscible option.
Sorm
Irreducible oil saturation (fraction) due to water blocking at the given water
saturation. (Only required for pseudo-miscible option.) There is no default
value. If the pseudo-miscible option is used then the fifth column is assumed
to be Sorm and capillary pressure hysteresis effects are not modelled.
User's Guide IMEX
Rock-Fluid Data 521
DEFAULTS:
Conditional keyword. There are no defaults.
CONDITIONS:
This keyword must be in the Rock-Fluid Data keyword group, immediately after the *RPT
keyword. *SWT is required with *RPT.
EXPLANATION:
Capillary pressure curves determine the height of the transition zones calculated by the
*VERTICAL option. Nonzero capillary pressure curves also influence the width of the
transition zones near sharp saturation fronts. This front widening mechanism tends to have a
stabilizing effect during the displacement of less mobile fluids (oil) by more mobile fluids
(water). The water-oil capillary pressure is defined as
Pcow = Po - Pw
Since water is assumed to be the wetting phase, the capillary pressure should be monotonically
declining with increasing water saturation. However, the Pcow value at Sw=1 can be nonzero.
Capillary pressure need not be entered, unless you are using the pseudo-miscible option with nonzero Sorm (and even then Pcow can be explicitly defined as 0.0). Otherwise the default is zero.
Entries must be in order of increasing water saturation. The first table entry must be for
Sw = Swcon (i.e. connate water saturation). It is not unusual to encounter water zones in
simulation studies, thus it is recommended that the last table entry have Sw = 1, with Krw = 1
and Krow = 0 (see Figure 8 in Appendix D).
THE *OILWET OPTION:
When the oil wet option is used the water-oil relative permeability table is altered. The Sw
column contains the wetting-phase saturation (oil saturation). The Krw column contains the
wetting-phase relative permeability (oil relative permeability). The Krow column contains
the nonwetting phase relative-permeability (water relative permeability). The Pcow column
now contains the capillary pressure between the wetting and nonwetting phases. Thus for the
oil wet system:
Pwater = Poil + Pcow
where:
Pwater = water phase pressure
Poil = oil phase pressure
Pcow = capillary pressure entered in the Pcow column.
By contrast for a water wet system:
Pwater = Poil - Pcow
Note that the values entered in the Pcow column are normally positive as the capillary
pressure is still the positive difference in pressure between non-wetting water and wetting oil.
Please refer to the tutorial section for a further discussion of the oil wet option used in IMEX.
522 Rock-Fluid Data
User's Guide IMEX
THE *GASWATER OPTION:

When using the Gas Water option the *SWT table is only altered slightly. Krow is not read
in, the Pcow column (if present is read as Pcgw) and the Pcowi column is not read. Thus a
typical Gas Water *SWT table would be:
*SWT
Sw
:
krw
:
(Pcgw)
:
It is important to ensure that the final value in the *SWT table, when using the
*GASWATER option is Sw =1.0, as the irreducible oil saturation must be 0.0. Pcgw is
defined as:
Pcgw = Pg - Pw
THE *GASWATER_WITH_CONDENSATE OPTION:
When using the *GASWATER_WITH_CONDENSATE model, the Pcow column, if present,
is read as Pcgl (capillary pressure between gas and liquid). If condensate appears in the
reservoir as liquid its saturation is added to the water saturation when Pcgl is determined. For
the condensate option, the output of Pcgl is sent to the Pcog output array (both for ascii and
SR2 output). The Pcow output array is zero. The condensate option initializes the reservoir in
fashion identical to the *GASWATER option. The *VOLATILE_OIL option does not alter
the meaning of Pcow, capillary pressures are treated normally and initialization is handled in
a fashion similar to the black oil model.
The acceptable range of values for water saturation is:
min
max
SI
Field
Lab
0.0
1.0
0.0
1.0
0.0
1.0
The acceptable range of values for water relative permeability is:
min
max
SI
Field
Lab
0.0
1.0
0.0
1.0
0.0
1.0
The acceptable range of values for oil relative permeability is:
min
max
SI
Field
Lab
0.0
1.0
0.0
1.0
0.0
1.0
The acceptable range of values for capillary pressure is:
min
max
User's Guide IMEX
SI
kPa
-1.0E+4
1.0E+4
Field
psi
-1450.0
1450.0
Lab
kPa
-1.0E+4
1.0E+4
Mod. SI
kg/cm2
-1.0E+2
1.0E+2
Rock-Fluid Data 523
The acceptable range of values for irreducible oil saturation is:
min
max
SI
Field
Lab
0.0
1.0
0.0
1.0
0.0
1.0
General Form of the *SWT keyword:

TABLE:
*SWT (*SMOOTHEND (*QUAD or *CUBIC or
*POWERQ or *POWERC or *OFF or *ON))
(*SMOOTH (*ALL or *ALLPL or *ALLINT or *ALLLIN or
or *NONE or *OFF or property_list))
(*PLAW pkrw pkrow (ppcowd (ppcowi or plsorm))
(*DSWTI dswti_value) (*SWTKRTHR krthr_value)
Sw
:
krw
:
krow
:
(Pcow)
:
(Pcowi or Sorm)
:
Uses of the General Form:

Definitions of individual subkeywords will follow, but it is useful to first highlight a few of
the many tasks the *SWT keyword can perform in situations where rock data is sparse and
the user wants to generate meaningful curves.
1. Using sparse data, generate curves which produce the best fit to an analytical
power-law model over the entire range of data.
*SWT *SMOOTH *ALL or SWT *SMOOTH *ALLPL
2. Using Sparse data, generate curves which produce the best fit to an analytical
power-law model over a smaller range of data which ensures matching of the
original data at critical points and when Pcow -> 0.0 (or becomes constant).
*SWT *SMOOTH *ALLINT
3. Using no data, specify power law curves explicitly using input power law
exponents and a two row relative perm. table (to define saturation end points and
relative permeability end points)
*SWT *SMOOTH *PLAW water_exponentoil_exponent pcow_exponent
4. Re-interpolate an existing table using equal spacing of approx. 0.01 to take
advantage of faster table look-up procedures.
*SWT *SMOOTH *ALLLIN *DSWTI 0.01
5. Alter the way the interpolation is performed in the table interval between kr=0 and
the first nonzero entry for kr. For Relative permeability, this is at Swcrit (for Krw)
and at Sorw (for Krow).
*SWT *SMOOTHEND *CUBIC
524 Rock-Fluid Data
User's Guide IMEX
DEFINITIONS:
(General Form which includes all Regression Options)
The *SWT keyword can be subdivided into 5 subkeywords
*SMOOTHEND - Smooth End-points of tables
*SWTKRTHR - Water Table Critical End-point Threshold
*SMOOTH- Regression Options
*DSWTI - Table Regeneration Options
*PLAW - Direct Power Law Input (Regression not Used)
*SWT
(*SMOOTHEND subkeywords)
(*SWTKRTHR krthr_value) (*SMOOTHsubkeywords)
(*DSWTI dswti_value)
(*PLAW subkeywords)

*SMOOTHEND
interval with kr=0 and kr>0 for krw and krow. These are the table intervals
where water or oil has just become mobile. The keyword may be followed
by a subkeyword as indicated below.
*OFF *ON *QUAD *CUBIC *POWERQ *POWERC
If *SMOOTHEND is not specified or it has no subkeyword, the default is
*SMOOTHEND *ON. This is equivalent to *SMOOTHEND *POWERQ.
cubic smoothing is used. For *OFF, no smoothing (that is, linear
interpolation) is used.
For *POWERQ and *POWERC end point smoothing is done using power
law if the exponent is between 1.5 and 4.0. Otherwise, for *POWERQ,
quadratic is used and for *POWERC, cubic is used for end point smoothing.
If the *SMOOTH regression option is not used, *POWERC And *POWERQ
default to *CUBIC and *QUAD respectively.
Critical End-Point Threshold Option:
*SWTKRTHR
Optional keyword to use krthr_value as a threshold value for krw and krow
for end-point determination.
User's Guide IMEX
Rock-Fluid Data 525
krthr_value
If a tabular value for krw or krow is less than or equal to krthr_value then it
is set to zero. If *SWTKRTHR is not specified, then a default value of 5.0E07 is used for krthr_value. Entered values of krthr_value are reset to lie
within the range 5.0E-16 to 5.0E-07.
Regression Options:
*SMOOTH
Perform regression on the relative permeability table entries and generate an
equally spaced table which best fits the data over the specified range.
Valid Regression Optionsubkeywords of the *SMOOTH Keyword are:
*ALL *ALLPL *ALLINT *ALLLIN *NONE *OFF Property_list
If the *ALL subkeyword is used, or no subkeywords are Found, then all
properties in the table are smoothed according to default actions. This tries
power law first. For kr's, if the resultant power is greater than 4.0 then linear
is used for that property. The smoothed ranges for *ALL are as follows:
Property
Range
Krw
Krow
Pcow
Swcrit to 1.0-Sorw
Swcon to 1.0-Sorw
Swcon to Sw(Pcow=0.0)
Pcowi
Sorm
Swcon to Sw(Pcow=0.0)
Swcon to Max Sw in Table
where
Swcrit
Swcon
Sw(Pcow=0.0)
=
=
=
Critical Saturation
Connate Saturation
Lowest Sw in table with Pcow constant
*ALLPL
Specifies that Power law is used for all properties in the table. If powers >
4.0 are calculated, they are used.
*ALLINT
Specifies that Power law is used between a reduced range of end points the
smoothed ranges for *ALLINT are as follows
Property
Range
Krw
Krow
Pcow
Swcrit to 1.0-Sorw
Swcrit to 1.0-Sorw
Swcrit to Sw(Pcow=0.0)
Pcowi
Sorm
Swcrit to Sw(Pcow=0.0)
Swcrit to Max Sw in Table
526 Rock-Fluid Data
User's Guide IMEX
where
Swcrit = Critical Saturation
Sw(Pcow=0.0) = Lowest Sw in table with Pcow constant
*ALLLIN
Generates an evenly spaced table using linear interpolation for all properties.
If the *NONE or *OFF keyword is used then none of the table properties are
smoothed. This is the default action if the *SMOOTH keyword is not
specified.
property_list allows the user to customize how regression is applied to a list
of properties.
For a *SWT table, valid properties to be regressed include:
*KRW
*KRWPL
*KRWINT
*KRWLIN
*KROW
*KROWPL
*KROWINT
*KROWLIN
*PCOW
*PCOWPL
*PCOWINT
*PCOWLIN
*PCOWI
*PCOWIPL
*PCOWIINT
*PCOWILIN
*SORM
*SORMPL
*SORMINT
*SORMLIN
If property_list is used and a particular, existing property is not specified,
then linear interpolation is used for that property.
The conventions for the above properties are the same as for *ALL, *ALLPL
and *ALLINT and *ALLLIN.
Table Regeneration Option:
*DSWTI
When the Regression options (*SMOOTH) are not used IMEX does not
regenerate equally spaced rel. perm. tables, i.e.:
Equally spaced tables not generated when,
1. the *SMOOTH keyword is not found,
2. the *SMOOTH *NONE keyword is found,
3. the *SMOOTH *OFF keyword is found.
When regression is employed, IMEX automatically creates equally spaced
tables for the newly created curves.
The *DSTWI keyword can be used to define the number of points in the table.
NOTE:
*SMOOTH *ALLLIN and *DSWTI dswti_value
Can be used together to generate new equally spaced tables without using
regression on table values. This can speed up relative permeability
calculation times as equally spaced tables are interpolated more efficiently.
User's Guide IMEX
Rock-Fluid Data 527
dswti_value
Generate an evenly space table with deltaSw approximately equal to
dswti_value. Calculate deltaSw as follows:
deltaSw=(1-Swc-Sorw)/int((1-Swc-Sorw)/dswti+eps)
using the input dswti_value and where int(x) is the truncated integer value of x.
If the *DSWTI keyword is not specified, the default for dswti_value is
calculated internally.
The minimum of half of the minimum table spacing and 0.016 is calculated:
1. If this value is greater than 0.001 then dswti_value is set to it.
2. If this value is less than 0.001 then dswti_value is set to 0.001.
Direct Input of Power Law Coefficients:
*PLAW
Input power law exponents for table properties for cases with only end points
input.
pkrw
Power law exponent for krw for cases with only end points input. A value of
0.0 results in a default value of 1.0.
pkrow
Power law exponent for krow for cases with only end points input. A value
of 0.0 results in a default value of 1.0.
ppcowd
Power law exponent for Pcowd for cases with only end points input. A value
of 0.0 results in a default value of 1.0. If it is not specified, then it is
assumed that Pcowd is zero.
ppcowi
Power law exponent for Pcowi for cases with only end points input. A value
assumed that Pcowi is zero.
psorm
Power law exponent for Sorm for cases with only end points input. A value
assumed that Sorm is zero.
528 Rock-Fluid Data
User's Guide IMEX
EXPLANATION:
In the regression option, power law or linear interpolation may be used for krw and krow and,
if they exist, for Pcow and Pcowi. If *SMOOTH keyword is omitted, or *SMOOTH *NONE
or *OFF is specified, no smoothing is done. Otherwise, the keywords
*SMOOTH (*ALL or *ALLPL or *ALLINT or *ALLLIN or
*NONE or *OFF property_list))
determine for which properties power law and for which properties linear interpolation are to
be used.
If *PLAW is specified, then all properties are assumed to be power laws using the specified
exponents. End points and maximum values are determined from the following table. All
other values, if they exist are ignored.
If *PLAW is not specified and power law correlations are specified, then there must be at two
Sw's given such that the particular property is nonzero. The power for that property is then
calculated using nonlinear least squares.
If the *SMOOTH keyword is specified, an equally spaced table is generated. Using a power
law or linear interpolation as specified by its subkeywords. Within the simulator, this equally
spaced table is used. Relative permeabilities are calculated using linear interpolation from
this internal table except over the interval between kr=0 and kr>0. On this interval, the type
of smoothing, if any, is controlled by the *SMOOTHEND keyword and optionally one of its
subkeywords.
Examples:
a) Generate power law correlations for Krw, Krow and Pcow using least squares fits
of the following table:
*SWT *SMOOTH
** sw
0.178350
:
0.881490
krw
0.0
:
0.490000
krow
0.991590
:
0.000000
pcow
27.930
:
-2.75
b) Generate power law correlations for Krw, Krow and Pcow (drainage) matching
endpoints and maximum values for a table where Swcrit = Swcon and Sorw =
Soirw:
1. Input the power law exponents for Krw, Krow and Pcowd on the *SWT
keyword.
*SWT * SMOOTH * PLAW pkrw pkrow ppcowd
2. Input the endpoints and values of Krwiro, Krowcw and minimum and
maximum values of drainage oil water capillary pressure in the two row
table.
Swcrit = Swcon
1 Sorw
User's Guide IMEX
0.0
Krwiro
Krowcw
0.0
Pcowmax
Pcowmin
Rock-Fluid Data 529
c) Generate power law correlations for Krw, Krow and Pcow (drainage) matching
endpoints and maximum values for a table where Swcrit > Swcon and Sorw =
Soirw:
1. Input the power law exponents as in B above.
2. Input the endpoints and values of Krwiro, Krwcw and minimum and
maximum values of oil water capillary pressure in the three row table.
Swcon
Swcrit
1 Sorw
0.0
0.0
Krwiro
Krowcw
*INT
0.0
Pcowmax
*INT
Pcowmin
Note: Any value may be used instead of *INT, as Krow and Pcow at Sw = Swcrit are
ignored When using *PLAW.
For the above (B and C), assuming *SMOOTH *ALL (the default) or *SMOOTH *ALLPL,
we obtain:
Sw Swcrit
Krw = Krwiro
1 Swcrit Sorw
Pkrw
1 Sorw Sw
Krow = Krocw
1 Swcon Sorw
Pcow D
Pkrow
1 Sorw Sw
= (Pcow (Sw = Swcon ) Pcow (Sw = 1 Sorw )

1 Swcon Sorw
+ Pcow (Sw = 1 Sorw )
530 Rock-Fluid Data
Ppcowd
User's Guide IMEX
Liquid-Gas Relative Permeability Table (Conditional)
*SLT,
*SGT
PURPOSE:
*SLT indicates the start of a liquid-gas relative permeability table dependent on liquid saturation.
*SGT indicates the start of a liquid-gas relative permeability table dependent on gas saturation.
In addition *SLT(*SGT) controls the use of a wide variety of relative permeability regression
and optimization functions.
In its most basic form (to define beginning of table).
TABLE:
*SLT (*NOSWC) (*SMOOTHEND Subkeywords) (*SLTKRTHR krthr_value)

Sl
:
krg
:
krog
:
(Pcog) (Pcogi)
:
*SGT (*NOSWC) (*SMOOTHEND Subkeywords) (*SLTKRTHR krthr_value)

Sg
:
krg
:
krog
:
(Pcog) (Pcogi)
:
*SLT(*SGT) subkeywords provide the user with a number of unique features which can:
1. Change the shape of the curves as they go to zero.
2. Create new curves obtained from nonlinear regression of available data.
3. Allow the user to interact with the regression procedure to specify the range over
which regression is performed.
4. Regenerate equally spaced tables using a user defined Interpolation interval.
See below for its most complete form:
DEFINITIONS:
*NOSWC
Keyword indicating that the total liquid residual saturation does not include
the connate water saturation.
If the *NOSWC option is used, it is assumed that the connate water
saturation is not part of the measured residual liquid saturation in the liquidgas table. Hence, Sorg is set equal to the residual liquid saturation obtained
from the *SLT (*SGT) table and Slcon equals irreducible oil (Soirg).
By default, it is assumed that the residual liquid saturation in the liquid-gas
table was measured in the presence of connate water. Hence, Sorg is
obtained by subtracting off Swcon from the residual liquid saturation
obtained from the *SLT (*SGT) table. Slcon is comprised of connate water
(Swcon) plus irreducible oil (Soirg). See the figures in the *SWCON
definition for more details.
User's Guide IMEX
Rock-Fluid Data 531
*SMOOTHEND
interval with kr=0 and kr>0 for krg and krog. These are the table intervals
where gas or oil has just become mobile. The keyword may be followed by a
subkeyword as indicated below.
*OFF *ON *QUAD *CUBIC

If *SMOOTHEND is not specified or it has no subkeyword the default is
*SMOOTHEND *ON. This is equivalent to *SMOOTHEND *QUAD.
Critical End-point Threshold Option:
*SLTKRTHR
Optional keyword to use krthr_value as a threshold value for krg and krog for
end-point determination.
krthr_value
If a tabular value for krg or krog is less than or equal to krthr_value, then it is
set to zero. If *SLTKRTHR is not specified, then a default value of 5.0E-07 is
used. Entered values of krthr_value are reset to lie within the range 5.0E-16 to
5.0E-07.
Sl
Total liquid saturation (fraction).
Sg
Gas saturation (fraction).
krg
Relative permeability to gas (fraction) for the given saturation.
krog
Relative permeability to oil (fraction) in the presence of gas and connate
water for the given saturation.
Pcog
Gas-oil capillary pressure (kPa | psi | kPa | kg/cm2). If this column is not
entered it is assumed to be zero and capillary pressure effects are not
included in the simulation.
532 Rock-Fluid Data
User's Guide IMEX
Pcogi
Imbibition Gas-oil capillary pressure (kPa | psi | kPa | kg/cm2). This column
is optional; Pcogi (column 5) must be entered alongside Pcog (column 4). If
Pcogi is not entered, it is set equal to Pcog and gas-oil capillary pressure
hysteresis is not modeled. Pcog hysteresis cannot be modelled when using
the pseudo-miscible option.
DEFAULTS:
Conditional keyword. There are no defaults.

CONDITIONS:
This keyword must be in the Rock-Fluid Data keyword group, immediately after the *SWT
keyword and table. Either *SLT or *SGT are required with *RPT.
EXPLANATION:
Capillary pressure curves determine the height of the transition zones calculated by the
*VERTICAL option. Nonzero capillary pressure curves also influence the width of the
transition zones near sharp saturation fronts. This front widening mechanism tends to have a
stabilizing effect during the displacement of less mobile fluids (oil) by more mobile fluids (gas).
The gas-oil capillary pressure is defined as
Pcog = Pg - Po
Since liquid is assumed to be the wetting phase, the capillary pressure should be
monotonically declining with increasing liquid saturation. However, the Pcg value at Sl=1 or
Sg=0 can be nonzero. If Pcog is not entered, the capillary pressure defaults to zero.
For *SLT, table entries must be in order of increasing liquid saturation. The last entered
value for Sl should be 1. The last entered krog value (krog at Sl=1) should equal the first
entered value of Krow in the water-oil relative permeability table (krow at connate water
saturation (Swcon)).
If the *NOSWC option is used, it is assumed that the connate water saturation is not part of
the measured residual liquid saturation in the liquid-gas table. Thus Slcon and Sorg do not
depend on the connate water saturation derived from the oil-water curve or entered using the
*SWCON keyword
When *NOSWC is not used, reading in *SWCON arrays will effect both oil-water and
liquid-gas tables.
When *NOSWC is used, reading in *SWCON will effect only the oil-water tables. The gasliquid table is assumed to have no connate water in place, thus the connate liquid (Slcon) is
equal to the irreducible oil (Soirg).
See the *SWCON card for further explanation.
Examples:
See examples in the *SWT keyword section. See Figure 8 in Appendix D for typical relative
permeability curves.
User's Guide IMEX
Rock-Fluid Data 533
THE *OILWET OPTION:
When the oil wet option is used with the BLOCK_CENTER initialization option, the gas
capillary pressure entered in the Pcog column is the capillary pressure between the
intermediate wetting phase (water) and the nonwetting phase (gas). Thus gas phase pressure
is calculated as:
Pgas = Poil + Pcow + Pcog
where:
Pgas
Poil
Pcow
Pcog
= gas phase pressure

= oil phase pressure
= capillary pressure from the water oil rel-perm table.
= capillary pressure from the gas liquid rel-perm table.
Compare this with the water wet case or the oil wet case when the DEPTH_AVE
initialization option is used where:
Pgas = Poil + Pcog
Please refer to the tutorial section for a further discussion of the oil wet option used in IMEX:
THE *GASWATER OPTION:
When using the Gas Water option the SGT/SLT table is only altered slightly. Krog is not
read in and the Pcog column is not read. Thus a typical Gas Water SGT/SLT table would be:
*SLT (*NOSWC)
Sl
krg
:
:
*SGT (*NOSWC)
Sg
krg
:
:
When using the Gas Water option, it is important to ensure that the value of Soirg (irreducible
oil in the liquid-gas table) calculated from this table is always 0.0 and that the value of Slcon
is equal to Swcon (or is equal to 0.0, if the *NOSWC option is used).
In a *SGT table this is accomplished by ensuring that the final value of Sg in the table is 1.0Swcon (or 1.0, if the *NOSWC option is used). In a *SLT table this is accomplished by
ensuring that the first value of Sl is equal to Swcon (or 0.0, if the *NOSWC option is used).
THE *GASWATER_WITH_CONDENSATE OPTION:
When using the Condensate option the Pcog column is not read. Pcgl (capillary pressure
between gas and liquid) is read in on the SWT table in place of Pcow (See *SWT).
The acceptable range of values for liquid or gas saturation is:
min
max
534 Rock-Fluid Data
SI
Field
Lab
0.0
1.0
0.0
1.0
0.0
1.0
User's Guide IMEX
The acceptable range of values for gas relative permeability is:
min
max
SI
Field
Lab
0.0
1.0
0.0
1.0
0.0
1.0
The acceptable range of values for oil relative permeability is:
min
max
SI
Field
Lab
0.0
1.0
0.0
1.0
0.0
1.0
The acceptable range of values for capillary pressure is:
min
max
SI
kPa
-2.5E+4
2.5E+4
Field
psi
-3625.0
3625.0
Lab
kPa
-2.5E+4
2.5E+4
Mod. SI
kg/cm2
-2.5E+2
2.5E+2
General Form of the *SLT(*SGT) Keyword:

TABLE:
*SLT (*NOSWC) (*SMOOTHEND (*QUAD or *CUBIC or

(*SMOOTH (*ALL or *ALLPL or *ALLINT or *ALLLIN
(*PLAW pkrg pkrog ppcog ppcogi)
(*DSLTI dslti_value) (*SLTKRTHR krthr_value)
Sl
:
krg
:
krog
:
(Pcog)
:
(Pcogi)
:
*SGT (*NOSWC) (*SMOOTHEND (*QUAD or *CUBIC or

(*SMOOTH (*ALL or *ALLPL or *ALLINT or *ALLLIN
(*PLAW pkrg pkrog ppcog ppcogi)
(*DSLTI dslti_value) (*SLTKRTHR krthr_value)
Sg
:
krg
:
krog
:
(Pcog)
:
(Pcogi)
:
Uses of the General Form:
Definitions of individual subkeywords will follow, but it is useful to first highlight a few of
the many tasks the *SLT(*SGT) keyword can perform in situations where rock data is sparse
and the user wants to generate meaningful curves.
User's Guide IMEX
Rock-Fluid Data 535
power-law model over the entire range of data.
*SLT *SMOOTH *ALL or SLT *SMOOTH *ALLPL
power-law model over a smaller range of data which ensures matching of the
original data at critical points and when Pcog -> 0.0 (or becomes constant).
*SLT *SMOOTH *ALLINT
3. Using no data, specify power law curves explicitly using input power law
exponents and a two row relative perm. table (to define saturation end points and
relative permeability end points)
*SLT *SMOOTH *PLAWgas_exponentoil_exponentpcog_exponent
4. Re-interpolate an existing table using equal spacing of approx. 0.01 to take
advantage of faster table look-up procedures.
*SLT *SMOOTH *ALLLIN *DSLTI 0.01
5. Alter the way the interpolation is performed in the table interval between where
kr=0 and the first nonzero entry in the table. For Relative permeability, this is at
Sgcrit (for Krg) and at Sorg (for Krog).
*SLT *SMOOTHEND *CUBIC
DEFINITIONS:
(General Form which includes all Regression Options)

The *SLT (*SGT) keyword can be subdivided into 5 subkeywords
*SMOOTHEND - Smooth End-points of tables
*SLTKRTHR - Gas Table Critical End-point Threshold Option
*SMOOTH - Regression Options
*DSLTI - Table Regeneration Options
*PLAW - Direct Power Law Input (Regression not Used)
*SLT (*SGT)
(*SMOOTHEND subkeywords)
(*SLTKRTHR krthr_value)
(*SMOOTHsubkeywords)
(*DSLTI dslti_value)
(*PLAW subkeywords)
*SMOOTHEND
interval with kr=0 and kr>0 for krg and krog. These are the table intervals
where gas or oil has just become mobile. The keyword may be followed by a
subkeyword as indicated below.
536 Rock-Fluid Data
User's Guide IMEX
*OFF *ON *QUAD *CUBIC *POWERQ *POWERC

If *SMOOTHEND is not specified or it has no subkeyword the default is
*SMOOTHEND *ON. This is equivalent to *SMOOTHEND *POWERQ.
For *POWERQ and *POWERC end point smoothing is done using power
law if the exponent is between 1.5 and 4.0. Otherwise, for *POWERQ,
quadratic is used and for *POWERC, cubic is used for end point smoothing.
If the *SMOOTH regression option is not used, *POWERC And *POWERQ
default to *CUBIC and *QUAD respectively.
Critical End-point Threshold Option:
*SLTKRTHR
Optional keyword to use krthr_value as a threshold value for krg and krog for
end-point determination.
krthr_value
If a tabular value for krg or krog is less than or equal to krthr_value, then it is
set to zero. If *SLTKRTHR is not specified, then a default value of 5.0E-07
is used. Entered values of krthr_value are reset to lie within the range 5.0E16 to 5.0E-07.
Regression Options:
*SMOOTH
Perform regression on the relative permeability table entries and generate an
equally spaced table which best fits the data over the specified range.
Valid Regression Option Subkeywords of the *SMOOTH Keyword:
*ALL *ALLPL *ALLINT *ALLLIN *NONE *OFF Property_list

If the *ALL subkeyword is used, or no subkeywords are found, then all
properties in the table are smoothed according to default actions. This tries
power law first. For kr's, if the resultant power is greater than 4.0 then linear
is used for that property. The smoothed ranges for *ALL are as follows:
Property
Krg
Krog
Pcog
Pcogi
User's Guide IMEX
Range
Sllow to 1.0-Sgcon
Slcrit to 1.0-Sgcon
Sllow to Sl(Pcog=0.0)
Sllow to Sl(Pcogi=0.0)
Rock-Fluid Data 537
where
Slcrit
Sllow
Sl(Pcog=0.0)
Sgcon
=
=
=
=
Critical Liquid Saturation

Lowest Sl in Table
Lowest Sl in table with Pcog constant
Connate Gas Saturation which is assumed equal
to Critical Gas Saturation
*ALLPL
Specifies that Power law is used for all properties in the table. If a power >
4.0 is calculated, the power is used.
*ALLINT
Specifies that Power law is used between a reduced range of end points the
smoothed ranges for *ALLINT are as follows
Property
Krg
Krog
Pcog
Pcogi
Range
Slcrit to 1.0-Sgcon
Slcrit to Slmax
Sllow to Sl(Pcog=0.0)
Sllow to Sl(Pcogi=0.0)
where
Slcrit
Sllow
Sl(Pcog=0.0)
Sgcon
=
=
=
=
Critical Liquid Saturation

Lowest Sl in Table
Lowest Sl in table with Pcog constant
Connate Gas Saturation which is assumed
equal to Critical Gas Saturation
*ALLLIN
Generates an evenly spaced table using linear interpolation for all properties.
If the *NONE or *OFF keyword is used then none of the table properties are
smoothed. This is the default action if the *SMOOTH keyword is not
specified.
property_list allows the user to customize how regression is applied to a list
of properties.
For a *SLT(*SGT) table, valid properties to be smoothed include:
*KRG
*KRGPL
*KRGINT
*KRGLIN
*KROG
*KROGPL
*KROGINT
*KROGLIN
*PCOG
*PCOGPL
*PCOGINT
*PCOGLIN
*PCOGI
*PCOGIPL
*PCOGIINT
*PCOGILIN
If property_list is used and a particular, existing property is not specified,
then linear interpolation is used for that property.
The conventions for the above properties are the same as for *ALL, *ALLPL
and *ALLINT and *ALLLIN.
538 Rock-Fluid Data
User's Guide IMEX
Table Regeneration Option:
*DSLTI
When the Regression options (*SMOOTH) are not used IMEX does not
regenerate equally spaced rel. perm. tables, i.e.:
Equally spaced tables not generated when,
1. the *SMOOTH keyword is not found,
2. the *SMOOTH *NONE keyword is found,
3. the *SMOOTH *OFF keyword is found.
When regression is employed IMEX automatically creates equally spaced
tables for the newly created curves.
The *DSTLI keyword can be used to define the number of points in the table.
NOTE:
*SMOOTH *ALLLIN and *DSLTI dslti_value

Can be used together to generate new equally Spaced tables without using
regression on table values. This can speed up relative permeability
calculation times as equally spaced tables are interpolated more efficiently.
dslti_value
Generate an evenly space table with deltaSl approximately equal to
dslti_value. Calculate deltaS as follows:
deltaSl=(1-Slc-Sgc)/int((1-Slc-Sgc)/dslti+eps)
Using the input dslti_value and where int(x) is the truncated integer value of x.
If the *DSLTI keyword is not specified, the default for dslti_value is
calculated internally. The minimum of half of the minimum table spacing
and 0.016 is calculated. If this value is greater than 0.001 then dslti_value is
set to it. Otherwise, if *DSLTI is not specified, dslti_value is set to 0.001.
Direct Input of Power Law Coefficients:
*PLAW
Input power law exponents for table properties for cases with only end points
input.
pkrg
Power law exponent for krg for cases with only end points input. A value of
0.0 results in a default value of 1.0.
pkrog
Power law exponent for krog for cases with only end points input. A value
of 0.0 results in a default value of 1.0.
User's Guide IMEX
Rock-Fluid Data 539
ppcog
Power law exponent for Pcog for cases with only end points input. A value
assumed that Pcog is zero.
ppcogi
Power law exponent for Pcogi for cases with only end points input. A value
assumed that Pcogi is zero.
EXPLANATION:
For optional smoothing, power law or linear interpolation may be used for krg and krog and,
if it exists, for Pcog. If *SMOOTH keyword is omitted, or *SMOOTH *NONE or *OFF is
specified, no smoothing is done. Otherwise, the keywords
*SMOOTH
(*ALL or *ALLPL or *ALLINT or *ALLLIN or

*NONE or *OFF or property_list))
determine for which properties power law and for which properties linear interpolation are to
be used.
If *PLAW is specified, then all properties are assumed to be power laws using the specified
exponents. End points and maximum values are determined from the following table. All other
values, if they exist are ignored. If *PLAW is not specified and power law correlations are
specified, then there must be at least two Sl's or Sg's given such that the particular property is
nonzero. The power for that property is then calculated using nonlinear least squares.
If the *SMOOTH keyword is specified, an equally spaced table is generated. Using a power
law or linear interpolation as specified by its subkeywords. Within the simulator, this equally
spaced table is used. Relative permeabilities are calculated using linear interpolation from
this internal table except over the interval between kr=0 and kr>0. On this interval, the type
of smoothing, if any, is controlled by the *SMOOTHEND keyword and optionally one of its
subkeywords.
Examples:
A. Generate power law correlations for Krg, Krog and Pcog (drainage) matching
endpoints and maximum values for a gas liquid (*SLT) table where Slcrit = Slcon
and Sgcrit = Sgcon:
1. Input the power law exponents for Krg, Krog and Pcogd on the *SLT
keyword.
*SLT * SMOOTH * PLAW pkrg pkrog Ppcog
2. Input the endpoints and values of Krgcl and Krogcg and minimum and
maximum values of drainage oil gas capillary pressure in the two row
table.
Slcrit = Slcon
1 Sgcrit
540 Rock-Fluid Data
Krgcl
0.0
0.0
Krogcg
Pcogmax
Pcogmin
User's Guide IMEX
B. Generate power law correlations for Krg, Krog and Pcog (drainage) matching
endpoints and maximum values for a gas liquid (*SLT) table where Slcrit > Slcon
and Sgcrit = Sgcon:
1. Input the power law exponents for Krg, Krog and Pcogd as in A above.
2. Input the endpoints and values of Krgcl and Krogcg and minimum and
maximum values of drainage oil gas capillary pressure in the three row
table.
Slcon
Slcrit
1 Sgcrit
Krgcl
*INT
0.0
0.0
0.0
Krogcg
Pcogmax
*INT
Pcogmin
Note: Any value may be used in place of *INT, as Krg and Pcog at Sl = Slcrit are ignored
when using *PLAW.
For the above (A and B), assuming *SMOOTH *ALL (the default) or *SMOOTH *ALLPL,
we obtain
Sl Slcrit
Krog = Krogcl
1 Slcrit Sgcrit
1 Sgcrit Sl
Krg = Krgcl
1 Slcon Sgcrit
Pcog D
Pkrog
Pkrg
1 Sgcrit Sl
= (Pcog (Sl = Slcon ) Pcog (Sw = 1 Sgcrit )

1 Slcon Sgcrit
+ Pcog (Sw = 1 Sgcrit )
User's Guide IMEX
Ppcog
Rock-Fluid Data 541
Hysteresis Parameters (Optional)
*EPSPC, *EPSPCG, *HYSKRG,
*HYSKRO
PURPOSE:
These keywords signal entry of the hysteresis parameters.

*EPSPC
Denotes the entry of epspc. This determines the transition between the
imbibition and drainage curves for oil-water capillary pressure.
*EPSPCG
Denotes the entry of epspcg. This determines the transition between the
imbibition and drainage curves for oil-gas capillary pressure.
*HYSKRG
Denotes that hysteresis effect on the gas relative permeability is modelled.
*HYSKRO
Denotes that hysteresis effect on the oil relative permeability (krow) is modelled.
FORMAT:
*EPSPC
*EPSPCG
*HYSKRG
*HYSKRO
epspc
epspcg
sgrmax
sormax
DEFINITIONS:
epspc
Dimensionless real number which determines the transition between the
imbibition and drainage curves for oil-water capillary pressure. Typical
values of epspc lie between 0.05 and 0.1. epspc cannot be less than 1.0d-05.
epspcg
Dimensionless real number which determines the transition between the
imbibition and drainage curves for oil-gas capillary pressure. Typical values
of epspcg lie between 0.05 and 0.1. epspcg cannot be less than 1.0d-05.
sgrmax
Sgrmax is the maximum possible imbibition residual gas saturation. It is the
endpoint of an imbibition branch which breaks off from the gas relative
permeability drainage curve at the maximum possible krg (at Sg = 1.0Slcon). This value is used to evaluate the shape and path of all imbibition
curves which leave the drainage curve at any saturation. Sgrmax must be
greater than Sgcrit and less than 1.0-Slcon. Even though Sgcrit is normally
close to zero, Sgrmax can be quite large, values of 0.3 to 0.4 are typical.
542 Rock-Fluid Data
User's Guide IMEX
sormax
Sormax is the maximum possible imbibition residual oil saturation (Oil-water
table). It is the endpoint of an imbibition branch which breaks off from the oil
relative permeability drainage curve at the maximum possible kro (at So = 1.0Swcon). This value is used to evaluate the shape and path of all imbibition
curves which leave the drainage curve at any saturation. Sormax must be
greater than Sorw and less than 1.0-Swcon. Sormax is normally significantly
larger than Sorg.
DEFAULTS:
If *HYSKRG is not entered the hysteresis effect on gas relative permeability is not included
in the simulation. If *HYSKRG is entered without the real number sgrmax following it,
sgrmax = Sgcrit + 0.5 * (1-Slcon-Sgcrit),
This default value corresponds to the midpoint of the allowable range.
If *HYSKRO is not entered the hysteresis effect on oil relative permeability (krow) is not
included in the simulation. If *HYSKRO is entered without the real number sormax
following it,
sormax = Sorw + 0.5 * (1-Swcon-Sorw),
This default value corresponds to the midpoint of the allowable range.
The water-oil capillary pressure hysteresis effect is included if the fifth column (Pcowi) of the
tables introduced by *SWT is specified. The default for epspc is 0.1. If *EPSPC is entered
without a real number following it, epspc = 0.1 is assumed.
The oil-gas capillary pressure hysteresis effect is included if the fifth column (Pcogi) of the
tables introduced by *SLT (or *SGT) is specified. The default for epspcg is 0.1. If
*EPSPCG is entered without a real number following it, epspcg = 0.1 is assumed.
CONDITIONS:
These keywords must be in the Rock-Fluid Data keyword group, after the *RPT keyword.
EXPLANATION:
The epspc, epspcg, sgrmax and sormax refers to the rock type set_number specified by *RPT
(which must be defined before these keywords are used). If more than one rock type is
present then *EPSPC, *EPSPCG, *HYSKRG and *HYSKRO must be specified before the
next *RPT keyword. After the next *RPT keyword, a new set of *EPSPC, *EPSPCG,
*HYSKRG and *HYSKRO can be specified for the new rock type set_number.
When Pcow hysteresis is included, the water-oil capillary pressure is expressed as follows. If
the capillary pressure is initially on the drainage curve (decreasing Sw) and Sw begins to
increase, then pcow is calculated from the drainage-to- imbibition scanning curve
pcow(Sw) = (f) x pcow(Im,Sw) + (1-f) x pcow(Dr,Sw);
here f lies between zero and one, and is given by
User's Guide IMEX
Rock-Fluid Data 543
Sw (max) Sw (hyst) + epspc
f =
Sw (max ) Sw (hyst )
Sw Sw (hyst )
Sw Sw (hyst ) + epspc
where Sw(hyst) is the value which Sw had when the reversal occurred and Sw(max) is the
maximum attainable water saturation 1-Soirw. pcow remains on this scanning curve until Sw
falls below Sw(hyst), where it reverts to the drainage curve. If the capillary pressure is
initially on the imbibition curve (increasing Sw) and Sw begins to decrease, then pcow is
calculated from the imbibition-to-drainage scanning curve:
pcow(Sw) = (f) x pcow(Dr,Sw) + (1-f) x pcow(Im,Sw),
where
Sw (hyst) Swr + epspc
f =
Sw (hyst ) Swr
Sw (hyst ) Sw
Sw (hyst ) Sw + epspc
with Swr the connate water saturation. Pcow remains on this scanning curve until Sw exceeds
Sw(hyst), where pcow reverts to the imbibition curve. Let dSmax be the maximum amount by
which water saturation changes in one continuous drainage or imbibition cycle. If epspc is much
larger than dSmax, the transition between curves is gradual, being nearly linear in Sw; if epspc is
much smaller than dSmax the transition is sudden, with the full transition to the second curve
occurring when Sw differs from Sw(hyst) by only a small amount. The typical values of 0.05 to
0.1 for epspc generally leads to a moderate transition. epspc cannot be less than 1.0d-05.
When Pcog hysteresis is included, the oil-gas capillary pressure is expressed as follows. Sl
below refers to liquid saturation in the oil (liquid) gas table. If the capillary pressure is
initially on the drainage curve (decreasing Sl) and Sl begins to increase, then pcog is
calculated from the drainage-to- imbibition scanning curve
pcog(Sl) = (f) x pcog(Im,Sl) + (1-f) x pcog(Dr,Sl);
here f lies between zero and one, and is given by
Sl(max) Sl(hyst) + epspcg
f =
Sl(max ) Sl(hyst )
Sl Sl(hyst )
Sl Sl(hyst ) + epspcg
where Sl(hyst) is the value which Sl had when the reversal occurred and Sl(max) is the
maximum attainable liquid saturation 1-Sgcon. Pcog remains on this scanning curve until Sl
falls below Sl(hyst), where it reverts to the drainage curve. If the capillary pressure is
initially on the imbibition curve (increasing Sl) and Sl begins to decrease, then pcog is
calculated from the imbibition-to-drainage scanning curve:
pcog(Sl) = (f) x pcog(Dr,Sl) + (1-f) x pcog(Im,Sl),
where
Sl(hyst) Slr + epspcg
f =
Sl(hyst ) Slr
544 Rock-Fluid Data
Sl(hyst ) Sl
(
)
Sl
hyst
Sl
+
epspcg
User's Guide IMEX
with Slr the connate liquid saturation. Pcog remains on this scanning curve until Sl exceeds
Sl(hyst), where pcog reverts to the imbibition curve. Let dSmax be the maximum amount by
which water saturation changes in one continuous drainage or imbibition cycle. If epspcg is
much larger than dSmax, the transition between curves is gradual, being nearly linear in Sl; if
epspcg is much smaller than dSmax the transition is sudden, with the full transition to the
second curve occurring when Sl differs from Sl(hyst) by only a small amount. The typical
values of 0.05 to 0.1 for epspcg generally leads to a moderate transition. epspcg cannot be
less than 1.0d-05.
The hysteresis effect on krg is calculated as follows:
krg(Sg) = krg(Dr;Sg),during drainage;
krg(Sg) = krg(Dr;Sg(shifted)), during imbibition;
where
Sg(shifted)=Sgcrit+(Sg-Sgrh)(Sgh-Sgcrit)/(Sgh-Sgrh).
krg is assumed to lie on the drainage curve initially; Sgh is the value of Sg when the shift to
imbibition occurs and Sgrh is the value of Sgcrit corresponding to Sgh via Land's equation,
1
1
1
1
S g r m a x Sgcrit
S g m a x Sgcrit
Sgrh Sgcrit
Sgh Sgcrit
where Sgmax = 1 - Swcon. Sgrmax has the value of the user-entered parameter sgrmax.
The hysteresis effect on krow is calculated as follows:
krow(Sw) = kro(Dr;Sw),during drainage;
krow(Sw) = kro(Dr;Sw(shifted)), during imbibition;
where
Sw(shifted)=1.0 Sorw (1-Sw-Sorh)(Soh-Sorw)/(Soh-Sorh).
krow is assumed to lie on the drainage curve initially; Soh is the value of So when the shift to
imbibition occurs and Sorh is the value of Sorw corresponding to Soh via Land's equation,
1
1
1
1
S o r m a x Sorw
So m a x Sorw
Sorh Sorw
Soh Sorw
where Somax = 1 - Swcon. Sormax has the value of the user-entered parameter sormax.
Example:
*EPSPC 0.05 *EPSPCG 0.05
*HYSKRG 0.4
*HYSKRO 0.4
THE *OILWET OPTION:
When the oil wet option is used, hysteresis effects can still be modelled, however the non
wetting phase relative permeability hysteresis (controlled by HYSKRO) now refers to water
relative permeability and somax refers to water saturation. The water capillary pressure
hysteresis now refers to non wetting phase (water) pressure wetting phase (oil) pressure.
Please refer to the Tutorial section for details.
User's Guide IMEX
Rock-Fluid Data 545
TRAPPED OIL HYSTERESIS OPTION:
If the rock type uses the trapped oil hysteresis option, Pcow hysteresis and Krow hysteresis
are coupled and hysteresis is handled using an approach that considers the trapped oil as a
immobile oil saturation in addition to Sorw for the imbibition process.
To enable trapped oil hysteresis, it is only necessary to give the imbibition Pcow in SWT
table with constant Pcow for water saturation points equal to and great than 1-Sormax. The
trapped oil hysteresis will be applied for grid blocks using that table.
When using the trapped oil hysteresis option, Krow hysteresis is the same as described above;
however, *HYSKRO is no longer necessary and is ignored if given to read in Sormax. The
Sormax value is instead read from the imbibition Pcow column of the *SWT table.
Keyword *EPSPC is still applied to calculate Pcow scanning curves.
See the Tutorial section for a complete description.
546 Rock-Fluid Data
User's Guide IMEX
Reduced Vertical Extent of Pcow Cycle for Trapped Oil

Hysteresis (Conditional)
*DAMP-PCOW-TROIL
PURPOSE:
Allows linear reduction of Pcow values when Soh towards Sorw so as to make a damped
Pcow hysteresis cycle.
FORMAT:
*DAMP-PCOW-TROIL
DEFAULTS:
Conditional keyword. By default the option is off.

CONDITIONS:
Used only when the rock type is trapped oil hysteresis enabled.
DEFINITION:
In the trapped oil hysteresis option for Pcow, all Pcow scanning cycles start from the historically
attained maximum oil saturation, Soh, and end at the residual oil saturation of the imbibition
process, Sorh.
If Soh is very close to Sorw (And therefore Sorh), and the difference between the imbibition
capillary pressures at Sorh and the drainage capillary pressure at Soh is large, a small change
in water saturation may result in very large change in capillary pressure.
This keyword provides an option to reduce the capillary pressure difference which might
result in a pressure oscillation. The reduction factor linearly varies from 1 to 0 as the
historical attained oil, Soh, changes from 1-Swcon to Sorw. Please see the tutorial The Trapped
Oil Hysteresis Option for Pcow and Krow (Oil-Water System) for discussions and graphs.
User's Guide IMEX
Rock-Fluid Data 547
Method for Evaluating 3-Phase Kro (Optional)
*KROIL
PURPOSE:
*KROIL introduces subkeywords which indicate how Kro is evaluated. The options are:
Stone's first model; Stone's second model as modified by Aziz and Settari; and the linear
isoperm method.
For Stone's models the Krow and Krog values may be looked up as functions either of Sw
and Sg, respectively, or of 1-So and 1-Swcon-So.
FORMAT:
*KROIL
(*STONE1)
(*STONE2)
(*LINEAR_ISOPERM)
(*SEGREGATED)
(*SWSG | *SO)
(*SWSG | *SO)
DEFINITIONS:
*STONE1
Subkeyword which indicates that Stone's First Model (see below) is to be
used to compute Kro.
*STONE2
Subkeyword which indicates that Stone's Second Model (see below) is to be
*LINEAR_ISOPERM
Subkeyword which indicates that the Linear Isoperm Model (see below) is to
be used to compute Kro.
*SEGREGATED
Subkeyword which indicates that a segregated model (see below) is to be
*SWSG
Subkeyword of *STONE1 and *STONE2 which indicates that Krow is
looked up as a function of Sw and Krog is looked up as a function of Sg for
use in the computation of Kro. *STONE2 *SWSG is the default when
*KROIL does not appear in the data set. When *KROIL *STONE2 or
*KROIL *STONE1 appear with no following subkeyword, the default
method for Krow and Krog is *SWSG. *STONE2 *SWSG is the default
when *KROIL appears with no following subkeyword.
*SO
Subkeyword of *STONE1 and *STONE2 which indicates that Krow is looked
up as a function of 1 - So and Krog is looked up as a function 1 - Swcon - So
for use in the computation of kro with Stone's second model. This can be used
to model Krow and Krog entered as functions of oil saturation only.
548 Rock-Fluid Data
User's Guide IMEX
DEFAULTS:
Optional keyword. If *KROIL is not encountered in the data set, the default is *KROIL
*STONE2 *SWSG. If *KROIL is encountered with no following subkeyword, the default is
*KROIL *STONE2 *SWSG. If *KROIL *STONE1 is encountered with no following
subkeyword, the default is *KROIL *STONE1 *SWSG. If *KROIL *STONE2 is
encountered with no following subkeyword, the default is *KROIL *STONE2 *SWSG.
CONDITIONS:
If it appears, this keyword must be in the Rock-Fluid Data keyword group. It can appear
anywhere within this group, but should only appear once. *KROIL is not rock type dependent
and applies to all tables. If *KROIL appears more than once, the last occurrence overwrites
previous occurrences.
EXPLANATION:
In STONE'S SECOND MODEL for kro as modified by Aziz and Settari, the oil relative
permeability is computed as
krow
kro = krocw *
+ krw
krocw
krog
+ krg krw krg
krocw
krw and krg are always looked up as functions of Sw and Sg, respectively. krow and krog are
read from water-oil and gas-liquid relative permeability tables, respectively. In IMEX, krow
is tabulated as a function of Sw and krog is tabulated as a function of Sg. If *KROIL
*STONE2 *SWSG is in effect and values of Sw, So, and Sg are given, krow is looked up in
the table as krow(Sw) and krog is looked up as krog(Sg). When *KROIL *STONE2 *SO is
in effect, krow is looked up from the table as krow(1 - So) and krog is looked up as krog(1Swcon-So). *SO corresponds to having krow and krog looked up directly as functions of So.
In some situations, the user may know krow and krog as functions of So rather than of Sw
and Sg. IMEX requires that the krow and krog curves be entered as functions of Sw and Sg,
respectively, but the user may enter krow(Sw = 1-So) as krow(So) and krog(Sg = 1-SwconSo) as krog(So) and specify *KROIL *STONE2 *SO. The effect is exactly the same as
having krog and krow tabulated directly as functions of So. In STONE'S FIRST MODEL as
modified by Aziz and Settari, kro is computed as
kro =
Sostar * krow * krog

krocw * (1 Swstar ) * (1 Sgstar )
where
Sostar =
Swstar =
(So Som )
(1 Swcon Som )
(Sw Swcon )
(1 Swcon Som )
User's Guide IMEX
Rock-Fluid Data 549
Sgstar =
Sg
(1 Swcon Som )
for the "minimal" value Som of the oil saturation, IMEX uses the linear function of Sg
proposed by Fayers and Matthews (SPEJ April 1984, pp. 224-232):
Som(Sg) = (1 - a(Sg)) * Sorw + a(Sg) * Sorg,
where
a (Sg ) =
Sg
(1 Swcon Sorg )
As in Stone's second model, when the subkeyword *SO is specified, krow(1-So) and krog(1Swcon-So) are used in the formula; otherwise, krow(Sw) and krog(Sg) are used.
The LINEAR ISOPERM model was proposed by L.E. Baker (paper SPE 17369, 1988). In
this method, kro(Sg,Sw) is defined in the region Swcon < Sw < 1 - Sorw, 0 < Sg < (1-SorwSw) * (1-Sorg-Swcon) / (1-Sorw-Swcon), by specifying curves (isoperms) along which kro
assumes a constant value. In particular, the isoperms are assumed to be straight line
segments. For example, if
kro(Sg=0,Sw) = krow(Sw) = 0.2
for Sw = Sw2
kro(Sg,Swcon) = krog(Sg) = 0.2
for Sg = Sg2,
and
then all points (Sg,Sw) on the line segment joining (0,Sw2) to (Sg2,Swc),
(Sg,Sw) = (0,Sw2) + b*(Sg2,Swcon-Sw2), 0 < b < 1,
also have kro(Sg,Sw) = 0.2 .
The *SEGREGATED model corresponds to a model which uses a segregated assumption for
the gas and water. So is assumed to be constant throughout the block and the water in the gas
zone is assumed to be equal to the connate water saturation. Using volume fraction
arguments and some algebraic manipulation
kro =
(Sg * krog + (Sw Swcon )* krow )

(Sg + Sw Swcon )
where krog is the oil relative permeability for a system with oil, gas and connate water
(created as in *SWSG but looked up as in *SO), and krow is the oil relative permeability for
a system with oil and water only (created as in *SWSG but looked up as in *SO). This model
gives similar results as STONE2 on the IMEX templates. However, on a large field problem,
significant differences were noted both in results and in CPU times.
This model was initially proposed by L.E Baker in SPE/DOE Paper (17369) presented at the
6-th Symposium on Enhanced Oil Recovery, Tulsa, Oklahoma.
550 Rock-Fluid Data
User's Guide IMEX
Rock Type (Optional)
*RTYPE
PURPOSE:
*RTYPE indicates the start of input of rock types. A relative permeability curve set_number
is assigned to each grid block (see *RPT).
ARRAY:
*RTYPE
DEFAULTS:
Optional keyword. The default is *RTYPE *CON 1, where 1 is a value for set_number; this
uses the first relative permeability curve set for all grid blocks.
CONDITIONS:
This keyword must be in the Rock-Fluid Data keyword group, after all the two phase relative
permeability curve tables have been entered.
This keyword can also be in the Well and Recurrent Data section as *RTYPE can be altered
during the run.
If saturation or relative permeability value end point array entry by rock type input is used,
then the *RTYPE array definition must be placed before any arrays (*SWCON etc) are
defined.
EXPLANATION:
If more than one set of corresponding water-oil and liquid-gas relative permeability tables is
input, then a set of curves must be assigned to each grid block.
*RTYPE indicates the input of an array of grid values. All array reading option subkeywords
are valid. See the section on Array Reading Options for more details.
When *RTYPE is redefined repeatedly in recurrent data, the effected block saturation end
points are individually reset to the new curve values. However, after an end point is read in
explicitly (say using the *SWCON keyword) this behavior changes.
After an end point (Swcon, Swcrit, Soirw, Slcon, Sorg, Sorw, Sgcrit, Sgcon, Krwiro, Krocw,
Krgcl, Krogcg, Pcwmax, Pcgmax, JFwmax, or JFgmax) is read in explicitly the individual
end point is always honored, even when the rock types change. End points not explicitly
defined are reset, but the user must explicitly alter an end point that has been entered using
one of the end point array keywords (*SWCON, *SWCRIT, *SOIRW, *SLCON, *SORW,
*SORG, *SGCRIT , *SGCON, *KRWIRO, *KROCW, *KRGCL, *KROGCG, *PCWMAX,
*PCGMAX, *JFWMAX, *JFGMAX ).
This procedure is different from the one used on the *PERF family of cards when rock type
and end points are read in. On the *PERF family of cards all values are reset to the values
defined. End points are always reset.
User's Guide IMEX
Rock-Fluid Data 551
Saturation Endpoints for each Grid Block (Optional)
*SWCON,
*SWCRIT, *SOIRW, *SORW, *SORMAX, *SGCON, *SGCRIT, *SLCON, *SORG
PURPOSE:
These keywords specify the end points of the relative permeabilities tables of each grid block.
The table curves are scaled based on these end points.
ARRAY:
*SWCON
*SWCRIT
*SOIRW
*SORW
*SORMAX
*SGCON
*SGCRIT
*SLCON
*SORG
DEFINITIONS:
*SWCON
Connate water saturation of a grid block. *SWCON is the first entry in the
*SWT table. 1.0-*SWCON is the maximum oil saturation possible in the
*SWT table.
*SWCRIT
Critical water saturation of a grid block. *SWCRIT is the saturation at which
water first becomes mobile in the two phase oil-water system. It is always
equal to or greater than *SWCON.
*SOIRW
Irreducible oil saturation of a grid block. 1-*SOIRW is the last water
saturation entry in the *SWT table. *SOIRW is the minimum oil saturation
possible in the *SWT table. 1-*SOIRW is the maximum water saturation in
the *SWT table.
*SORW
Residual oil saturation of a grid block. *SORW is the oil saturation at which
oil first becomes mobile in the two phase oil-water table. It is always equal
to or greater than *SOIRW.
*SORMAX
Maximum residual oil saturation of a grid block. Imbibition end point of the
water-oil hysteresis model. *SORMAX is the oil saturation at which oil first
becomes immobile for a boundary/primary imbibition process (imbibition
starting at oil saturation equals to 1-Swcon). Its value is always between 1*SWCON and *SORW. *SORMAX is only effective for blocks which have
552 Rock-Fluid Data
User's Guide IMEX
Krow hysteresis or trapped oil hysteresis enabled. Please see the tutorial
Trapped Oil Hysteresis for Pcow and Krow for discussions and graphs.
Oil-Water Table
P
cow
K
w
ro
0.0
rw
1.0
Sw
S wcon
S wcrit
1-S orw
1-S oirw
*SGCON
Connate gas saturation of a grid block. *SGCON is the minimum gas
saturation in the two phase gas-liquid system. It is the first entry in a *SGT
table and 1-*SGCON is the last entry in a *SLT table.
*SGCRIT
Critical gas saturation of a grid block. *SGCRIT is the gas saturation at which
gas first becomes mobile. It is always equal to or greater than *SGCON.
*SLCON
Connate liquid saturation of a grid block in the gas-liquid table. *SLCON
includes connate water (*SWCON) unless the *NOSWC option is employed.
It is the minimum liquid saturation in the two phase gas-liquid system and
the first entry in the *SLT table and one minus the last entry in the *SGT
table.
*SORG
Residual oil saturation of a grid block in the two phase gas liquid system. It
does not include connate water, thus *SORG + *SWCON = residual liquid
saturation. If the *NOSWC option is active *SORG = residual liquid
saturation.
User's Guide IMEX
Rock-Fluid Data 553
rog
cog
Liquid-gas Table (*NOSWC not active)
K
rg
S org
S oirg
0.0
1.0
Sl
S wcon
S lrg
1-S gcrit
1-S gcon
P
rog
cog
Liquid-gas Table (*NOSWC active)
S lcon
K
rg
S org
S oirg
0.0
1.0
Sl
S lcon
S lrg
1-S gcrit
1-S gcon
where Slrg is the residual liquid in the gas-liquid table.

Soirg is the irreducible oil in the gas liquid table.
554 Rock-Fluid Data
User's Guide IMEX
If *NOSWC is not active:

Slrg
Sorg + Swcon
Soirg
Slcon - Swcon
If *NOSWC is active:
Slrg
Sorg
Soirg
Slcon
DEFAULTS:
Optional keyword. The default is to use the end points from the relative permeability tables
corresponding to the *RTYPE region to which the grid block belongs. Any or all of the
above keywords may be omitted. In addition, when used in Well and Recurrent Data section
it is not necessary to specify the end point values for all the grid blocks. The grid blocks that
are not specified will have the default values of the end points (from tables corresponding to
the blocks *RTYPE region).
CONDITIONS:
This keyword can be in the Rock-Fluid Data keyword group after all the two phase relative
permeability tables have been entered and/or in the Well and Recurrent Data section as all of
the end points can be altered during the run.
The use of these end point keywords changes how the *RTYPE keyword in the Well and
Recurrent Data section chooses end point values for tables being reassigned to new grid
blocks. Please see *RTYPE for an explanation of how the end point keywords in the Well
and Recurrent Data section interact.
If saturation end point array entry by rock type input is used, then the *RTYPE array
definition must be placed before any end point arrays (*SWCON etc) are defined.
EXPLANATION:
The end point keywords indicate the input of an array of grid values. All array reading option
subkeywords are valid. See the section on Array Reading Options for more details.
These values override the values of the saturation end points. Once defined these values
remain in effect until explicitly overridden by another instance of these keywords. Changing
a blocks *RTYPE will not change a blocks saturation end point if it has been overridden by
these keywords (see *RTYPE Keyword). Without the use of these keywords, if a blocks
*RTYPE is changed, its saturation endpoints will be changed as well.
In addition to the standard array reading options, the eight saturation end point arrays can be
entered according to rock type. The format used when using this entry technique is:
Saturation End Point Array *RTYPE Rock Type Table number End Point Value
For example, if the value of connate water saturation for table number 3 is to be changed to
0.12, the format would be:
*SWCON *RTYPE 3 0.12
User's Guide IMEX
Rock-Fluid Data 555
This end point array reading format searches all blocks whose rock type is 3 and assigns the
connate water saturation of those blocks to the value of 0.12. The result is identical to the
result one would obtain if an *IJK array reading option was used to explicitly change each
block whose *RTYPE was 3 to 0.12.
Multiple occurrences of this format are valid, for example:
In addition it is possible to define a saturation end point array using a standard array reading
option and override the input values for specific rock types using this format, for example:
*SWCON *IJK 1:5 1:10 1 0.10
Initially all blocks connate water saturation are set to the saturation end points defined by
each rock types relative permeability table (the default), the keywords above override blocks
in the IJK range 1:5 1:10 1 with the connate water saturation of 0.10 and further set the
connate water saturation of all blocks in rock type 3 to 0.15.
SCALING
The relative permeabilities and capillary pressures are looked up on their respective tables
after the lookup saturation (either Sw or Sl) have been scaled appropriately.
For the oil-water table:
K rw , S 'w
S 'wcrit
+ (S w
1 S 'oirw S 'wcrit
1 S oirw S wcrit
S wcrit ) *
1 S 'orw S 'wcon
K row , S 'w' = S 'wcon + (S w S wcon ) *
1 S orw S wcon
1 S 'oirw S 'wcon
Pcow , S 'w'' = S 'wcon + (S w S wcon )
1 S oirw S wcon
Sw is the blocks water saturation

S 'w , S 'w' , S 'w'' are the saturations used to interpolate the original user entered tables with respect
to Krw, Krow and Pcow respectively.
S wcrit , S wcon , S oirw and S orw are the user entered grid block end point values (from the
keywords of the same name)
556 Rock-Fluid Data
User's Guide IMEX
S 'wcrit , S 'wcon , S 'oirw and S 'orw are the original table end point values (calculated directly from
the relative permeability tables)
For the gas-liquid table:
1 S 'lrg S 'gcon
K rog , S 'l = S 'lrg + S l S lrg *
1 S lrg S gcon
1 S 'lcon S 'gcrit
+ (S l S lcon ) *
1 S lcon S gcrit
1 S 'lcon S 'gcon
Pcog , S 'l'' = S 'lcon + (S l S lcon )
1 S lcon S gcon
K rg ,
S 'l'
S 'lcon
Sl = So + Sw
Slrg = Swcon + Sorg, if the *NOSWC option is not used.
Slrg = Sorg, if the *NOSWC option is used.
In the latter case Sorg includes all residual liquid in the liquid-gas system.
S 'l , S 'l' , S 'l'' are the saturations used to interpolate the original user entered tables with respect
to Krog, Krg and Pcog respectively.
Sorg, Slcow, Sgcon and Sgcrit are the user entered grid block end point values (from the keywords
of the same name)
S 'org , S 'lcon , S 'gcon and S 'gcrit are the original table end point values (calculated directly from the
relative permeability tables)
Note that each relative permeability curve is now scaled with respect to different end points
which define the range of each curve.
Krw
thus Krw is scaled by Swcrit and Soirw
Swcrit Sw 1 Soirw,
thus
Krow is scaled by Swcon and Sorw
Krow
Swcon Sw 1 Sorw,
Pcow
thus Pcow is scaled by Swcon and Soirw
Swcon Sw 1 Soirw,
thus if *NOSWC is not active, Krog is scaled by
Krog
Slrg Sl 1 Sgcon,
Swcon, Sorg and Sgcon. If *NOSWC is active Krog
is scaled by Sorg and Sgcon
Krg
thus Krg is scaled by Slcon and Sgcrit
Slcon Sl 1 Sgcrit,
Pcog
thus Pcog is scaled by Slcon and Sgcon
Slcon Sl 1 Sgcon,
THE *OILWET OPTION:
User's Guide IMEX
Rock-Fluid Data 557
When the oil wet option is active the column which normally contains water saturation
should now contain the saturation of the wetting phase (oil). The Krw column should contain
the wetting phase relative permeability (oil relative permeability). The Krow column contains
the nonwetting phase relative permeability (water relative permeability).
The meanings of the columns of the gas-liquid tables are not altered.
The use of the oil-wet option also changes the meaning of the user specified grid block
specific end point arrays
Swcon
Swcrit
Soirw
Sorw
Sormax
Sorg
Slcon
=
=
=
=
=
=
=
Connate wetting phase (oil) saturation

Critical wetting phase (oil) saturation
Irreducible nonwetting phase (water) saturation
Maximum residual nonwetting phase (water) saturation
Connate liquid saturation which is equal to connate oil
(Swcon) plus irreducible water saturation (Soirg) if the *NOSWC
option is not used, and equal to irreducible water saturation (Soirg)
alone, if the *NOSWC option is used
Please refer to the Tutorial section for a discussion of the IMEX oil wet option.
558 Rock-Fluid Data
User's Guide IMEX
Maximum Relative Permeability/Capillary Pressure Values for

Each Grid Block (Optional)
*KRWIRO, *KROCW, *KRGCL, *KROGCG,
*PCWMAX, *PCGMAX, *JFWMAX, *JFGMAX
PURPOSE:
These keywords specify the end point values of the relative permeabilities tables of each grid
block. The table curves are scaled based on these end points.
ARRAY:
*KRWIRO
*KROCW
*KRGCL
*KROGCG
*PCWMAX
*PCGMAX
*JFWMAX
*JFGMAX
DEFINITIONS:
*KRWIRO
Relative permeability value of water at the irreducible oil saturation in the
oil-water table (the last row in the table).
*KROCW
Relative permeability value of oil at the connate water saturation in the oilwater table (the first row in the table).
*KRGCL
Relative permeability value of gas at the connate liquid saturation in the gasliquid table (the first row in the table if SLT is used or the last row in the
table if SGT is used).
*KROGCG
Relative permeability value of oil at the minimum gas saturation in the gasliquid table (the last row in the table if SLT is used or the first row in the
table if SGT is used).
*PCWMAX
Oil-water capillary pressure value at the connate water saturation in the oilwater table (the first row in the table). PCWMAX is used to represent Gaswater capillary pressure if the GASWATER option is used and Gas-liquid
capillary pressure if the GASWATER_WITH_CONDENSATE option is
used. Do not use if the J Function option is specified.
User's Guide IMEX
Rock-Fluid Data 559
*PCGMAX
Oil-gas capillary pressure value at the connate liquid saturation in the gasliquid table. PCGMAX has no effect if the GASWATER or
GASWATER_WITH_CONDENSATE options are used. Do not use if the J
Function option is specified.
*JFWMAX
Oil-water J Function value at the connate water saturation in the oil-water
table (the first row in the table). JFWMAX is used to represent Gas-water J
Function if the GASWATER option is used and Gas-liquid J Function if the
GASWATER_WITH_CONDENSATE option is used. Only use if the J
Function option is specified.
*JFGMAX
Oil-gas J Function value at the connate liquid saturation in the gas-liquid
table. JFGMAX has no effect if the GASWATER or
GASWATER_WITH_CONDENSATE options are used. Only use if the J
function option is specified.
DEFAULTS:
Optional keyword. The default is to use the values from the relative permeability tables
corresponding to the *RTYPE region to which the grid block belongs.
CONDITIONS:
permeability tables have been entered or in the recurrent data section.
Any or all of the above keywords may be omitted. If any of the above keywords are used,
values for all grid blocks must be specified if input in the Rock-Fluid section. If input in the
recurrent data section it is not necessary to define all values, only those altered.
If relative permeability value end point array entry by rock type input is used, then the *RTYPE
array definition must be placed before any end point arrays (*KRWIRO etc) are defined.
EXPLANATION:
The end point keywords indicate the input of an array of grid values. All array reading option
subkeywords are valid. See the section on Array Reading Options for more details.
These values override the values of relative permeability, capillary pressures and J functions
at the endpoints. Once defined these values remain in effect until explicitly overridden by
another instance of these keywords. Changing a blocks *RTYPE will not change a blocks
endpoint if it has been overridden by these keywords (see *RTYPE Keyword). Without the
use of these keywords, if a blocks *RTYPE is changed, its endpoints will be changed as well.
In addition to the standard array reading options, the eight relative permeability end point
value arrays can be entered according to rock type. The format used when using this entry
technique is:
End Point Value Array *RTYPE Rock Type Table Number End Point Value
560 Rock-Fluid Data
User's Guide IMEX
For example, if the value of *KRWIRO for table number 3 is to be changed to 0.80, the
format would be:
*KRWIRO *RTYPE 3 0.80
This end point value array reading format searches all blocks whose rock type is 3 and
assigns the water relative permeability at irreducible oil of those blocks to the value of 0.8.
The result is identical to the result one would obtain if an *IJK array reading option was used
to explicitly change each block whose *RTYPE was 3 to 0.80.
Multiple occurrences of this format are valid, for example:
In addition it is possible to a define relative permeability end point array using a standard
array reading option and override the input values for specific rock types using this format,
for example:
*KRWIRO *IJK 1:5 1:10 1 0.90
Initially all blocks water relative permeability at irreducible oil are set to the values defined
by each rock types relative permeability table (the default), the keywords above override
blocks in the IJK range 1:5 1:10 1 with the water relative permeability at irreducible oil of
0.90 and further set the water relative permeability at irreducible oil of all blocks in rock type
3 to 0.80.
THE *OILWET OPTION:
The use of the oil-wet option also changes the meaning of the user specified grid block
specific end point arrays. *KRWIRO, *KROCW, *KRGCL, *KROGCG, *PCWMAX,
*PCGMAX, *JFWMAX, *JFGMAX. Please refer to the Tutorial section for a discussion of
the IMEX oil wet option.
User's Guide IMEX
Rock-Fluid Data 561
Water Oil-Capillary Pressure (J Function) Shift for Each Grid

Block (Optional)
*PCOW_SHIFT, *JFW_SHIFT
PURPOSE:
These keywords specify the amount that the Water-oil capillary pressure or water-oil J
Function curve is shifted up or down. This keyword can be used to account for the effects of a
tilted water-oil contact (See Tutorial section).
ARRAY:
*PCOW_SHIFT
*JFW_SHIFT
DEFINITIONS:
*PCOW_SHIFT
*PCOW_SHIFT specifies the amount that a grid blocks water-oil capillary
pressure curve is shifted up or down from the table value. PCOW_SHIFT is used
to represent Gas-water capillary pressure if the GASWATER option is used and
Gas-liquid capillary pressure if the GASWATER_WITH_CONDENSATE
option is used. Do not use if the J Function option is specified.
*JFW_SHIFT
*JFW_SHIFT specifies the amount that a grid blocks water-oil J Function
curve is shifted up or down from the table value. JFW_SHIFT is used to
represent the Gas-water J Function if the GASWATER option is used and the
Gas-liquid J Function if the GASWATER_WITH_CONDENSATE option is
used. Only use if the J Function option is specified.
DEFAULTS:
Optional keyword. The default is to zero.

CONDITIONS:
permeability tables have been defined. This keyword may only be used if the DEPTH_AVE
initialization option is used.
EXPLANATION:
The PCOW_SHIFT keyword indicates the input of an array of grid values. All array reading
option subkeywords are valid. See the section on Array Reading Options for more details.
These values are added to the value of Pcow or water J functions. The shift is applied after
the maximum capillary pressure/J Function of a block has been set. For example, if a
maximum Pcow has been defined using PCWMAX and a Pcow shift has been defined using
PCOW_SHIFT, the actual block maximum is the sum of the two. PCOW_SHIFT can be
used to model tilting water-oil or water-gas contacts (see the Tutorial section on how to
model tilting contacts).
562 Rock-Fluid Data
User's Guide IMEX
Non-Darcy Flow in Reservoir (Optional)
*NONDARCY
PURPOSE:
*NONDARCY signals the use of the non-Darcy reservoir flow option.

FORMAT:
*NONDARCY
(*GEERTSMA)
(*FG1)
(*FG2)
-or*NONDARCY *GENERAL
Coefficient Table
wt_ndarcy
DEFINITIONS:
*GEERTSMA
*GEERTSMA signals the use of Geertsma's (1974) correlation for gas phase
non-Darcy flow.
*FGI
*FG1 signals the use of Frederick and Graves (1994) first correlation for gas
phase non-Darcy flow.
*FG2
*FG2 signals the use of Frederick and Graves (1994) second correlation for
gas phase non-Darcy flow.
*GENERAL
*GENERAL signals that a table of coefficients follows which will define
how non-Darcy flow is a function of saturation, porosity and permeability for
each phase.
wt_ndarcy
A weighting factor which controls how much the non Darcy Forchheimer
number is allowed to change over a timestep. The default is 1.00. Excellent
results are obtained when a value of 0.50 is used. Values above 0.50 are not
recommended.
Coefficient Table
A table of coefficients for (beta) the non-Darcy flow coefficient for each phase (defined
below). (beta) always has units of 1/ft. regardless of the unit system used in the simulator.
It is always assumed that we are using Permeability in mD and a g so that we obtain a
factor with units of 1/ft.
User's Guide IMEX
Rock-Fluid Data 563
Background:
For any phase the Forchheimer equation can be written as

p =
v
v
+ p p U p U p
K K rp
where p is the potential of phase P and p is the non-Darcy coefficient for that phase.
The above equation can be rewritten as:
p =
with
p
K K rp
(1 + Fo p ) Uv p
v
Fo p = p K K rp p U p / p
Fo is referred to as the Forchheimer Number.

The correlations for p are all of the form (K in md, and p in 1/ft.)
p =
(K K ) (S )
N1p
rp
N 2p
When using the *NONDARCY *GEERTSMA, *FG1 or *FG2 option for all phases except
gas is assumed to be zero. For gas the following parameters for g, N1g and N2g are used.
Correlation
N1g
*Geertsma
48511.34
0.5
5.5
*FG1
7.89E10
1.6
0.404
*FG2
2.11E10
1.55
1.0
N2g
When using the *NONDARCY *GENERAL option the user can enter a table of , N1, N2
and Forchmax (defined below) for each phase in the reservoir.
It is always assumed that we are using Permeability in mD and a g so that we obtain a
factor with units of 1/ft.
The first three columns of the table represent , N1 and N2 respectively, while the rows
indicate which phase the coefficients correspond to. The last column (4th) is Forchmax, which
is normally defaulted to 1.00e+04, can be used to input the maximum Forchheimer no. for
each phase. A Forchheimer no. calculated to be greater than Forchmax is limited to Forchmax. It
is recommended that Forchmax is allowed to default to 1.00e+04.
The Coefficient Tables are of the following form, the form differs slightly for different
*MODEL (model) types
564 Rock-Fluid Data
User's Guide IMEX
For models with gas, water and oil phases present (no solvent) the form of the coefficient
tables is:
*NONDARCY *GENERAL
0.50
gas
N1g
N2g
Forchmaxg
water
N1w
N2w
Forchmaxw
oil
N1o
N2o
Forchmaxo
For models with gas and water phases present (no oil or solvent) the form of the coefficient
tables is:
*NONDARCY *GENERAL
0.50
gas
N1g
N2g
Forchmaxg
water
N1w
N2w
Forchmaxw
For models with solvent phase present the form of the coefficient tables is:
*NONDARCY *GENERAL
0.50
gas
N1g
N2g
Forchmaxg
water
N1w
N2w
Forchmaxw
oil
N1o
N2o
Forchmaxo
solvent
N1s
N2s
Forchmaxs
For models with oil and water phases present (no gas or solvent) the form of the coefficient
tables is:
*NONDARCY *GENERAL
0.50
water
N1w
N2w
Forchmaxw
oil
N1o
N2o
Forchmaxo
Note: A value of N2p = 0.0 will eliminate the dependence of p on ( x Sp) and allow a
correlation for p of the form:
p =
(K K )
N1p
rp
This form is common when modelling non Darcy flow in fractures.
User's Guide IMEX
Rock-Fluid Data 565
Non-Darcy Coefficient Correction Factor (Conditional)

*NDARCYCOR
PURPOSE:
Allows the input of block by block correction factors for the non-Darcy factor calculated as
described in the background section for the *NONDARCY keyword.
ARRAY:
*NDARCYCOR
DEFAULTS:
Conditional keyword. By default all values are 1.0.

CONDITIONS:
Used only when non-Darcy flow is needed.

DEFINITION:
This array allows the user to modify the Forchheimer Number for any block. The correction
is only applied to the gas phase unless the *NONDARCY *GENERAL option is used. If
*NONDARCY *GENERAL is used, the correction is applied to all phases.
Setting *NDARCYCOR to zero in a region of the reservoir will force that region to exhibit
Darcy flow and significantly reduce the non Darcy models calculations. The factor is
dimensionless.
*NDARCYCOR can also be used to effectively model non Darcy flow in thin grid blocks
which are set up to represent fractures. The proper definition of *NDARCYOR, which is a
function of the fracture pseudo permeability and of N1p (in the correlation for non Darcy
factor ), will allow the user to model non Darcy flow in 0.10 inch thick fractures using grid
blocks as wide as 1 foot. See the tutorial section on modelling non Darcy flow in fractures
for a discussion of how this is done.
*NDARCYCOR may be redefined in recurrent data.
566 Rock-Fluid Data
User's Guide IMEX
Fracture Velocity Width Correction for PolymerWater Mixture

Viscosity (Conditional)
*FRWIDTHCOR
PURPOSE:
Allows the input of block by block correction factors for the modification of the velocity used
to interpolate in the *PMIX *VELTABLE table.
ARRAY:
*FRWIDTHCOR
DEFAULTS:
Conditional keyword. By default all values are 1.0.

CONDITIONS:
Used only when *PMIX *VELTABLE is input.

DEFINITION:
This array allows the user to modify the water velocity used to interpolate in the *PMIX
*VELTABLE polymerwater mixture viscosity table.
It is envisioned that this correction factor would be used in blocks which are attempting to
represent fractures. It is assumed that these simulated fracture blocks are small, on the order
of 0.2 meter wide (for example), but are not small enough to accurately model the velocities
in a fracture which (we will assume) has an actual fracture width of 0.1 millimeters.
The user might ensure that the actual fracture KA and the simulated fracture block KA (and
hence the flow) are similar, but the velocity in the simulated fracture block would be lower by
the ratio of the real fracture area to the simulated fracture block area (in this example 2,000.0
times lower)
A value of FRWIDTHCOR of 2,000.0 could be used to multiply the calculated velocity in a
simulated fracture block by 2,000.0 before interpolating in the *PMIX *VELTABLE table to
find the velocity dependent viscosity of the waterpolymer mixture.
A value of FRWIDTHCOR = 0.0 in a grid block sets the velocity used in the *PMIX
*VELTABLE viscosity calculation to 0.0. This is a convenient way to limit the region where
the polymer water viscosity calculation is a function of velocity.
A value of FRWIDTHCOR = 1.0 in a grid block sets the velocity used in the *PMIX
*VELTABLE viscosity calculation to the intrinsic block velocity.
User's Guide IMEX
Rock-Fluid Data 567
Pressure Gradient Threshold Arrays
*PTHRESHI, *PTHRESHJ,
*PTHRESHK
PURPOSE:
Defines pressure gradients on interblock flow boundaries below which flow is not initiated.
ARRAY:
*PTHRESHI
*PTHRESHJ
*PTHRESHK
DEFINITIONS:
*PTHRESHI
Pressure gradient threshold for the flow between block I,J,K and I+1,J,K.
(kPa/m | Psi/ft | kPa/cm | kg/cm2/m).
*PTHRESHJ
Pressure gradient threshold for the flow between block I,J,K and I,J+1,K.
*PTHRESHK
Pressure gradient threshold for the flow between block I,J,K and I,J,K+1.
The pressure gradient threshold is a one time switch. Once it is exceeded between two blocks
in a specific direction, the interblock connection remains open to flow ever more.
A zero value of the thresholds indicate the connection is always open to flow.
Note that the thresholds are defined on connections/interfaces between blocks and not on the
blocks themselves.
DEFAULTS:
The defaults for all values are 0.0. This allows flow between connections without a pressure
gradient to be overcome.
CONDITIONS:
This keyword must be in the Rock-Fluid Data keyword group. This keyword may be
introduced in this section (not in recurrent data) at a restart to model injection followed by
production in a fracture clean-up simulation (See Tutorial on Using the Fracture Clean-up
Model).
EXPLANATION:
The pressure gradient thresholds are used to approximately model yield stress. Once the
pressure gradient for flow has been exceeded on an inter-block connection, the threshold for
that connection is not checked again.
568 Rock-Fluid Data
User's Guide IMEX
The pressure gradient which initiates flow is corrected for the initial (usually time = 0.0)
pressure in the reservoir. This prevents flow due to the difference in head in the reservoir.
When the pressure gradient threshold keywords are introduced on a restart, the initial
reservoir pressure used to correct the pressure gradient is the pressure in the reservoir at the
date of the first restart when the pressure gradient thresholds were introduced.
User's Guide IMEX
Rock-Fluid Data 569
Leverett J Function Option
*JFUNC
PURPOSE:
Activates the Leverett J Function option. This option allows the user to replace water-oil
and/or oil-gas capillary pressures entered with Leverett J Functions. When the *GASWATER
model option is used, the entered water-gas capillary pressure is replaced by a water-gas J
Function. When the *GASWATER_WITH_CONDENSATE model option is used, the
entered liquid-gas capillary pressure is replaced by a liquid-gas J Function.
FORMAT:
*JFUNC
(*BOTH)
(*WATER)
(*GAS)
DEFINITIONS:
*JFUNC
Activate J function option. All capillary pressure entries in relative
permeability tables are now interpreted as J Functions. All rock types are
effected by a single *JFUNC keyword which must precede all *RPT
keywords.
*BOTH
Both water-oil and oil-gas capillary pressures are interpreted as J functions
for all rock types defined (Default).
*GAS
Oil-gas capillary pressures are interpreted as J functions. Water-oil capillary
pressures are still interpreted as capillary pressures.
*WATER
Water-oil capillary pressures are interpreted as J functions. Oil-gas capillary
pressures are still interpreted as capillary pressures.
When using the *GASWATER model option, use of any of the *JFUNC,
*JFUNC *BOTH, *JFUNC *WATER, or *JFUNC *GAS keywords will
enable the interpretation of water-gas capillary pressures as J functions.
When using the *GASWATER_WITH_CONDENSATE model option, use
of any of the *JFUNC, *JFUNC *BOTH, *JFUNC *WATER, or *JFUNC
*GAS keywords will enable the interpretation of liquid-gas capillary
pressures as J functions.
DEFAULTS:
Optional keyword, capillary pressure table entries are interpreted as capillary pressures unless
*JFUNC is encountered, If *JFUNC is found, some capillary pressure tables (based on
subkeywords) may be read in as J functions. If *JFUNC is encountered without an optional
subkeyword, *JFUNC *BOTH is assumed.
570 Rock-Fluid Data
User's Guide IMEX
CONDITIONS:
This keyword must be in the Rock-Fluid Data section and must be placed before the first
*RPT keyword.
EXPLANATION:
The J function option allows the user to scale capillary pressure to account for differences in
block porosity and permeability. When used with the *SRFTNW and *SRFTNG keywords
to enter individual block surface tension terms (water-oil surface tension using *SRFTNW
and oil-gas surface tension using *SRFTNG), the user can additionally scale capillary
pressure based on block location.
When enabled, J functions replace capillary pressure in the relative permeability table reading
options (*SWT, *SGT, *SLT). When the *GASWATER model option is used, water-gas
capillary pressure is interpreted as water-gas J function. Water-gas surface tension is entered
using the *SRFTNW keyword.
When the *GASWATER_WITH_CONDENSATE model option is used, liquid-gas capillary
pressure is interpreted as liquid-gas J function. Liquid-gas surface tension is entered using
the *SRFTNW keyword.
The effect of block dependent porosity, permeability and surface tension is accounted for in
the Gravity Equilibrium initialization options and will produce spatially varying transition
zones which are a result of spatially varying capillary pressures.
Defining:
J(SW)
J(Sg) or J(Sl)
*SRFTNW
*SRFTNG
CF
Block Porosity at Reference Pressure

Block Permeability, normally
K x * K y unless K x * K y
is zero, then the largest of Kx, Ky, and Kz is used (md)
Dimensionless J function entered in the *SWT table. This
replaces the capillary pressure column for Pcow (or Pcwg)
Dimensionless J function entered in the *SGT or *SLT table.
This replaces the capillary pressure column for Pcog
Oil-water or water-gas Surface tension or Surface tension *
Cosine (Contact Angle) (dyne/cm)
Oil-gas Surface tension or Surface tension * Cosine (Contact
Angle) (dyne/cm)
Conversion Factor,
for Capillary pressure in kPa = 31.8316
for Capillary pressure in Psi = 4.61678
Then the true capillary pressure for the oil-water table is:
Pcow = J (Sw ) * SRFTNW *
User's Guide IMEX
* CF
K
Rock-Fluid Data 571
For the water-gas capillary pressure:

Pcwg = J (Sw ) * SRFTNW *
* CF
K
Similarly for the oil-gas table:

Pcog = J (Sg ) * SRFTNG *
* CF
K
Pcog = J (Sl ) * SRFTNG *
* CF
K
or
If oil-gas surface tension is defined as a function of pressure by entering an appropriate

nonzero column in one of the PVT table entry options and by entering a single reference
pressure (*REFPST ) to be used in all PVT regions, then the Pcog calculations for each block
is corrected by:
PVT Table Surface tension (function of P and PVT region )
PVT Table Surface tension at Reference pressure (function of * REFPST and PVT region )
Note: Although there is a single reference pressure for surface tension which is shared by all
PVT tables. Each table may have a different surface tension at the reference pressure.
572 Rock-Fluid Data
User's Guide IMEX
J Function Surface Tension Arrays
*SRFTNW, *SRFTNG
PURPOSE:
Defines Surface tensions to be used in the J Function Option.

ARRAY:
*SRFTNW
*SRFTNG
DEFINITIONS:
*SRFTNW
In the calculation of Pcow or Pcwg (Pclg - condensate option) when using the J
function option, the capillary pressure is a function of J function, porosity,
permeability and surface tension (or surface tension * Cosine (Contact
Angle). This grid block array can be used to enter a different surface tension
term for each block in the reservoir (dyne/cm).
*SRFTNG
In the calculation of Pcog when using the J function option, the capillary
pressure is a function of J function, porosity, permeability and surface
tension (or surface tension * Cosine (Contact Angle). This grid block array
can be used to enter a different surface tension term for each block in the
reservoir (dyne/cm).
DEFAULTS:
Optional keywords which are used only when the J function option is active. When the J
function option is in use, the default values are 1.00. This implies the surface tension effects
are constant with grid block and are already multiplied into the J function terms entered in the
*SWT, *SGT, and *SLT tables.
When using the *GASWATER or *GASWATER_WITH_CONDENSATE option, only
*SRFTNW can be used to enter surface tension.
User's Guide IMEX
Rock-Fluid Data 573
Scale Deposition Table (Optional)
*SCLDPS
PURPOSE:
*SCLDPS indicates the start of a seawater option scale deposition table.

TABLE:
*SCLDPS (Scale Deposition Table Set Number)

Seawater fraction
:
Precipitation
:
*SCLDPS tables allow the user to enter scale deposition as a function of seawater volume
fraction for individual scale deposition set numbers. These sets can then be applied to
individual wells and/or well layers using *SCLTBL-WELL and SCLTBL-LAYER as
described in the Recurrent Data section
DEFINITIONS:
*SCLDPS
Scale Deposition Table. A table which defines how seawater volume fraction
influences the precipitation of scale onto a well.
Scale Deposition Table Set Number
The set number of the table. Each well or layer of a well can be assigned to a
deposition table through this set number. The first set number must be 1 and
each additional deposition table must have a set number which increments by 1.
Seawater fraction
Seawater volume fraction. Seawater fraction must be no less than zero and
no greater than one. In addition the seawater fraction must monotonically
increase down the tables rows (fraction).
Precipitation
Amount of scale deposition per unit flow rate of water flow though a well
perforation (kg/m3 | lb/ft3| g/cm3| kg/m3).
DEFAULTS:
Optional keyword. There are no defaults.

CONDITIONS:
This keyword must be in the Rock-Fluid Data keyword group.

EXPLANATION:
This table defines the relationship between the seawater volume fraction and the amount of
scale precipitated per unit flow rate of water produced.
574 Rock-Fluid Data
User's Guide IMEX
The amount of scale deposited on a perforation over a timestep is calculated as the product of
the amount of scale precipitated per unit flow rate of water and the water flow through the
perforation. If the *BLACKOIL_SEAWATER or *OILWATER_SEAWATER options are
used, the amount precipitated per unit flow rate of water is also a function of seawater volume
fraction produced. Without the seawater injection (tracking) options, the tables will always
use the 0.0 volume fraction seawater value of precipitation to work out the amount deposited
over a timestep.
User's Guide IMEX
Rock-Fluid Data 575
Scale Damage Table (Optional)
*SCLDMG
PURPOSE:
*SCLDMG indicates the start of a seawater option scale damage table.

TABLE:
*SCLDMG (Scale Damage Table Set Number)

Scale Deposited
:
Damage Factor
:
*SCLDMG tables allow the user to enter scale well damage as a function of scale deposited
per perforation length for individual scale damage set numbers. These sets can then be
applied to individual wells and/or well layers using *SCLTBL-WELL and SCLTBL-LAYER
as described in the Recurrent Data section.
DEFINITIONS:
*SCLDMG
Scale Damage Table. A table which defines how a layers or wells productivity
is reduced as a function of the scale deposited per perforation length.
Scale Damage Table Set Number
The set number of the table. Each well or layer of a well can be assigned to a
damage table through this set number. The first set number must be 1 and each
additional deposition table must have a set number which increments by 1.
Scale Deposited
The amount of scale deposited per perforation length (kg/m | lb/ft| g/cm|
kg/m).
Damage Factor
The layer or well index multiplier which represents a reduction in
productivity caused by the amount of scale deposited (fraction)
DEFAULTS:
Optional keyword. There are no defaults.

CONDITIONS:
This keyword must be in the Rock-Fluid Data keyword group.

EXPLANATION:
When seawater and formation water flow simultaneously into a well, scale is precipitated.
The amount of scale deposited is a function of rate of water flow through the perforations and
the relative amounts of formation water and seawater. The amount of scale which is deposited
accumulates through the simulation (unless removed using the *SCLRMV-WELL keyword
in the Recurrent Data section). This table refines the relationship between the cumulative
scale deposited per perforation length and the well damage.
576 Rock-Fluid Data
User's Guide IMEX
Subdomain Reinfiltration (Optional)
*SD_REINF
PURPOSE:
*SD_REINF works with TRANSD to modify the Inter-Subdomain connection (defined by

TRANSD > 0.0) to represent reinfiltration.
ARRAY:
*SD_REINF
DEFAULTS:

CONDITIONS:
This keyword may be in the Rock-Fluid Data keyword group only and is only valid when the
*SUBDOMAIN dual porosity option is in effect and TRANSD is defined for at least one
Inter-Subdomain connection. Use of this keyword is enhanced if the *FRACVOL Reservoir
Description Keyword is used to reduce the relative size of the top and bottom blocks in the
subdomain.
EXPLANATION:
The use of the TRANSD array allows the matrix stacks in each subdomain to be connected
vertically together. This allows gravity drainage to occur directly from subdomain to
subdomain.
There is normally capillary continuity between subdomains (see *TRANSD).
The *SD_REINF array introduces capillary discontinuity between the bottom block of a
matrix subdomain and other subdomain blocks it is connected to (including the connection
created using *TRANSD
The lowest subdomain block stands in for the fracture when modeling flow from the matrix
into a vertical fracture then reinfiltration from the vertical fracture back into the matrix.
In order to be valid, it is important that *FRACVOL be used to reduce the thickness of the
last subdomain block, as it is this subdomain block which is used to represent the vertical
fracture between matrix blocks. See keywords *FRACVOL and *TRANSD.
Normally values of *SD_REINF = 1.0 or 0.0 are used. This turns the reinfiltration model on
(1.0) or off (0.0). However in certain instance it may be desirable to use this model to
approximately account for limited reinfiltration which occurs alongside gravity drainage. In
this case, fractional values of *SD_REINF may be entered.
*SD_REINF introduces a capillary discontinuity by modifying the bottom-most subdomain
blocks Pcog and Pcow using the following formula (for each subdomains lowest block).
A scale factor (PCSCALE) is calculated for both oil-water and oil-gas capillary pressure in
the lowest subdomain block.
PCSCALE = 1.00 *SD_REINF
User's Guide IMEX
Rock-Fluid Data 577
The capillary pressures of that single subdomain block are adjusted by PCSCALE.
PCOW = PCOW*PCSCALE
PCOG = PCOG*PCSCALE
Examples:
The following provides an example of TRANSD and *SD_REINF usage:

**In Reservoir Description
** Turn on direct gravity drainage between subdomains
** for all blocks
*TRANSD *CON 1.0
:
** In Rock-Fluid Data
** Introduce capillary discontinuity in inter
** Subdomain flow.
*SD_REINF CON 1.0
The suggested range of values for SD_REINF is:

min
max
578 Rock-Fluid Data
SI
0.0
1.0
Field
0.0
1.0
Lab
0.0
1.0
User's Guide IMEX
Initial Conditions
Notes on Initial Conditions

IMEX currently provides seven different ways to initialize the reservoir.
1. *USER_INPUT, formerly *VERTICAL *OFF, which is still accepted by IMEX.
Pressure, water saturation and oil saturation are specified for each grid block. Gas
saturations are then determined by subtraction. Use the following keywords for
this option:
*PRES, *PB or *PBT, *SW and *SO (*API or *APIT if the API tracking model is
used, *PDEW or *PDEWT if the *VOLATILE_OIL or
*GASWATER_WITH_CONDENSATE model is used).
2. *VERTICAL *BLOCK_CENTER *WATER_OIL formerly *VERTICAL *ON,
which is still accepted by IMEX, with gas-oil contact depth set outside the
reservoir. In this approach the water-oil contact depth, together with a reference
pressure at a reference depth are specified. Bubble point pressure for each grid
block or as a table versus depth must be assigned by the user. Saturations for
entire grid blocks are set equal to their values at the grid block center.
Gravity-capillary equilibrium calculations are performed to calculate all grid block
pressures and saturations.
For this approach use the following keywords:
*DWOC, *REFDEPTH, *REFPRES, *PB or *PBT (*API or *APIT if the API
tracking model is used, *PDEW or *PDEWT if the *VOLATILE_OIL model is
used).
3. *VERTICAL *BLOCK_CENTER *WATER_GAS, formerly *VERTICAL *ON,
which is still accepted by IMEX, with water-oil and gas/oil contact depths set
equal. This is gravity initialization for reservoirs containing only water and gas
phases. In this approach the water-gas contact depth, together with a reference
pressure at a reference depth are specified. Saturations for entire grid blocks are
set equal to their values at the grid block center.
pressures and saturations.
*DWGC, *REFDEPTH, *REFPRES, *PB or *PBT (*PDEW or *PDEWT
*GASWATER_WITH_CONDENSATE model)
User's Guide IMEX
Initial Conditions 579
4. *VERTICAL *BLOCK_CENTER *WATER_OIL_GAS, formerly *VERTICAL

*ON, which is still accepted by IMEX. The water-oil contact and gas-oil contact
depth, together with a reference pressure at a reference depth must be specified.
Saturations for entire grid blocks are set equal to the value at the block center.
pressures and all oil, gas and water saturations.
*DWOC, *DGOC, *REFDEPTH, *REFPRES, *PB or *PBT (*API or *APIT if
the API tracking model is used, *PDEW or *PDEWT if the *VOLATILE_OIL
model is used).
To use this option to initialize a water-gas reservoir, set the water-oil contact depth
equal to the gas-oil contact depth. The result in this case will be the same as that
obtained under the *BLOCK_CENTER *WATER_GAS option listed above. Note
that the *GASWATER and *GASWATER_WITH_CONDENSATE options
require the use of the *WATER_GAS initialization option.
5. *VERTICAL *DEPTH_AVE *WATER_OIL. Gravity initialization for an undersaturated reservoir with saturations averaged over the depth interval covered by a
grid block. In this approach, for example, if a grid block has its block center
slightly below the water-oil contact depth, the water saturation assigned to the
block is the average over the block volume of the local saturations, and not simply
the water saturation value below the water-oil contact.
For this option, the water-oil contact depth, together with a reference pressure and
a reference depth must be specified. Note that these are the same data required
when using option (2) above (*VERTICAL *BLOCK_CENTER *WATER_OIL).
When this initialization is performed, a table is constructed giving oil phase
pressure, water-oil capillary pressure, and several other properties as functions of
depth. This table is constructed by Runge-Kutta integration of the equation
determining the hydrostatic pressure. Water saturation is then determined as a
function of depth by inverting the table which gives water-oil capillary pressure as
a function of saturation. Block water saturations are then determined by numerical
quadrature of the local water saturation.
For this approach use the following associated keywords:
*DWOC, *REFDEPTH, *REFPRES, *PB or *PBT (*API or *APIT if the API
tracking model is used, *PDEW or *PDEWT if the *VOLATILE_OIL model is
used).
6. *VERTICAL *DEPTH_AVE *WATER_OIL_GAS. Gravity initialization for a
saturated reservoir with a gas cap, with saturations averaged over the depth interval
covered by a grid block. In this approach, for example, if a grid block has its block
center slightly above the gas-oil contact depth, the gas saturation assigned to the
block is the average over the block volume of the local gas saturation, and not
simply the gas saturation value in the gas cap.
580 Initial Conditions
User's Guide IMEX
For this option, the water-oil and gas-oil contact depths, together with a reference
pressure and a reference depth must be specified. Note that these are the same data
required when using option (4) above (*VERTICAL *BLOCK_CENTER
*WATER_OIL_GAS).
When this initialization is performed, a table is constructed giving oil-phase
pressure, water-oil capillary pressure, oil- gas capillary pressure, and several other
properties as functions of depth. This table is constructed by Runge-Kutta
integration of the equation determining the hydrostatic pressure. Water and gas
saturations are then determined as functions of depth by inverting the tables which
give water-oil capillary pressure and oil/gas capillary pressure as functions of
saturation. Block water and gas saturations are then determined by numerical
quadrature of the local water saturation.
*DWOC, *DGOC, *REFDEPTH, *REFPRES, *PB or *PBT (*API or *APIT if
the API tracking model is used, *PDEW or *PDEWT if the *VOLATILE_OIL
model is used).
7. *VERTICAL *DEPTH_AVE *WATER_GAS. Gravity initialization for a
reservoir in which all of the hydrocarbon fluid is initially in the gas phase, with
saturations averaged over the depth interval covered by a grid block. In this
approach, for example, if a grid block has its block center slightly below the watergas contact depth, the water saturation assigned to the block is the average over the
block volume of the local water saturation, and not simply the water saturation
value in the water zone.
For this option, the water-gas contact depth, together with a reference pressure and
a reference depth must be specified.
When this initialization is performed, a table is constructed giving water or gas
phase pressure and several other properties as functions of depth. This table is
constructed by Runge-Kutta integration of the differential equation determining the
hydrostatic pressure. Water and gas saturations are then determined as functions
of depth. Block water and gas saturations are then determined by numerical
quadrature of the local saturations.
In order to properly account for the effects of gas-water capillary pressure (Pcgw),
the *TRANSZONE subkeyword must be used with *VERTICAL *DEPTH_AVE
*WATER_GAS.
Note that the *GASWATER and *GASWATER_WITH_CONDENSATE options
require the use of the *WATER_GAS initialization option.
*DWGC, *REFDEPTH, *REFPRES, *PB or *PBT (*PDEW or *PDEWT
*GASWATER_WITH_CONDENSATE model)
User's Guide IMEX
Initial Conditions Identifier (Required)
*INITIAL
PURPOSE:
*INITIAL indicates the beginning of initial condition values.
FORMAT:
*INITIAL
DEFAULTS:
CONDITIONS:
This keyword must be the first keyword in the Initial Conditions keyword group, which must
come immediately after the Rock-Fluid Data keyword group.
User's Guide IMEX
User Input or Vertical Equilibrium Selection (Required)

*USER_INPUT, *VERTICAL
PURPOSE:
*USER_INPUT and *VERTICAL control the type of reservoir initialization calculation.
FORMAT:
*USER_INPUT
*VERTICAL (*BLOCK_CENTER) (*WATER_OIL)
(*WATER_GAS)
(*WATER_OIL_GAS)
(*DEPTH_AVE)
(*WATER_OIL)
(*WATER_GAS)
(*WATER_OIL_GAS)
(*EQUIL)
(*NOEQUIL)
(*TRANZONE)
(*NOTRANZONE)
DEFINITIONS:
*USER_INPUT
Specify pressure and saturations of each grid block. Use this subkeyword
only when the pressure and saturations at each grid block need to be defined
individually. This keyword replaces, and is exactly equivalent to, the
*VERTICAL *OFF option of earlier versions, which is still accepted by
IMEX.
*VERTICAL
Indicates that pressures are determined from the hydrostatic equation and
saturations from the capillary pressure tables.
*BLOCK_CENTER
Sub option of *VERTICAL. Assign block saturations as the saturation
prevailing at the block center.
*DEPTH_AVE
Sub option of *VERTICAL. Assign block saturations as averages over the
depth interval spanned by the grid block.
*WATER_OIL
Perform gravity-capillary equilibrium initialization of a reservoir initially
containing no gas.
*WATER_GAS
Perform gravity-capillary equilibrium initialization for reservoirs with only
water and gas phases initially present.
User's Guide IMEX
*WATER_OIL_GAS
Perform gravity-capillary equilibrium initialization of a reservoir initially
containing water, oil, and gas phases.
*EQUIL
With *DEPTH_AVE options, add a pressure correction to each phase (during
the simulation) in order that the reservoir initially be in gravitational
equilibrium. The averaging of saturations over depth prevents gravitational
equilibrium from being established simply by setting saturations from capillary
pressure tables. The calculated pressure corrections can be printed by using
the *OUTPRN *RES *ALL keywords in the Input/Output Control section.
*NOEQUIL
Do not add a phase pressure correction when a *DEPTH_AVE option is used.
*TRANZONE
Subkeyword for the *WATER_GAS option only, which specifies that the wateroil capillary pressure curves entered by the user should be used to determine a
water-gas transition zone in which the gas saturation is above its critical value
and the water saturation is between 1.0 and Swcon (It is assumed for the
*WATER_GAS option that *SOIRW and *SORW are both equal to 0.0). The
presentation of this subkeyword in the format line above was constrained by
page width and may be misleading. Note that *TRANZONE should be used
with the *WATER_GAS option only. *NOTRANZONE is the default.
*NOTRANZONE
A subkeyword for the *WATER_GAS option only, which specifies that the
transition from the water to the gas zone should be sharp, before the
averaging over grid block thicknesses. *NOTRANZONE is the default. If
neither *TRANZONE nor *NOTRANZONE is encountered after the
*WATER_GAS subkeyword, then *NOTRANZONE is assumed.
DEFAULTS:
One of *USER_INPUT or *VERTICAL is required. No default.
If *VERTICAL is entered without any subkeyword, *VERTICAL *BLOCK_CENTER
*WATER_OIL_GAS is assumed, which is also the case if *VERTICAL *BLOCK_CENTER
is entered. If *VERTICAL *DEPTH_AVE is entered, then *VERTICAL *DEPTH_AVE
*WATER_OIL_GAS *EQUIL is assumed.
CONDITIONS:
This keyword must be the second keyword in the Initial Conditions keyword group.
EXPLANATION:
These keywords indicate how the initial conditions of the reservoir are to be determined.
User's Guide IMEX
*USER_INPUT
Under *USER_INPUT (formerly *VERTICAL *OFF, which is still accepted
by IMEX), pressure and saturations at each grid block must be specified by
the user under the *PRES, *SW, *SO keywords. Gas saturations are then
determined by subtraction.
*VERTICAL *BLOCK_CENTER
Under *VERTICAL *BLOCK_CENTER *WATER_OIL_GAS (formerly
*VERTICAL *ON, which is still accepted by IMEX), the water-oil-contact
depth (*DWOC), the gas- oil-contact depth, together with a reference
pressure (*REFPRES) at a reference depth (*REFDEPTH) must be specified.
Bubble point pressures for the reservoir must also be specified by using
either *PBT or *PB. *API or *APIT must also be specified if the API
tracking model is used (*API-INT).
When using the *GASWATER; *PBT or *PB need not be input.
To use this option to initialize a gas reservoir, set *DGOC = *DWOC or use
*VERTICAL *BLOCK_CENTER *WATER_GAS as outlined below.
To use this option to initialize an under-saturated oil reservoir, set *DGOC to be
above the top of the reservoir. When this is done the results will be identical
from those obtained using *VERTICAL *BLOCK_CENTER *WATER_OIL.
Under *VERTICAL *BLOCK_CENTER *WATER_GAS, the water-gas
contact depth (*DWGC) together with a reference pressure (*REFPRES) at a
reference depth (*REFDEPTH) must be specified. Bubble point pressures for
the reservoir must also be specified by using either *PBT or *PB unless the
*GASWATER or *GASWATER_WITH_CONDENSATE option is used.
When using the *GASWATER or *GASWATER_WITH_CONDENSATE
*MODEL option, *PBT or *PB need not be input. In addition, the
*WATER_GAS initialization option and the *DWGC keyword must be used
to define the gas water contact.
Under *VERTICAL *BLOCK_CENTER *WATER_OIL (formerly
*VERTICAL *ON, which is still accepted by IMEX, with the gas-oil contact
depth set outside the reservoir), the water-oil contact (*DWOC), together
with a reference pressure (*REFPRES) at a reference depth (*REFDEPTH)
must be specified. Bubble point pressures for the reservoir must also be
specified by using either *PBT or *PB. *API or *APIT must also be
specified if the API tracking model is used (*API-INTOW).
*VERTICAL *DEPTH_AVE
Under *VERTICAL *DEPTH_AVE *WATER_OIL, the water-oil contact
depth (*DWOC) together with a reference pressure (*REFPRES) at a reference
depth (*REFDEPTH) must be specified. Bubble point pressures for the
reservoir must also be specified by using either *PBT or *PB. *API or *APIT
must also be specified if the API tracking model is used (*API-INTOW). This
User's Guide IMEX
option is used for undersaturated oil reservoirs in which the added accuracy in
initial amounts-in-place made possible by the depth averaging is wanted.
Under *VERTICAL *DEPTH_AVE *WATER_GAS (*TRANZONE |
*NOTRANZONE), the water-gas contact depth (*DWGC) together with a
reference pressure (*REFPRES) at a reference depth (*REFDEPTH) must be
specified. Bubble point pressures for the reservoir must also be specified by
using either *PBT or *PB, unless the *GASWATER or
*GASWATER_WITH_CONDENSATE options are used. When using the
*GASWATER or *GASWATER_WITH_CONDENSATE *MODEL option,
*PBT or *PB need not be input. In addition, the *WATER_GAS initialization
option and the *DWGC keyword must be used to define the gas water contact.
This option is used for water-gas reservoirs in which the added accuracy in
initial amounts in place made possible by depth averaging of saturations is
wanted.
Under *VERTICAL *DEPTH_AVE *WATER_OIL_GAS, the water-oil
contact (*DWOC) and gas-oil contact (*DGOC) depths, together with a
reference pressure (*REFPRES) at a reference depth (*REFDEPTH) must be
specified. Bubble point pressures for the reservoir must also be specified by
using either *PBT or *PB. *API or *APIT must also be specified if the API
tracking model is used (*API-INT). This option is used for water-oil-gas
reservoirs in which the added accuracy in initial amounts in place made
possible by depth averaging of saturations is wanted.
When using the *GASWATER_WITH_CONDENSATE model, *PDEW or
*PDEWT also must be entered. When using the *VOLATILE_OIL model,
*PDEW or *PDEWT must also be entered.
See the notes in the beginning of this section for more information.
Example:
For a water-gas reservoir with a transition zone and gravity equilibrium, enter:
*VERTICAL *DEPTH_AVE *WATER_GAS *TRANZONE *EQUIL
User's Guide IMEX
Initial Oil Phase Reservoir Pressure (Conditional)
*PRES
PURPOSE:
*PRES indicates the input of reservoir pressures.
(kPa | psi | kPa | kg/cm2)
ARRAY:
*PRES
DEFAULTS:
CONDITIONS:
This keyword must be in the Initial Conditions keyword group. Required with
*USER_INPUT.
EXPLANATION:
The *PRES keyword defines the initial reservoir block pressures (of the oil phase). Any of
the array reading options can be used.
The acceptable range of values for block pressures is:
min
max
User's Guide IMEX
SI
kPa
0.0
1.0E+6
Field
psi
0.0
145,038.0
Lab
kPa
0.0
1.0E+6
Mod. SI
kg/cm2
0.0
1.0E+4
Initial Bubble Point Pressure (Conditional)
*PB
PURPOSE:
*PB indicates the input of bubble point pressure in array format.
ARRAY:
*PB
DEFAULTS:
Required keyword unless using PVT models discussed in CONDITIONS or *PBT. No
default values.
If zero is entered, then the block bubble-point pressure is initialized to the block pressure if
any of the *VERTICAL options are used. If *USER_INPUT initialization is used *PB must
be explicitly set, and input of zero does not default to block pressure.
If the fluid model is *MODEL *OILWATER, a fixed bubble-point pressure is used. The
bubble-point pressure is set to the first pressure value in the *PVT table if *PB is not found.
If multiple PVT regions are specified and the initial bubble point pressure is set with *PB and
*VERTICAL *BLOCK_CENTER is enabled, the bubble point pressure used to calculate
pressures at the woc's and goc's is the value for the grid block closest to the WOC and the
GOC. If bubble point pressure varies as a function of depth, use *PBT instead of *PB.
CONDITIONS:
This keyword must be in the Initial Conditions keyword group and is required unless
*MODEL *OILWATER, *MODEL *GASWATER, *MODEL
*GASWATER_WITH_CONDENSATE or *PBT keywords are used.
If the *GASWATER or *GASWATER_WITH_CONDENSATE models are used, it is not
necessary to define *PB or *PBT.
EXPLANATION:
The *PB keyword defines the initial bubble point pressure of each grid block. Any of the
array reading options may be used.
The acceptable range of values for bubble point pressure is:
min
max
SI
kPa
0.0
1.0E+6
Field
psi
0.0
145,038.0
Lab
kPa
0.0
1.0E+6
Mod. SI
kg/cm2
0.0
1.0E+4
User's Guide IMEX
Initial Bubble Point Pressure vs. Depth (Conditional)
*PBT
PURPOSE:
*PBT indicates the input of bubble point pressure in tabular format as a function of depth.
TABLE:
*PBT set_number
depth
:
Pb
:
DEFINITIONS:
set_number
Set_number for this set of bubble-point pressure vs. depth table. This is the
number used with *PTYPE to assign initial bubble point pressures to grid
blocks. The set_number must correspond to the set_number specified with
*PVT and must increase monotonically.
depth
Depth (m | ft | cm)
Pb
Bubble point pressure (kPa | psi | kPa | kg/cm2)
DEFAULTS:
Required keyword unless using the models discussed in CONDITIONS or *PB. No default
values.
This keyword is to be used only when the bubble-point pressure is a function of depth. If the
bubble-point pressure is constant, use *PB instead.
If the fluid model is *MODEL *OILWATER, a fixed bubble-point pressure is used. The
bubble-point pressure is set to the first pressure value in the *PVT table, if *PBT is not
specified.
CONDITIONS:
Required unless *MODEL *OILWATER, *PB or the two gas-water options are used. This
keyword must be in the Initial Conditions keyword group and must consist of two columns of
numerical entries. The number of *PBT keywords must be equal to the number of PVT
regions defined in the Component Properties section. Each PVT region can have at most one
Pb vs. Depth table.
If the *GASWATER or *GASWATER_WITH_CONDENSATE options are used, it is not
necessary to define *PB or *PBT.
User's Guide IMEX
EXPLANATION:
The *PBT keyword defines the initial bubble point pressure as a function of depth for a PVT
region. Values of bubble point pressure are then assigned to each grid block by use of the
*PTYPE keyword.
Examples:
*PBT 1
** depth Pb
3000.0
1200.0
3200.0
1300.0
The acceptable range of values for bubble point pressure is:
min
max
SI
kPa
0.0
1.0E+6
Field
psi
0.0
145,038.0
Lab
kPa
0.0
1.0E+6
Mod. SI
kg/cm2
0.0
1.0E+4
min
max
SI
m
-1.0E+4
1.0E+4
Field
ft
-32,808.0
32,808.0
Lab
cm
-1.0E+6
1.0E+6
Note that the reservoir depth and the depth for the woc and goc must fall within the depths
specified in the table.
User's Guide IMEX
Initial Dew Point Pressure (Conditional)
*PDEW
PURPOSE:
*PDEW indicates the input of dew point pressure in array format.
ARRAY:
*PDEW
DEFAULTS:
Required keyword only when using *MODEL *GASWATER_WITH_ CONDENSATE or
*MODEL *VOLATILE_OIL, unless the *PDEWT option is used instead. No default values.
If zero is entered, then the block dew point pressure is initialized to the block pressure if any
of the *VERTICAL options are used. If *USER_INPUT initialization is used, *PDEW must
be explicitly set, and input of zero does not default to block pressure.
If multiple PVT regions are specified and the initial dew point pressure is set with *PDEW and
*VERTICAL *BLOCK_CENTER is enabled, the dew point pressure used to calculate
pressures at the gas water contact is the value for the grid block closest to the gas water contact.
If dew point pressure varies as a function of depth, use *PDEWT instead of *PDEW.
CONDITIONS:
This keyword must be in the Initial Conditions keyword group and is required only when
*MODEL *GASWATER_WITH_CONDENSATE or *MODEL *VOLATILE_OIL is used.
*PDEW cannot be used when the *PDEWT option is used.
EXPLANATION:
The *PDEW keyword defines the initial dew point pressure of each grid block. Any of the
array reading options may be used.
The acceptable range of values for dew point pressure is:
min
max
User's Guide IMEX
SI
kPa
0.0
1.0E+6
Field
psi
0.0
145,038.0
Lab
kPa
0.0
1.0E+6
Mod. SI
kg/cm2
0.0
1.0E+4
Initial Dew Point Pressure vs. Depth (Conditional)
*PDEWT
PURPOSE:
*PDEWT indicates the input of dew point pressure in tabular format as a function of depth.
TABLE:
*PDEWT set_number
depth
:
Pdew
:
DEFINITIONS:
set_number
Set_number for this set of dew point pressure vs. depth table. This is the
number used with *PTYPE to assign initial dew point pressures to grid
blocks. The set_number must correspond to the set_number specified with
*PVT and must increase monotonically.
depth
Depth (m | ft | cm)
Pdew
Dew point pressure (kPa | psi | kPa | kg/cm2)
DEFAULTS:
Required keyword when using *MODEL *GASWATER_WITH_CONDENSATE or
*MODEL *VOLATILE_OIL unless the *PDEW option is used instead. No default values.
This keyword to be used only when the dew point pressure is a function of depth. If the dew
point pressure is constant, use *PDEW instead.
CONDITIONS:
This keyword must be in the Initial Conditions keyword group and is required only when
*MODEL *GASWATER_WITH_CONDENSATE or *MODEL *VOLATILE_OIL option is
used. *PDEWT cannot be used when the *PDEW option is used. The number of *PDEWT
keywords must be equal to the number of PVT regions defined in the Component Properties
section. Each PVT region can have at most one Pdew vs. Depth table
EXPLANATION:
The *PDEWT keyword defines the initial dew point pressure as a function of depth for a
PVT region. Values of dew point pressure are then assigned to each grid block by use of the
*PTYPE keyword.
User's Guide IMEX
Examples:
*PDEWT 1
** depth
3000.0
3200.0
Pdew
1200.0
1300.0
The acceptable range of values for dew point pressure is:
min
max
SI
kPa
0.0
1.0E+6
Field
psi
0.0
145,038.0
Lab
kPa
0.0
1.0E+6
Mod. SI
kg/cm2
0.0
1.0E+4
min
max
SI
m
-1.0E+4
1.0E+4
Field
ft
-32,808.0
32,808.0
Lab
cm
-1.0E+6
1.0E+6
Note that the reservoir depth and the depth for the water gas contact must fall within the
depths specified in the table.
User's Guide IMEX
Initial Solvent Bubble Point Pressure (Conditional)
*PBS
PURPOSE:
*PBS indicates input of solvent bubble point pressure (kPa | psi | kPa | kg/cm2).
ARRAY:
*PBS
DEFAULTS:
Conditional keyword. Default is 0.0 for all grid blocks.
CONDITIONS:
This keyword must be in the Initial Conditions keyword group. Used with *MODEL
*MISNCG or *MODEL *MISCG.
EXPLANATION:
The *PBS keyword defines the initial solvent saturation pressure. Any of the array reading
options can be used.
If rss (entered in *PVTS) is approaching zero, that is, there is no solvent in the water phase,
then the values for the solvent bubble point pressures are reset to the block pressures.
The acceptable range of values for solvent bubble point pressure is:
min
max
SI
kPa
0.0
1.0E+6
Field
psi
0.0
145,038.0
Lab
kPa
0.0
1.0E+6
Mod. SI
kg/cm2
0.0
1.0E+4
User's Guide IMEX
Initial Oil Saturation (Conditional)
*SO
PURPOSE:
*SO indicates input of initial oil saturation (fraction).
ARRAY:
*SO
DEFAULTS:
CONDITIONS:
This keyword must be in the Initial Conditions keyword group and is required with the
*USER_INPUT initialization option except when the *MODEL *GASWATER fluid model
is used.
EXPLANATION:
The *SO keyword defines the initial oil saturation at each block. Any of the array reading
User's Guide IMEX
Initial Water Saturation (Conditional)
*SW
PURPOSE:
*SW indicates input of initial water saturation (fraction).
ARRAY:
*SW
DEFAULTS:
CONDITIONS:
This keyword must be in the Initial Conditions keyword group. Required with
*USER_INPUT initialization option except when the *MODEL *OILWATER fluid model is
used.
EXPLANATION:
The *SW keyword defines the initial water saturation at each block. Any of the array reading
User's Guide IMEX
Initial Polymer Concentration (Conditional)
*POLYCONC
PURPOSE:
*POLYCONC indicates input of initial polymer concentration (kg/m3 | lb/bbl | g/cm3).
ARRAY:
*POLYCONC
DEFAULTS:
Conditional keyword. If the *POLYCONC keyword is missing then the initial polymer
concentration in the reservoir is set to zero.
CONDITIONS:
This keyword must be in the Initial Conditions keyword group. Required when the polymer
option is being used and nonzero initial polymer concentrations are required.
EXPLANATION:
The *POLYCONC keyword defines the initial polymer concentration at each block. Any of
the array reading options can be used.
The acceptable range of values for polymer concentration is:
min
max
User's Guide IMEX
SI
Kg/m3
0.0
29.0
Field
Lb/stb
0.0
10.0
Lab
g/cm3
0.0
0.029
Mod. SI
Kg/m3
0.0
29.0
Initial Seawater Volume Fraction (Conditional)
*SEAWATFRC
PURPOSE:
*SEAWATFRC indicates input of initial seawater volume fraction (fraction).
ARRAY:
*SEAWATFRC
DEFAULTS:
Conditional keyword. If the *SEAWATFRC keyword is missing then the initial seawater
fraction in the reservoir is set to zero.
CONDITIONS:
This keyword must be in the Initial Conditions keyword group. Used when the seawater
injection option is being used and nonzero initial seawater volume fractions are required.
EXPLANATION:
The *SEAWATFRC keyword defines the initial seawater volume fraction at each block.
Any of the array reading options can be used.
The acceptable range of values for seawater volume fraction is:
min
max
SI
Fraction
0.0
1.0
Field
Fraction
0.0
1.0
Lab
Fraction
0.0
1.0
Mod. SI
Fraction
0.0
1.0
User's Guide IMEX
Initial Light Oil Volume Fraction (Conditional)
*API
PURPOSE:
*API indicates input of initial light oil volume fraction at surface conditions (fraction).
ARRAY:
*API
DEFAULTS:
CONDITIONS:
This keyword must be in the Initial Conditions keyword group and is Required if *MODEL
*API-INT or *API-INTOW is used and *APIT is not used.
EXPLANATION:
The *API keyword defines the initial light oil volume fraction at surface conditions at each
block. Any of the array reading options can be used.
User's Guide IMEX
Initial Light Oil API Volume Fraction vs. Depth (Conditional)

*APIT
PURPOSE:
*APIT indicates the input of Initial light oil volume fraction (Surface Conditions) in tabular
format as a function of depth (fraction).
TABLE:
*APIT set_number
depth
:
API
:
DEFINITIONS:
set_number
Set_number for this set of light oil volume fractions vs. depth table. This is
the number used with *PTYPE to assign PVT Regions to grid blocks. The
set_number must correspond to the set_number specified with *PVTAPI and
must increase monotonically.
depth
Depth (m | ft | cm).
API
Initial Light Oil Volume Fraction (fraction).
DEFAULTS:
CONDITIONS:
This keyword must be in the Initial Conditions keyword group and is Required if *MODEL
*API-INT or *API-INTOW is used and *API is not used.
This keyword must be in the Initial Conditions keyword group and must consist of two columns
of numerical entries. The number of *APIT keywords must be equal to the number of PVT
regions defined in the Component Properties section. Each PVT region can have at most one
API vs. Depth table. Light oil volume fraction is limited to a range between 0.0 and 1.0.
EXPLANATION:
The *APIT keyword defines initial light oil volume fraction as a function of depth for a PVT
region. Values of light oil volume fraction are then assigned to each grid block by use of the
*PTYPE keyword.
User's Guide IMEX
Examples:
*APIT 1
** depth
3000.0
3200.0
Light oil volume fraction

0.8
0.6
Note that the reservoir depth and the depth for the woc and goc must fall within the depths
specified in the table.
User's Guide IMEX
Reference Depth and Reference Pressure (Conditional)

*REFDEPTH, *REFPRES
PURPOSE:
*REFDEPTH indicates input of reference depth.
*REFPRES indicates input of reference pressure.
FORMAT:
*REFDEPTH depth
*REFPRES pressure
DEFINITIONS:
depth
Depth of reference pressure (*REFPRES) (m | ft | cm). If the value is
negative, then *REFDEPTH is interpreted as an altitude above some datum.
For accurate initialization, the reference depth should be within the reservoir.
pressure
The reference pressure (kPa | psi | kPa | kg/cm2) at the reference depth
(*REFDEPTH) is used to set reservoir block pressures.
DEFAULTS:
CONDITIONS:
This keyword must be in the Initial Conditions keyword group. Required with all
*VERTICAL subkeywords.
EXPLANATION:
The reference pressure at the reference depth are used in the vertical equilibrium option,
along with the capillary pressures, fluid densities, water-oil contact and the gas-oil contact
depths, to calculate the block pressures and saturations.
If multiple PVT regions have been specified, then a corresponding number of reference
depths and pressures must also be specified. If the number of reference depths and pressures
is less than the number of PVT regions then the last entered values will be assigned to the
remaining PVT regions.
Each PVT region can have at most one reference depth and pressure. Multiple reference
depths and pressures can be entered by specifying a corresponding number of PVT regions
with identical PVT data for each region.
User's Guide IMEX
For multiple PVT regions *REFDEPTH and *REFPRES can be entered in either of two ways
1. Separate keywords for each region:
Example:
*REFPRES Pressure for Region 1

*REFPRES Pressure for Region 2
2. Multiple regions on one keyword:

Example:
*REFPRES Pressure for Region 1 Pressure for Region 2
The acceptable range of values for reference depth is:
min
max
SI
m
-1.0E+4
1.0E+4
Field
ft
-32,808.0
32,808.0
Lab
cm
-1.0E+6
1.0E+6
min
max
SI
kPa
0.0
1.0E+6
Field
psi
0.0
145,038.0
Lab
kPa
0.0
1.0E+6
Mod. SI
kg/cm2
0.0
1.0E+4
If using the *PBT keyword, the reference depth must be within the range of depth values
entered in the bubble point pressure vs. depth table.
User's Guide IMEX
Depth to Water-Oil Contact (Conditional)
*DWOC, *DGOC, *DWGC
PURPOSE:
*DWOC indicates input of water-oil contact(s).
*DGOC indicates input of gas-oil contact depth.
*DWGC indicates input of water-gas contact depth.
FORMAT:
*DWOC woc_depth
*DGOC goc_depth
*DWGC wgc_depth
DEFINITIONS:
woc_depth
Depth (m | ft | cm) to water-oil contact plane. If the value is negative, then
*DWOC is interpreted as an altitude above some datum.
goc_depth
Depth (m | ft | cm) to gas-oil contact plane. If the value is negative, then
*DGOC is interpreted as an altitude above some datum.
wgc_depth
Depth (m | ft | cm) to water-gas contact plane. If the value is negative, then
*DWGC is interpreted as an altitude above some datum.
DEFAULTS:
CONDITIONS:
This keyword must be in the Initial Conditions keyword group. The use of the *DWGC
keyword is not allowed in a *WATER_OIL_GAS initialization. Use equal *DWOC and
*DGOC values instead.
*DWOC is required with:
*DGOC is required with:
*DWGC is required with:
*WATER_OIL
*WATER_OIL_GAS
*WATER_OIL
*WATER_OIL_GAS
*WATER_OIL_GAS
*WATER_OIL_GAS
*WATER_GAS
*WATER_GAS
User's Guide IMEX
EXPLANATION:
The *DWOC keyword defines the depth to the water-oil plane. If there is no water-oil
contact set depth to a value greater than the depth of the bottom of the reservoir.
The *DGOC keyword defines the depth to the gas-oil contact plane. If there is no gas-oil
contact set depth to a value less than the depth of the top of the reservoir.
The *DWGC keyword defines the depth to the water-gas contact in a reservoir in which no
oil is present. If there is no water-gas contact set depth to a value greater than the depth of
the bottom of the reservoir.
The definition for the term 'contact depth' as used in the *BLOCK_CENTER options differs
slightly from the definition used in the *DEPTH_AVE options. This is unfortunate, but was
necessary in order that the capillary pressure vs. depth tables could be constructed for the
*DEPTH_AVE options.
For the *BLOCK_CENTER options, the water-oil contact depth is defined as the smallest
depth at which the water saturation assumes its maximum value (unity or nearly unity). This
definition is consistent with the one from Slider, H.C., "Practical Petroleum Reservoir
Engineering Methods", Chapter 5, 1976, PennWell Books, Tulsa, pp. 266.
For the *DEPTH_AVE options, the water-oil contact depth is defined as the depth at which the
water-oil capillary pressure assumes a prescribed value. The default value for this capillary
pressure is zero, but the user can set a nonzero value for the water- oil capillary pressure at the
water-oil contact in the *DEPTH_AVE options by using the *WOC_PC keyword.
For the *BLOCK_CENTER options, the gas-oil contact depth is defined as the smallest depth
at which the gas saturation has the value zero. This definition is consistent with the one from
Slider, H.C., "Practical Petroleum Reservoir Engineering Methods", Chapter 5, 1976,
PennWell Books, Tulsa, pp. 266.
For the *DEPTH_AVE options, the gas-oil contact depth is defined as the depth at which the
oil-gas capillary pressure assumes a prescribed value. The default value for this capillary
pressure is zero, but the user can set a nonzero value for the oil-gas capillary pressure at the
gas-oil contact in the *DEPTH_AVE options by using the *GOC_PC keyword.
Under the *BLOCK_CENTER initialization options, the gas-oil transition zone is ABOVE
the gas-oil contact, and the water-oil transition zone is above the water-oil contact. In the
*DEPTH_AVE options, some gas may lie below the gas-oil contact and some oil may lie
below the water-oil contact, depending on the capillary pressure values chosen at the contacts
(see the explanations with the *WOC_PC and *GOC_PC keywords.).
depth
GOC
Gas zone
Gas-oil transition zone
Oil zone
WOC
User's Guide IMEX
Water-oil transition zone

Water zone
If the number of water-oil, gas-oil or gas-water contacts is less than the number of PVT
regions then the last entered value will be assigned to the remaining PVT regions.
Each PVT region can have at most one water-oil, gas-oil and water-gas contact only.
Multiple contacts can be entered by specifying a corresponding number of PVT regions with
identical PVT data for each region.
For multiple PVT regions *DWOC *DGOC and *DWGC can be entered in either of two ways
Example:
*DWOC contact for Region 1

*DWOC contact for Region 2

Example:
*DWOC contact for Region 1 contact for Region 2
The value for depths entered must fall within the range of depths entered with *PBT (if
bubble point pressure is a function of depth).
The acceptable range of values for depth of contacts is:
min
max
SI
m
-1.0E+4
1.0E+4
Field
ft
-32,808.0
32,808.0
Lab
cm
-1.0E+6
1.0E+6
User's Guide IMEX
Datum Depth Specification (Optional)
*DATUMDEPTH
PURPOSE:
*DATUMDEPTH specifies the datum depth for pressure printout corrected to datum for each
PVT region. A *DATUMDEPTH keyword should be entered for each PVT region. The first
*DATUMDEPTH applies to PVT region 1, the second to PVT region 2 and so on.
FORMAT:
*DATUMDEPTH
depth
(*INITIAL)
(*REFDENSITY density)
(*REFDENSITY *GRIDBLOCK)
DEFINITIONS:
depth
A real number denoting the depth to the datum (m | ft).
*INITIAL
Keyword indicating that the initial equilibrium pressure distribution in the
reservoir will be used to calculate the corrected datum pressures. This is the
default.
*REFDENSITY
Keyword indicating that the entered density should be used to calculate the
corrected datum pressures.
density
A real number denoting the density to be used for calculating the corrected
pressures (Kg/m3 | lb/ft3). This entered density will be used to correct for
the gravity head for every grid block (in the gas cap; in the oil zone; and in
the water zone).
*GRIDBLOCK
Keyword specifying that the grid block density should be used to calculate
the datum pressures.
DEFAULTS:
Optional keyword. If this keyword is not present and the output of datum pressures is
requested then no datum corrected pressure printout will appear. If *DATUMDEPTH is
present without *INITIAL or *REFDENSITY then the default is *INITIAL.
If *REFDENSITY is specified without density or *GRIDBLOCK then the default is
*GRIDBLOCK.
User's Guide IMEX
If one *DATUMDEPTH keyword is found in a model with multiple PVT regions, the single
datum depth specification will be applied to all PVT regions. If fewer *DATUMDEPTH
keywords than PVT regions are found (but more than a single *DATUMDEPTH is specified)
all remaining PVT regions will default to the last specified *DATUMDEPTH and a warning
will be issued.
CONDITIONS:
This keyword if present must be in the Initial Conditions keyword group.
EXPLANATION:
*DATUMDEPTH modifies the individual grid block pressures to a common depth by
removing the gravity head from the pressures. This removal of the gravity head can be
accomplished in three ways:
1. If *INITIAL is used (this is the default) then the pressures are corrected based on
the initial equilibrium pressure distribution of the reservoir.
datum_pressure(ib) = pressure(ib) - pressure_0(ib) + datum_pressure_0
Here: datum_pressure(ib) is the corrected datum pressure for grid block ib.
pressure(ib) is the actual pressure at grid-block ib.
pressure_0(ib) is the grid block pressure at initial time (time = 0).
datum_pressure_0 is the datum pressure at initial time.
Please note that the above calculation assumes that:
a) The phase densities do not change by a large amount during the simulation.
b) The water-oil and gas-oil contacts do not move by a large amount.
The above assumptions are valid for most black-oil simulation studies.
2. If *REFDENSITY density is used then the pressures are corrected to datum as follows.
datum_pressure(ib) = pressure(ib) + density*g*(datumdepth - depth(ib))
Here: density is the density specified using the *REFDENSITY keyword above.
depth(ib) is the depth of grid block ib.
g is the acceleration due to gravity.
3. If *REFDENSITY *GRIDBLOCK is used then the pressures are corrected to
datum as follows.
datum_pressure(ib) = pressure(ib) + density(ib)*g*(datumdepth - depth(ib))
Here: density(ib) is the oil phase density of grid block ib if the oil phase is present
in the block. If the oil phase is not present in that grid block then it is the gas
phase density of grid block ib if the gas phase is present. If neither the oil nor the
gas phases are present then density(ib) is the water density of grid block ib.
When *DATUMDEPTH is specified for a multiple PVT region model, each PVT
region can use a different *DATUMDEPTH calculation method and/or use
different method parameters.
User's Guide IMEX
Examples:
1.
2.
3.
4.
5.
*DATUMDEPTH
*DATUMDEPTH
*DATUMDEPTH
*DATUMDEPTH
*DATUMDEPTH
1000.0
1000.0
1000.0
1000.0
1000.0
*INITIAL
*REFDENSITY
*REFDENSITY *GRIDBLOCK
*REFDENSITY 871.0
Examples 1 and 2 above are equivalent.

Examples 3 and 4 above are equivalent.
The above 5 examples could also define the datum depths for a single run with 5 PVT
regions.
User's Guide IMEX
Specification of Capillary Pressure Values at Contacts

(Optional)
*GOC_PC, *WOC_PC
PURPOSE:
*GOC_PC AND *WOC_PC introduce the value of the oil-gas capillary pressure at the gasoil contact and the value of the water-oil capillary pressure at the water-oil contact,
respectively. These values are used in the *VERTICAL *DEPTH_AVE initialization
options. Values of *GOC_PC AND *WOC_PC can be entered for each PVT region.
FORMAT:
*GOC_PC pcggoc
*WOC_PC pcwwoc
DEFINITIONS:
pcggoc
A real number specifying the value of the oil-gas capillary pressure at the
gas-oil contact (kPa | psi | kPa | kg/cm2) for a PVT region.
pcwwoc
A real number specifying the value of the water-oil capillary pressure at the
water-oil contact (kPa | psi | kPa | kg/cm2) for a PVT region.
DEFAULTS:
Optional keywords. If this keyword is present then the *USER_INPUT and all *VERTICAL
*BLOCK_CENTER initialization options are unaffected because pcggoc and pcwwoc do not
enter these initializations. For the *VERTICAL *DEPTH_AVE options the default values
are pcggoc = 0 and pcwwoc = 0.
OILWET Option:
If the OILWET option is used, pcwwoc is defaulted to the value of the largest capillary
pressure in all of the *SWT tables in the PVT region. This is normally the value at the
minimum wetting phase (oil) saturation in each table. The user can always explicitly input a
desired value to override this default. Please see the Tutorial section for details.
CONDITIONS:
These keywords, if present, must be in the Initial Conditions keyword group. No error
message is printed if a *VERTICAL *DEPTH_AVE option is chosen but *GOC_PC or
*WOC_PC does not appear in the data; the values of pcggoc or pcwwoc are defaulted and the
initialization proceeds.
User's Guide IMEX
EXPLANATION:
Under the *DEPTH_AVE initialization options, the water-oil contact is defined as the depth at
which the water-oil capillary pressure has the value pcwwoc. Normally this value is zero, which
is the default value in IMEX, but the *WOC_PC keyword allows the user to set any desired value
for the water-oil capillary pressure at the water-oil contact depth. Under the *BLOCK_CENTER
initialization options the water-oil contact depth is defined by saturation values rather than by
capillary pressure values and pcwwoc does not enter the computation, with capillary pressures
determined from the established values of the saturations. If *WOC_PC is entered when a
*BLOCK_CENTER option is invoked, the entered pcwwoc value is ignored. Similarly,
*GOC_PC allows the user to specify a value for pcggoc, the value of the oil-gas capillary
pressure at the gas-oil contact. Again the default value is zero. Under the *DEPTH_AVE
initialization options, the gas-oil contact depth is defined as the depth at which the oil-gas
capillary pressure has the value pcggoc. This keyword is ignored under the *USER_INPUT
and the *VERTICAL *BLOCK_CENTER options.
Examples:
*GOC_PC -0.5
*WOC_PC 10.0
The acceptable range of values for *GOC_PC is between the minimum and maximum values
of gas- oil capillary pressure from gas-liquid relative- permeability tables in the PVT region it
applies to.
The acceptable range of values for *WOC_PC is between the minimum and maximum values
of water- oil capillary pressure from water-oil relative- permeability tables in the PVT region
it applies to.
If the *WOC_PC for a PVT region is less than the minimum value of water oil capillary
pressures in the PVT region, a warning message will be issued. The run will continue to
initialize using the following logic:
Below the WOC: Water saturation is set to 1.0
From the WOC to the height above the water oil contact where the head
difference between reservoir condition water and oil equals the minimum
water oil capillary pressure in the relative permeability table which
applies to a particular grid block minus the *WOC_PC, the water
saturation is set to 1.0 Soirw (oil saturation = Soirw).
Moving up from the height above the water oil contact where the head
difference between oil and water equals the minimum water oil capillary
pressure in the applicable relative permeability table minus the *WOC_PC,
the water and oil saturation in the transition zone is determined from the
water oil capillary pressure curve, in the standard fashion.
A similar logic is applied if the *GOC_PC is less than the minimum value o gas oil capillary
pressure in a PVT region.
If one *WOC_PC or *GOC_PC keyword is found then this value will be assigned to all PVT
regions. If there are fewer values entered than PVT regions (but more than one) then an error
will be raised.
User's Guide IMEX
Each PVT region can have at most one *WOC_PC or *GOC_PC. For multiple PVT regions
*WOC_PC or *GOC_PC can be entered in either of two ways
Example:
*WOC_PC pcwwoc for Region 1

*WOC_PC pcwwoc for Region 2

Example:
* WOC_PC pcwwoc for Region 1 pcwwoc for Region 2
User's Guide IMEX
Specification of Method for Computing Oil Saturations in Gas

Zone
*GASZONE
PURPOSE:
*GASZONE specifies whether a residual oil saturation is to be set initially in the uppermost
part of the reservoir where the difference in the oil and gas phase pressures exceed the largest
tabulated value the oil-gas capillary pressure. This keyword is exactly equivalent to the
previous *GASCAP, which is still accepted by IMEX.
FORMAT:
*GASZONE
(*NOOIL)
(*OIL)
DEFINITIONS:
*OIL
Indicates that the oil/gas residual saturation Sorg is to be set initially in the
uppermost regions of the reservoir where the computed difference in the
pressures of the oil and gas phases exceeds the largest value which appears in
the oil/gas capillary pressure table.
*NOOIL
Indicates that an initial oil saturation of zero is to be assigned in the
uppermost regions of the reservoir where the computed difference in the
pressures of the oil and gas phases exceeds the largest value which appears in
the oil/gas capillary pressure table.
DEFAULTS:
Optional keywords. If *GASZONE is not present then *NOOIL is assumed and oil
saturations of zero are assigned initially in the gas zone.
CONDITIONS:
This keyword, if present, must be in the Initial Conditions keyword group. It applies for all of the
*VERTICAL initialization options. If *GASZONE is entered when the *USER_INPUT
initialization option is specified, then *GASZONE is ignored and no message is issued.
EXPLANATION:
Normally in black-oil simulation, reservoirs are initialized such that the primary gas-cap does
not have any residual oil. However if a nonzero oil saturation is desired in the gas-cap at
initial time then the *GASZONE *OIL option should be used. The oil saturation in the gas
cap is set to the residual oil saturation derived from the liquid-gas relative-permeability table.
This is normally calculated as the residual liquid saturation in the gas-liquid table minus the
connate water saturation. When the *NOSWC option is used, the residual oil saturation is
equal to the residual liquid saturation.
Example:
*GASZONE
User's Guide IMEX
*OIL
Specification of Water Saturation Above GOC or Below WOC

(Optional)
*GOC_SW, *WOC_SW
PURPOSE:
*GOC_SW and *WOC_SW allow the user to define the water saturation above the gas oil
contact and below the water-oil contact for each PVT Region.
FORMAT:
*GOC_SW
*WOC_SW
sw_gascap
sw_waterzone
DEFINITIONS:
sw_gascap
A real number specifying the value of the water saturation above the gas oil
contact for a PVT Region (Fraction).
sw_waterzone
A real number specifying the value of the water saturation below the wateroil contact for a PVT Region (Fraction).
DEFAULTS:
Optional keywords. If this keyword is present and any of the *BLOCK_CENTER initializations
is being used, the values will be accepted. WOC_SW is also accepted for DEPTH_AVE
initialization. If *USER_UNPUT initialization is being used a warning will be issued and the
input values will be ignored.
CONDITIONS:
These keywords, if present, must be in the Initial Conditions keyword group. One GOC_SW
(WOC_SW) card must appear for each PVT Region. The first appearance assigns a value to
the first PVT region. The next appearance of GOC_SW (WOC_SW) assigns a value to the
next PVT Region (and so on). One card must appear for each PVT Region defined.
*WOC_SW can be used with *BLOCK_CENTER and *DEPTH_AVE initialization options.
*GOC_SW can only be used with the *BLOCK_CENTER initialization options.
*GOC_SW is not to be used with the *GASWATER or *GASWATER_WITH_CONDENSATE
options.
Care must be exercised when using this option as it is possible to create mobile oil in the
water zone or mobile water in the gas cap by poor choices of input. In general, *GOC_SW
should be less than the PVT region's critical water saturation and *WOC_SW should be
greater than one minus the PVT region's residual oil saturation (Sorw).
User's Guide IMEX
EXPLANATION:
By default IMEX defines the initial water saturation of each block in a Gas Cap to the block's
connate water saturation. This value is either defined by the relative permeability table assigned
to each block in the Gas Cap or by the value explicitly defined on the *SWCON card. Also, by
default, IMEX assigns the initial water saturation of each block in the water zone to 1.0.
The *GOC_SW and WOC_SW keywords allow the user to easily override these defaults for
each PVT region. *GOC_SW and *WOC_SW will only alter water saturation in the gas cap
and water zone respectively. They allow the user to set initial water saturation to a single
value above (or below) the oil zone.
If defined, one *GOC_SW (or *WOC_SW) must appear for each PVT region. The order of
input defines which PVT Region water saturation is assigned to. The first card assigns
*GOC_SW or *WOC_SW to the first PVT Region, the last card assigns *GOC_SW or
*WOC_SW to the last PVT Region.
*GOC_SW and *WOC_SW override all values calculated in the gas cap or water zone, even
those defined on the *SWINIT card.
For multiple PVT regions *GOC_SW and *WOC_SW can be entered in either of two ways
Example:
*GOC_SW Saturation for Region 1

*GOC_SW Saturation for Region 2

Example:
*GOC_SW Saturation for Region 1 Saturation for Region 2
Examples:
*GOC_SW 0.20
*WOC_SW 0.25
User's Guide IMEX
Water Saturation Used with Block_Center Initialization

(Conditional)
*SWINIT
PURPOSE:
*SWINIT indicates input of initial water saturation to be used with *BLOCK_CENTER
gravity-capillary equilibrium initialization. *SWINIT values can be different from connate
saturation (fraction).
ARRAY:
*SWINIT
DEFAULTS:
CONDITIONS:
This keyword must be in the Initial Conditions keyword group. It must be used only with the
*BLOCK_CENTER initialization options.
Care has to be taken when this option is employed, reasonable values of saturation must be
entered, as highly altered Pcow curves will significantly effect other aspects of the model.
The user must be using rock curves with oil-water capillary pressure for this option to have
the desired effect, as the *SWINIT keyword scales Pcow to establish vertical equilibrium at
*SWINIT.
Various limitations exist on how *SWINIT can modify vertical equilibrium, these are listed
below:
In the Gas Cap, *SWINIT can vary between 0.0 and the critical (not connate) water
saturation. If *SWINIT greater than critical water saturation is found, Sw in that block will
be set equal to the critical water saturation. The *GOC_SW overrides the *SWINIT and can
be used to modify gas zone Sw further.
In the oil zone, *SWINIT must be greater than the connate water saturation. If *SWINIT less
than connate is found, Sw in the block will be set equal to connate, Pcow will be scaled but
the block will not be in equilibrium.
If, in an oil zone block, original Pcow or scaled Pcow is less than 0.1 kPa (0.0145 Psi) no
scaling will be done and a warning will be issued.
If Pcow, in an oil zone block, must be scaled by a factor greater than 200 or by a factor less
than 0.005 to be in equilibrium, the factor will be limited to 200 or 0.005 and a warning will
be issued. Pcow will be scaled by the reduced factor, but the fluids will not be in equilibrium.
In the water zone, *SWINIT values are ignored. The value of water saturation in the water
zone is normally 1.0. The *WOC_SW keyword overrides the *SWINIT keyword and can be
used to modify water zone water saturation.
User's Guide IMEX
In summary
Gas Cap:
0 < Swinit < Sw(critical)
Sw reset to Sw(critical) if > Sw(critical)
Oil Zone:
Sw reset to Sw(connate) if Sw < Sw(connate)
Pcow scaled only if both original Pcow and scaled Pcow > 0.1 kPa
Pcow scaling is limited by a maximum scaling factor of 200 and a minimum scaling
factor of 0.005. Scaling will occur with the maximum or minimum factors but the
grid blocks will not be in equilibrium.
Water Zone:
Swinit entry is ignored
EXPLANATION:
The *SWINIT keyword is used with the *BLOCK_CENTER equilibrium. It allows the user
to input all of the required keywords for *BLOCK_CENTER equilibrium option as well as
initial water sat.
This initial water saturation is used to scale the Pcow curve in such a way so that the initial
water saturation values are honored when the *BLOCK_CENTER equilibrium initialization
is solved for.
User's Guide IMEX
Non Equilibrium Water and Oil Saturation Used with GravityCapillary Equilibrium Initialization (Conditional)
*SWNEQ,
*SONEQ
PURPOSE:
*SWNEQ and/or *SONEQ indicates input of non equilibrium initial saturation overrides to
be used with gravity-capillary equilibrium initialization.
ARRAY:
*SWNEQ
*SONEQ
DEFAULTS:
CONDITIONS:
This keyword must be in the Initial Conditions keyword group.
*SWNEQ and *SONEQ need not be defined for every block in the reservoir, just the blocks
where the user requires that the initial saturations obtained from the gravity-capillary
initialization are overridden. In models where both oil and gas phases could initially exist,
both *SWNEQ and *SONEQ need to be defined in each block the user overrides.
If oil saturation is never present (model *GASWATER) or not input initially (model
*GASWATER_WITH_CONDENSATE), then *SWNEQ alone is entered. *SONEQ is
internally set to zero.
If the gas phase is not modeled, in a model like model *OILWATER, then *SONEQ alone is
entered. *SWNEQ is internally set to (1.0 *SONEQ).
EXPLANATION:
The *SWNEQ/*SONEQ keywords allow the user to override the equilibrium saturations
obtained from gravity-capillary initialization in a portion of the reservoir. An example of its
use would be to initialize the saturations in and near a water filled fracture, created during a
well stimulation procedure which is not to be explicitly modeled. The entire reservoir would
be initialized using equilibrium calculations while a user defined area surrounding the
fractured well might be given very high water saturation. The pressures in the overridden
blocks are still obtained from the equilibrium initialization.
The *IJK reading option is used to precisely define the region of saturation override. Areas
not to be overridden must not be defined in the SONEQ/SWNEQ *IJK range.
The *ALL array input option maybe used with *SONEQ and *SWNEQ. Values of
*SONEQ and *SWNEQ which are not to be overridden, but which must be included due to
the *ALL input format, must be given the value of -1.0.
User's Guide IMEX
EXAMPLE:
*SWNEQ *IJK
4:28 14:18 4:6
5:27 15:17 4:6
6:26 16:16 5:5
0.8
0.9
0.99
*SONEQ *IJK
4:28 14:18 4:6
5:27 15:17 4:6
6:26 16:16 5:5
0.2
0.1
0.01
User's Guide IMEX
Numerical Methods Control Identifier (Optional)
*NUMERICAL
PURPOSE:
*NUMERICAL identifies the beginning of all numerical methods control keywords.
FORMAT:
*NUMERICAL
DEFAULTS:
Optional keyword. It is required if you wish to override the defaults for one or more of the
keywords in this group.
CONDITIONS:
The Numerical Methods Control keyword group follows the Initial Conditions keyword
group in the data file.
EXPLANATION:
The defaults used in the numerical solution techniques provide a robust and efficient solution
to most simulation problems. You should override the defaults only if you have a very good
understanding of the solution methods involved. Inappropriate overriding of the defaults may
result in the use of much more CPU time than would otherwise be required for a problem.
For detailed explanations of the matrix solution controlling keywords (*NORTH, *SORDER,
*PIVOT, *SDEGREE, *RELTOL, *ITERMAX), please refer to the AIMSOL Technical
Manual.
User's Guide IMEX
Numerical Methods Control 621
Maximum Timesteps (Optional)
*MAXSTEPS
PURPOSE:
*MAXSTEPS is used to specify the maximum number of timesteps for the simulation run.
FORMAT:
*MAXSTEPS num
DEFINITIONS:
num
An integer to specify the maximum number of time steps allowed.
DEFAULTS:
Optional keyword. If it is not present in the input-data-file, the default is that the number of
time steps for the simulation is not explicitly limited.
CONDITIONS:
This keyword must be located in the Numerical Methods Control keyword group. If the
command line options "-onestep" or "-maxsteps n" are used they will override this keyword.
If both "-onestep" and "-maxsteps n" are specified then "-onestep" will be used.
EXPLANATION:
Simulation will be terminated when the number of timesteps is equal to the number specified
by *MAXSTEPS.
Examples:
** Limit the number of timesteps to 100
*MAXSTEPS 100
622 Numerical Methods Control
User's Guide IMEX
Maximum CPU Seconds (Optional)
*MAXCPU
PURPOSE:
*MAXCPU is used to specify the maximum CPU seconds for the simulation run.
FORMAT:
*MAXCPU maxcpu
DEFINITIONS:
maxcpu
A number to specify the maximum CPU time in seconds allowed for this
simulation run.
DEFAULTS:
Optional keyword. If it is not present in the input-data-file, the default is that CPU time for
the simulation is not explicitly limited.
CONDITIONS:
This keyword must be located in the Numerical Methods Control keyword group.
EXPLANATION:
Simulation will be terminated when the CPU time exceeds the number specified by
*MAXCPU.
This allows the program to terminate gracefully, having written output and restart records
before running out of CPU time.
This option requires the user to supply a subroutine called GETCPU which makes a system
call and returns with the current CPU time from the system. Please see the installation
instructions and notes section of the User's manual.
Examples:
** Limit the CPU time to one hour.
*MAXCPU 3600.0
User's Guide IMEX
Maximum Timestep Size (Optional)
*DTMAX
PURPOSE:
*DTMAX identifies the maximum time-step size.
FORMAT:
*DTMAX max_time_size
DEFINITIONS:
max_time_size
A real number to specify the maximum time-step size allowed (day | day |
min).
DEFAULTS:
Optional keyword. If it is not present in the input-data-file, the maximum time step size is
365 days.
CONDITIONS:
This keyword may be located in the Numerical Methods Control keyword group, and may
also be part of recurrent data. When restarting a run, the *DTMAX keyword must be in
recurrent data, after the restart time, to change from the value in the restarted run.
EXPLANATION:
If the time-step size calculated by the automatic time-step selector is larger than
max_time_size, it is set to max_time_size.
The time-step size is always automatically adjusted so that a timestep will end exactly at the
time specified by a *TIME or *DATE keyword in recurrent data. The maximum time-step
size has no real preset minimum or maximum. But to avoid computer round-off problems a
min. of 1.0E-10 days and a max. of 1.0E+20 days has been imposed.
Examples:
** Limit the maximum time-step size to half a day
*DTMAX 0.5
User's Guide IMEX
Minimum Timestep Size (Optional)
*DTMIN
PURPOSE:
*DTMIN identifies the minimum time-step size.
FORMAT:
*DTMIN min_time_size
DEFINITIONS:
min_time_size
A real number to specify the minimum time-step size allowed (day | day |
min).
DEFAULTS:
Optional keyword. If it is not present in the input-data-file, the minimum time-step size is
0.001 days.
CONDITIONS:
This keyword may be located in the Numerical Methods Control keyword group, and may
also be part of recurrent data. The minimum timestep size should be set to a reasonable
value. Do not simply reduce the minimum time-step to unrealistic sizes to try to converge.
*DTMIN must be smaller than *DTWELL.
EXPLANATION:
If the automatic time-step selector cuts the time step size to smaller than min_time_size
specified by *DTMIN, the time-step size is set to min_time_size. If the timestep does not
converge, the simulation is stopped. The maximum time-step size has no real preset
minimum or maximum. But to avoid computer round-off problems a min. of 1.0E-10 days
and a max. of 1.0E+20 days has been imposed.
Examples:
** Limit the minimum time-step size to 6 hours
*DTMIN 0.25
User's Guide IMEX
Normal Variation in Variables per Timestep (Optional)
*NORM,
*MAXCHANGE
PURPOSE:
*NORM identifies the typical changes in the basic variables over a timestep.
*MAXCHANGE identifies whether a time step is cut if the maximum change in pressure or
saturation is greater than that specified by *MAXCHANGE.
FORMAT:
*NORM
(*PRESS)
(*SATUR)
(*PBUB)
(*PDW)
value
*MAXCHANGE
(*PRESS)
(*SATUR)
(*PBUB)
(*PDW)
value
DEFINITIONS:
*PRESS
This subkeyword identifies pressure (kPa | psi | kPa | kg/cm2).
*SATUR
This subkeyword identifies saturations (fraction, dimensionless).
*PBUB
This subkeyword identifies bubble point pressure (kPa | psi | kPa | kg/cm2).
*PDW
This subkeyword identifies dew point pressure (kPa | psi | kPa | kg/cm2)
(*GASWATER_WITH_CONDENSATE and *VOLATILE_OIL model
only).
value
A real number to specify the controlled value.
DEFAULTS:
Optional keyword. If *NORM is not present in the input-data-file the defaults are:
Pressure Saturation Bubble Point Pressure Dew Point Pressure -
3000.0 kPa, 435.0 psia or 30.0 kg/cm2 for pressure or

0.1 for oil, water, and gas saturation.
defaults to the same value as pressure
10,000 kpa, 1450 psi or 100 kg/cm2
User's Guide IMEX
For all Models, except *MISCG and *MISNCG, the defaults for *MAXCHANGE are two
times the *NORM values (i.e. if the *MAXCHANGE keyword is missing).
For the *MISCG and the *MISNCG options defaults for *MAXCHANGE are:
Pressure Saturation Bubble Point Pressure -
1.0E+30 kPa, 1.45E+29 psia or 1.0E+28 kg/cm2

for pressure or bubble point pressure.
1.0 for oil, water, gas or solvent saturation.
defaults to the same value as Pressure
CONDITIONS:
EXPLANATION:
The *NORM keyword specifies the typical changes in the basic variables during a timestep.
These are used for automatic time-step selection.
The *MAXCHANGE keyword specifies the typical changes in the basic variables during a
timestep. If the change is greater than that specified by *MAXCHANGE then the time step is
cut in order to maintain the changes specified by *NORM.
The values for pressure and saturation changes MUST always be greater for
*MAXCHANGE than *NORM.
For field-scale water injection simulation, the pressure variation may be set to 7000 kPa
(1000 psi or 70 kg/cm2) and the saturation variation may be set to 0.20.
For single well coning studies a pressure variation of 1000 kPa (145 psi or 10 kg/cm2) and a
saturation variation between 0.05 and 0.10 may be used.
Examples:
*NORM *PRESS 145.0
*NORM *SATUR 0.08
User's Guide IMEX
Relaxation Options (Optional)
*RELAX
PURPOSE:
*RELAX selectively enables various relaxation options.
FORMAT:
*RELAX (*ACC (*OFF)) (*GAS (*OFF))
DEFINITIONS:
*ACC
This subkeyword enables over-relaxation in the accumulation term. This
may be helpful in gas- appearance, gas-disappearance problems.
*GAS
This subkeyword enables the under-relaxation for the gas saturation
disappearance.
*OFF
Disables the corresponding relaxation option.
DEFAULTS:
Optional keyword. If *RELAX is not present in the data set then the default is *OFF for
*GAS relaxation. *ACC relaxation is set by default to on, except when the pseudo-miscible
options are used. When the *MISCG or *MISNCG models are used, *ACC relaxation is
turned off by default.
CONDITIONS:
EXPLANATION:
The *RELAX *GAS enables the gas-saturation under-relaxation option. This option can lead
to faster convergence for simulations where the overall pressure in the reservoir is declining
and gas is being liberated due to the pressure falling below the bubble point.
This option prevents the gas from disappearing in a grid block for three Newtonian iterations.
If IMEX predicts that gas in a grid block will disappear for more than three iterations, only
then will the gas be allowed to disappear.
Conversely this option can slow the convergence for those simulations where there is free gas
present, and the reservoir pressure is rising. This is because for such situations when IMEX
predicts gas disappearance in a grid block the gas indeed should be allowed to disappear.
The *RELAX *ACC option performs over-relaxation of the accumulation term. This
normally improves model performance.
User's Guide IMEX
Examples:
*RELAX *GAS *OFF
*RELAX *GAS
User's Guide IMEX
**
**
**
**
Do not use gas saturation

under-relaxation
Use gas saturation
under-relaxation
Adaptive Implicit Switching (Optional)
*AIM
PURPOSE:
*AIM controls the adaptive implicit switching option.
FORMAT:
*AIM
(*OFF)
(*STAB)
(*STAB)
(*STAB)
(*THRESH)
(*BACK freq)
(*ALL-BLOCKS)
(*AND-THRESH)
(fraction1 fraction2)
DEFINITIONS:
*OFF
Adaptive implicit option is not used. The problem will be solved with fixed
implicitness.
*STAB
Adaptive implicit option with stability switching criterion is used.
*BACK
This subkeyword specifies backward switching. It enables switching from
implicit to explicit based on the stability switching criterion.
*ALL-BLOCKS
By default, the stability based switching algorithm only checks explicit
neighbours of implicit blocks for stability. This feature checks all blocks for
explicit to implicit switching. This feature is more robust than the *STAB
option on its own but can add 10-15% to CPU times when many explicit
blocks are present.
*AND-THRESH
A hybrid technique which checks explicit neighbours of implicit blocks for
stability switching and checks all blocks for threshold switching using the
default *NORM fraction of 0.25 (fraction1=fraction2=0.25). This technique
is nearly as robust as the *ALL-BLOCKS method and requires no extra CPU
time.
freq
An integer number to specify the frequency of checking backward switching.
(timesteps)
*THRESH
This subkeyword identifies threshold values for adaptive implicit switching.
User's Guide IMEX
fraction1
A real number specifying the fraction of values set by *NORM for checking
the bubble point pressure (the primary variable).
fraction2
A real number specifying the fraction of values set by *NORM for checking the
oil, water and gas saturations and to the polymer concentration when applicable
(primary variables). The value fraction2 is required with *THRESH.
DEFAULTS:
Optional keyword. If it is not present in the input-data-file, the default is:
*AIM *STAB *AND-THRESH
Default values for *AIM *THRESH are:
*AIM *THRESH 0.25 0.25
CONDITIONS:
EXPLANATION:
This keyword controls the switching of a grid block between an IMPES and a fully implicit
formulation.
Switching is based on either the numerical stability of a local amplification matrix or on
threshold changes in primary variables or on both. It is recommended that the stability
switching criterion *STAB or the hybrid criterion *STAB *AND-THRESH be used.
The variable fraction1 applies to bubble point pressures, and fraction2 applies to saturations.
The variable fraction2 also applies to the polymer reference concentration (*PREFCONC)
when *POLYMER option is used.
One of two cases may exist when using fraction1 and fraction2: either both are entered or
none are entered and the defaults apply.
Examples:
** Threshold values for adaptive implicit switching
*AIM *THRESH 0.25 0.25
** stability adaptive implicit switching
*AIM *STAB
** Hybrid stability-threshold adaptive implicit
** switching
*AIM *STAB *AND-THRESH
User's Guide IMEX
Convergence of Newton's Method (Optional)
*CONVERGE
PURPOSE:
*CONVERGE is used to specify the convergence criteria for Newton's Method when solving
the non-linear conservation equations in the simulator.
FORMAT:
*CONVERGE
(*PRESS) tol
(*SATUR)
-or*CONVERGE
(*MAXRES) value
(*MAXRES *OIL) value
(*MAXRES *WATER) value
(*MAXRES *GAS) value
(*MAXRES *SOLVENT) value
(*MAXRES *POLYMER) value
(*MAXRES *LIGHTOIL) value
(*MAXRES *SEAWATER) value
(*MAXRES *OTHER) value
-or*CONVERGE
(*TOTRES) value
(*WELLRES)
DEFINITIONS:
*PRESS
Indicates changes in convergence tolerance for pressure (kPa | psi | kPa |
kg/cm2).
*SATUR
Indicates changes in convergence tolerance for saturation. (fraction,
dimensionless).
tol
A real number indicating the tolerance value.
*MAXRES
Indicates the maximum scaled residual allowed for any single equation at
convergence.
User's Guide IMEX
Form
Applies Value to
*MAXRES value
*MAXRES *OIL value
*MAXRES *WATER value
*MAXRES *GAS value
*MAXRES *SOLVENT value
*MAXRES *POLYMER value
*MAXRES *LIGHTOIL value
*MAXRES *SEAWATER value
*MAXRES *OTHER value
All Equations
Oil Equation
Water Equation
Gas Equation
Solvent Equation
Polymer Equation
API Tracking Light Oil Component
Seawater Equation
Solvent, Polymer, Seawater or Light Oil
Component
*TOTRES
Indicates the maximum average scaled residual for all equations for a
particular phase or component. (i.e. the average of all oil equations in IMEX,
or the average of all flow equations for a particular component in GEM).
*WELLRES
Indicates the maximum scaled residual allowed for any single well equation
at convergence.
value
A real number indicating the residual value. (saturation fraction for IMEX)
DEFAULTS:
Optional keyword. The default is to use a residual convergence criteria:
Maximum average residual 0.001
Maximum well residual 0.001
Maximum Residual 0.1 (Single Porosity)
Maximum Residual 0.5 (Dual Porosity)
Polymer Maximum Residual is reduced to 0.005 times the default value if not explicitly
specified using the *MAXRES *POLYMER keyword
Seawater Maximum Residual is reduced to 0.005 times the default value if not
explicitly specified using the *MAXRES *SEAWATER keyword
Oil Maximum Residual (when using the *GASWATER_WITH_CONDENSATE model
only) is reduced to 0.1 times the default value if not explicitly specified using the
*MAXRES *OIL keyword
If a changes convergence criteria is specified, the default for the other changes tolerance is:
Pressure 10 kPa, 2.0 psia or 0.10 kg/cm2
Saturation 0.005
CONDITIONS:
This keyword must be located in the Numerical Methods Control keyword group. You
cannot specify both changes convergence criteria and residual convergence criteria.
User's Guide IMEX
EXPLANATION:
*CONVERGE specifies the convergence criteria and tolerances for conservation equations
during Newton iterations.
The default is to use a residual criteria to specify convergence. Equations are considered
converged when the maximum residual is less than a specified value, and the L1 norm
(average absolute value) of the residuals for each phase is less than a specified value.
The material conservation equation is:
(Flow Terms) + (Source Terms) - (Accumulation Term) = R
Using Newton's Method, the equation is not solved exactly to zero, but to some residual value
R, less than the specified residual for convergence. R is first scaled to give a "saturation
equivalent" error. Well equation residuals are expressed as a fraction of the specified rate.
If the format *CONVERGE *MAXRES value is used, or the default values above are used,
then:
1. The Polymer or Seawater Maximum residual is modified to be 0.005 x value.
2. When using the GASWATER_WITH_CONDENSATE model, the Oil residual is
modified to be 0.10 x value.
The format *CONVERGE *MAXRES component value has been introduced to allow the
user complete control over Maximum Residual convergence tolerances. The component
keyword can be one of *OIL, *WATER, *GAS, *SOLVENT, *POLYMER, *SEAWATER,
*LIGHTOIL, or *OTHER. When a components *Maximum residual tolerance is specified
explicitly in this fashion it is used directly. This format may be used to override the default
tolerance or a tolerance specified on a preceding *MAXRES value keyword. See examples
below.
Alternately, the changes in primary variables can be used as a convergence criteria. THIS
WAS THE METHOD USED IN IMEX 4.0 AND PRIOR VERSIONS.
Equations are considered converged when changes during Newtonian iterations are less than tol.
Examples:
*CONVERGE
*CONVERGE
*CONVERGE
*CONVERGE
*PRESS
*SATUR
*MAXRES
*MAXRES
2.50
0.005
0.01
*GAS 0.01
*CONVERGE *MAXRES 0.05

*MAXRES *OIL 0.002
*MAXRES *GAS 0.05
*CONVERGE *MAXRES 0.05
*CONVERGE *MAXRES *OIL 0.002
*CONVERGE *MAXRES *GAS 0.05
**Set tolerance to 0.01 globally

**Override Default for gas
**component, use defaults for
**other components
**Set tolerances to 0.05 globally
**Set Oil Component tolerance
**to 0.002
**Set Gas Component tolerance
**to 0.05
**Same action as preceding
**example
For a comparison of residual criteria and changes criteria, see CMG report 90.02.R.
User's Guide IMEX
Maximum Newtonian Cycles (Optional)
*NEWTONCYC
PURPOSE:
*NEWTONCYC is used to specify the maximum number of Newtonian cycles.
FORMAT:
*NEWTONCYC maxn
DEFINITIONS:
maxn
An integer indicating the maximum value.
DEFAULTS:
Optional keyword. If it is not present in the input-data-file, the default maximum number of
Newtonian cycles is 10.
CONDITIONS:
EXPLANATION:
This keyword specifies the maximum number of Newton iterations allowed in a timestep for
solution of the basic reservoir equations. If a solution within the convergence tolerances is
not achieved, the simulator will reduce the timestep size and repeat the timestep.
Examples:
*NEWTONCYC 8
User's Guide IMEX
Minimum Number of Newtonian Cycles (Optional)
*MINCYC
PURPOSE:
*MINCYC is used to specify the minimum number of Newtonian cycles when using the
*DRSDT or *DPBDT option.
FORMAT:
*MINCYC minn
DEFINITIONS:
minn
An integer indicating the minimum value.
DEFAULTS:
Newtonian cycles is 1.
CONDITIONS:
This keyword must be located in the Numerical Methods Control keyword group and is only
used when the *DRSDT or *DPBDT options are used.
EXPLANATION:
This keyword specifies the minimum number of Newton iterations allowed in a timestep for
solution of the basic reservoir equations.
Examples:
*MINCYC 2
User's Guide IMEX
Number of Cuts (Optional)
*NCUTS
PURPOSE:
*NCUTS controls the number of time-step size cuts allowed in a single timestep.
FORMAT:
*NCUTS value
DEFINITONS:
value
A real number to specify the maximum number of cuts allowed.
DEFAULTS:
Optional keyword. If it does not appear in the data set, then the default is 4 cuts.
CONDITIONS:
User's Guide IMEX
Convergence Tolerance for Linear Solver (Optional)
*RELTOL
PURPOSE:
*RELTOL is used to specify the convergence tolerance for the linear equation solver
(AIMSOL), when using *CONVERGE *PRESS or *CONVERGE *SATUR to specify
Newton Iteration convergence.
FORMAT:
*RELTOL tol
DEFINITIONS:
tol
DEFAULTS:
Optional keyword. If it is not present in the input-data-file, the default is: 0.1
CONDITIONS:
EXPLANATION:
Convergence of the linear equation solver is when, for all primary variables, the change in the
update value for the primary variable during a linear solver iteration is less than the Newton
Iteration convergence value specified by *CONVERGE times the value specified by
*RELTOL.
Examples:
*CONVERGE *PRESS 5.0** kPa
*CONVERGE *SATUR 0.1
*RELTOL 0.05
The linear solver stop it's iterations when all the changes in pressure primary variable updates
are less than 0.25 kPa (5.0 * 0.05) and all the changes in saturation primary variable updates
are less than 0.005 (0.1 * 0.05).
The linear solver will also stop iterating if the criteria specified by *PRECC is satisfied.
User's Guide IMEX
Convergence Tolerance for Linear Solver (Optional)
*PRECC
PURPOSE:
*PRECC is used to specify the convergence tolerance for the linear equation solver
(AIMSOL).
FORMAT:
*PRECC precc
DEFINITIONS:
precc
DEFAULTS:
Optional keyword. If it is not present in the input-data-file, the default is: 1.0E-04
CONDITIONS:
EXPLANATION:
Convergence of the linear equation solver is when the r.m.s. value of the residuals of the
equations has been reduced to precc times its original value.
If the set of linear equations is
Av = b
the r.m.s. residual is
r = || b - Av ||
Convergence is when, for the ith iteration,
r(i) / r(0) < precc
User's Guide IMEX
Orthogonalization (Optional)
*NORTH
PURPOSE:
*NORTH controls the maximum number of orthogonalizations to be performed before
resetting for the iterative solution method.
FORMAT:
*NORTH num
DEFINITIONS:
num
An integer defining the maximum number of orthogonalizations allowed.
DEFAULTS:
Optional keyword. If it is not present in the input-data-file, the default value is: 30
CONDITIONS:
EXPLANATION:
NOTE: The maximum number of orthogonalizations The user is not advised to alter solver
dimensioning without first contacting IMEX support at CMG for guidance.
Examples:
*NORTH 60
User's Guide IMEX
Solver Equation Ordering (Optional)
*SORDER
PURPOSE:
*SORDER controls the ordering of equations in ILU factorization.
FORMAT:
*SORDER
(*NATURAL)
(*REDBLACK)
(*RCM)
(*RCMRB)
DEFINITIONS:
*NATURAL
Use natural ordering.
*REDBLACK
Red-black reduced system preconditioning is used, similar to a D4 reordering
scheme. Elimination is performed on the reduced system of grid blocks
labeled "black", saving storage and computer time.
*RCM
Reverse Cuthill-McKee ordering is used. The ordering algorithm attempts to
minimize the bandwidth in the L & U factors. Compared to *NATURAL,
use of this scheme can result in significant CPU savings for higher values of
degree, especially *GAUSS. Savings are more modest when a low value of
IDEG is used, since matrix fill is small. For a regular grid with no wells and
no null blocks this method is equivalent to D2 ordering.
*RCMRB
Use reverse Cuthill-McKee ordering, then red/black ordering. For a regular
grid with no wells and no null blocks this method is equivalent to D4
ordering.
DEFAULTS:
*SORDER *REDBLACK
CONDITIONS:
EXPLANATION:
Examples:
*SORDER *NATURAL
User's Guide IMEX
Solver Factorization Degree (Optional)
*SDEGREE
PURPOSE:
*SDEGREE controls the maximum degree of fill terms used in the factorization.
FORMAT:
*SDEGREE
(max_deg)
(*GAUSS)
DEFINITIONS:
max_deg
An integer to specify maximum degree of fill terms.
*GAUSS
Keyword specifying that Gaussian elimination be used.
DEFAULTS:
*SDEGREE 1 (first degree with non-natural ordering)
*SDEGREE 2 (second degree with natural ordering)
CONDITIONS:
EXPLANATION:
This keyword controls the maximum degree of fill terms used in the calculation of the LU
factors via incomplete Gaussian elimination, where max_deg is an integer. A value of
max_deg greater than the matrix bandwidth or use of the *GAUSS keyword results in
complete Gaussian elimination. In general, larger values of max_deg may be required for
more difficult problems (extreme permeability contrasts). Larger values of max_deg result in
more calculations and a longer simulation run time.
Starting with IMEX 97.00, the solver efficiency for higher order (>1) has been improved
significantly. Thus it may be feasible to use the *SDEGREE 2 factorization without a large
increase in CPU time. However, memory usage has not changed and may limit the use of
higher order factorizations when running large models.
Examples:
** Use Gaussian elimination
*SDEGREE *GAUSS
User's Guide IMEX
Pivot Stabilization (Optional)
*PIVOT
PURPOSE:
*PIVOT controls the diagonal submatrix inversion pivot stabilization.
FORMAT:
*PIVOT
(*ON)
(*OFF)
DEFINITIONS:
*ON
Pivot stabilization is performed.
*OFF
No pivot stabilization is performed.
DEFAULTS:
Optional keyword. If it is not present in the input-data-file, the default is no pivot
stabilization:
*PIVOT *OFF
Specifying *PIVOT without a subkeyword will default to:
*PIVOT *ON
CONDITIONS:
EXPLANATION:
This keyword selects the pivot stabilization of diagonal submatrix inversion.
Examples:
** Use pivot stabilization.
*PIVOT *ON
User's Guide IMEX
Maximum Iterations (Optional)
*ITERMAX
PURPOSE:
*ITERMAX is used to specify the maximum number of iterations allowed in the Jacobian
matrix solution routine.
FORMAT:
*ITERMAX maxn
DEFINITIONS:
maxn
An integer indicating the maximum value.
DEFAULTS:
iterations is:
IMEX - 40
CONDITIONS:
EXPLANATION:
This keyword specifies the maximum number of iterations allowed in the iterative solution of
the linearized set of equations.
The minimum allowable value for maxn is 1. This is recommended with *GAUSS only.
The maximum allowable value for maxn is 200.
User's Guide IMEX
AIMSOL/PARASOL Switch (Optional)
*SOLVER
PURPOSE:
Choose which solver to use, AIMSOL or PARASOL (In order to use PARASOL, the parallel
computing licensing feature must be active).
FORMAT:
*SOLVER (*AIMSOL | *PARASOL)
DEFINITIONS:
*AIMSOL
CMGs non-Parallel Iterative Solver.
*PARASOL
CMGs Parallel Iterative Solver.
DEFAULTS:
Optional keyword. If it is not present in the input-data-file, *SOLVER *AIMSOL is
assumed.
CONDITIONS:
EXPLANATION:
This keyword specifies which solver to use. *SOLVER *PARASOL is required in order to
solve the linear system of equations in parallel. *JACPAR, *JACDOMS, *DPLANES or
*DTYPE is required to solve the Jacobian problem in parallel.
When installed in the Launcher, the Parallel version of IMEX is controlled via the command
line arguments:
-jacpar
*JACPAR *ON
and overrides the DTYPE/DPLANES keywords in the data set, and
-parasol n
*SOLVER *PARASOL
*PNTHRDS m
keywords
User's Guide IMEX
where n is an integer greater than 0, representing the desired target number of threads, and m
is the minimum of n and the number of logical CPU's. Note, for previous versions of IMEX,
-parasol n would only work for n=2, 4, 8, 16 or 32 and would use PPATTERN l, where l = 2,
3, 4, 8 or 9 respectively.
In addition the command line option -aimsol can be used to override *SORDER
*PARASOL and any PARASOL keywords in the data.
User's Guide IMEX
Red-Black Ordering Check for Parasol (Optional)
*CHECKRB
PURPOSE:
Choose when to abandon using Red-Black Ordering for a PARASOL class (a class is defined
as a disjoint set of blocks).
FORMAT:
*CHECKRB ( *ON | *OFF)
DEFINITIONS:
*CHECKRB *OFF
Always use Red-Black ordering for a class.
*CHECKRB *ON
Red-Black ordering is abandoned for a class in which the fraction of red
blocks is too small (when the number of red blocks is less than 60% of the
number of black blocks).
DEFAULTS:
Optional keyword. If it is not present in the input-data-file, the default is *OFF, thus RedBlack ordering is always performed when specified as the ordering method.
CONDITIONS:
This keyword must be located in the Numerical Methods Control keyword group and used
with *SOLVER *PARASOL.
EXPLANATION:
It may not always be efficient to perform Red-black ordering in situations where the number of
red blocks is small. This keyword allows the user to abandon Red-Black ordering in this
situation. This keyword has no effect if a red-black ordering has not been specified or defaulted.
User's Guide IMEX
Factorization Degree within PARASOL Classes (Optional)

*PDEGAA
PURPOSE:
Choose the factorization degree within PARASOL classes (a class is defined as a disjoint set
of blocks).
FORMAT:
*PDEGAA idegaa
DEFINITIONS:
idegaa
Factorization degree within each class
DEFAULTS:
Optional keyword. Default is the value of *SDEGREE, which by default is 1 for Red-Black
and RCMRB ordering, and 2 for natural ordering (see *SDEGREE in the Numerical Methods
Section).
CONDITIONS:
EXPLANATION:
This keyword allows the user to control the degree of factorization used within a class for
*SOLVER *PARASOL.
User's Guide IMEX
Factorization Degree between PARASOL Classes (Optional)

*PDEGAB
PURPOSE:
Choose the factorization degree between PARASOL classes (a class is defined as a disjoint
set of blocks).
FORMAT:
*PDEGAB idegab
DEFINITIONS:
idegab
Factorization degree between classes
DEFAULTS:
Optional keyword. Defaults to idegaa + 1 (See *PDEGAA).
CONDITIONS:
EXPLANATION:
This keyword allows the user to control the degree of factorization used between classes for
*SOLVER *PARASOL. PARASOL allows red-black elimination only within a class; thus
when 1st degree red-black ordering is used it is important that factorization of at least degree 2
be used between classes to raise the quality of the factorization above that obtained with the
1st degree natural ordering.
User's Guide IMEX
Parasol Class Partitioning Pattern (Optional)
*PPATTERN
PURPOSE:
*PPATTERN sets the basic partitioning of the reservoir into non-connected regions and
separators that makes possible the parallelization of the linear solution.
FORMAT:
*PPATTERN
or
*PPATTERN
*AUTOPSLAB inum
( ipatrn | *PARTITION | *PPARTITION | *GPARTITION |
*APARTITION )
DEFINITIONS:
*AUTOPSLAB inum
inum is the target number of level-1 classes. There are inum-1 separating
plane classes, and 0 or 1 class containing demotions. The direction taken is
one with the maximum number of planes that do not cut the dominant
transmissibility direction. Thus inum corresponds to the target number of
processors desired. The partitioning the reservoir under this method takes
care of the presence of null blocks, if any, thus distributing the active blocks
as equally as possible amongst the domains. This is the recommended way of
specifying the class partitioning.
ipatrn
ipatrn can have the values 0, 1, 2, 3, 4, 5, 6,7, 8 or 9. Figure 1 provides a
geometrical representation of different classes under ipatrn 1 through 7.
Table 1 summarizes the class distribution in the ipatrns. Column 5 of the
table shows the number of level-1 classes under each ipatrn, corresponding
to the target number of threads desired.
Please note that unlike *AUTOPSLAB above, the specification under
*PPATTERN using ipatrn does not adjust the partitioning automatically due
to presence of null blocks. The user is expected to select a particular ipatrn
based on the reservoir geometry, dominant flow direction, and distribution of
null blocks in the reservoir.
User's Guide IMEX
*PARTITION
'class partitioned' '1st major new class' '2nd major new class' 'separator class'
(*I|*J|*K) ind.
:
Each line directs the partitioning of the first class into two major and one
separator class, with the original class no longer existing after the partition.
The partitioning is planar, with the separator parallel to the I, J, or K axis as
specified, with index value ind. Initially there is the one class 'FIELD'; each
line creates three new classes and destroys one, for a net gain of two classes
per line. The names serve only to identify the classes while the pattern is being
created; they are not referred to thereafter. In principle any number of lines
may be entered after *PPATTERN *PARTITION;
Current limit is 64 lines per *PPATTERN keyword.
*PPARTITION
:
Like *PARTITION but simulator decides automatically what direction plane
to use and where to put it. The decisions are made to equalize class sizes as
much as possible and to minimize the size of the separator class.
*GPARTITION
:
Uses Alan George's rooted-level-structure method to partition 'class
partitioned' into 3 parts, 2 large classes and a separator. Like *PPARTITION
but doesn't use planes.
*APARTITION
:
"agglomeration partition" -- like *GPARTITION but provides classes
somewhat more nearly equal in size, but somewhat less regular in shape.
User's Guide IMEX
Example:
The 3-class, 2 level partitioning given in the initial description of the PPATTERN
keyword can be realized either with
*PPATTERN *AUTOPSLAB 2
or
*PPATTERN 2
or
*PPATTERN *PARTITION
'FIELD' 'Class 1' 'Class 2' 'Class 3' *I 51
DEFAULTS:
Optional keyword. If *PPATTERN is absent then *AUTOPSLAB 2 is assumed.
CONDITIONS:
This keyword is used only with *SOLVER *PARASOL.
EXPLANATION:
The parallelization of the solver requires the partitioning of the reservoir into disjoint sets of
blocks (classes). The classes are further organized into levels.
There must be no flow between blocks which are in different classes at the same level.
Example:
Consider a 101 x 50 x 10 reservoir and consider partitioning it into 3 classes as follows:
Class 1:
I = 1:50
J = 1:50
K = 1:10;
Class 2:
I = 52:101
J = 1:50
K = 1:10;
Class 3:
I = 51
J = 1:50
K = 1:10.
The large classes, 1 and 2, have no direct flow interactions because all flow between them
must go through blocks in class3. Classes 1 and 2 are at Level 1; class 3 is at level 2.
User's Guide IMEX
ipatrn
Description
Single class
(like AIMSOL)
Two crossed vertical

planes
Single vertical plane

cutting the reservoir
into two
Three parallel vertical

separating planes
Seven parallel vertical

separating planes
Four parallel vertical

separating planes
Like ipatrn 2, but

separating plane is
horizontal
Geometrical Representation
Legend
7
Like ipatrn 2, but

with 2 vertical
separating planes
Level - 1
Level - 2
Level - 3
Figure 1: Geometrical representation of class distribution under different ipatrns
User's Guide IMEX
Table 1: Summary of number of levels and classes under different ipatrns.

ipatrn
Description
Total
levels
Total
classes
Level
1
Class
Distribution
Level
2
Remarks
Level
3
Single Class (like AIMSOL)
Two crossed vertical planes
Single vertical plane cutting

the reservoir into two
Three parallel vertical

separating planes
Level-3 contains
demotions only
Seven parallel vertical

separating planes
16
Level-3 contains
demotions only
Four parallel vertical

separating planes
Level-3 contains
demotions only
Like ipatrn 2, but separating

plane is horizontal
Like ipatrn 2, but with 2

vertical separating planes
Level-3 contains
demotions only
Fifteen parallel vertical

separating planes
32
16
15
Level-3 contains
demotions only
Thirty-one parallel vertical

separating planes
64
32
31
Level-3 contains
demotions only
User's Guide IMEX
Target number of Planes per Jacobian Domain (Optional)

*DPLANES
PURPOSE:
Choose the target number of planes per Jacobian domain. (In order to use *DPLANES, the
parallel computing licensing feature must be active).
FORMAT:
*DPLANES imxdom
DEFINITIONS:
imxdom
Target number of planes per domain.
DEFAULTS:
Optional keyword. Planes are chosen in the dimension direction with the largest number of
non-trivial planes. imxdom is the number of corresponding non-trivial planes in this
direction per domain. If *DPLANES is specified and imxdom is not, then the default is 4.
CONDITIONS:
EXPLANATION:
This keyword allows the user to control the number of planes per Jacobian domain.
*SOLVER *PARASOL is required in order to solve the linear system of equations in
parallel. *JACPAR, *JACDOMS, *DPLANES or *DTYPE is required to solve the Jacobian
problem in parallel.
When installed in the Launcher, the Parallel version of IMEX is controlled via the command
line arguments:
-doms
*DPLANES
and overrides the JACPAR/JACDOMS/DTYPE/DPLANES keywords in the
data set, and
-parasol
which is equivalent to the keywords
*SOLVER *PARASOL
Note: Changing between parallel and serial options or among parallel options such as
*DPLANES/JACDOMS/JACPAR or different values of PPATTERN AUTOPSLAB n on
a restart run is currently not supported.
If the original run was parallel (for example parasol n jacpar) and you wish to run a restart
on a machine with fewer CPUs m, you may use parasol n pnthrds m jacpar.
User's Guide IMEX
Number of Threads to be Used (Optional)
*PNTHRDS
PURPOSE:
Choose the number of threads to be used for the simulation.
FORMAT:
*PNTHRDS ithrds
DEFINITIONS:
ithrds
Number of threads used.
DEFAULTS:
Optional keyword. If parallel Jacobian building and PARASOL are not specified, then it is
defaulted to one. If parallel Jacobian building or PARASOL is specified, then it is defaulted
to the minimum of 2 and the number of processors in the current machine.
CONDITIONS:
EXPLANATION:
If *PNTHRDS is set to a number greater than the number of processors, performance will
degrade. If *PNTHRDS is set to greater than two, then the solver *PPATTERN should be
changed in order to load balance properly, otherwise poor performance is likely to occur.
User's Guide IMEX
Complete Storage Grid Array of Jacobian Domain Numbers

(Optional)
*DTYPE
PURPOSE:
*DTYPE explicitly defines the domain numbers of individual blocks.
ARRAY:
*DTYPE
DEFINITIONS:
values
A number indicating a given blocks domain.
DEFAULTS:
Optional keyword.
CONDITIONS:
This keyword must be located in the Numerical Methods Data keyword group.
EXPLANATION:
This keyword explicitly sets the Jacobian domains of individual blocks.
User's Guide IMEX
Target Number of Jacobian Domains (Optional)
*JACDOMS
PURPOSE:
Choose the target number of Jacobian domains. (In order to use *JACDOMS, the parallel
FORMAT:
*JACDOMS ndoms
DEFINITIONS:
ndoms
Target number of domains.
DEFAULTS:
Optional keyword. If *JACDOMS is specified and ndoms is not, then the default is 2.
CONDITIONS:
EXPLANATION:
This keyword allows the user to control the number of Jacobian domains. For ndoms, ndoms
domains, plus, if needed, an additional garbage can domain may be set up. *SOLVER
*PARASOL is required in order to solve the linear system of equations in parallel. *JACPAR,
*JACDOMS, *DPLANES or *DTYPE is required to solve the Jacobian building in parallel.
If the number of threads, *PNTHRDS, is not specified, then the number of threads is set to
min (ndoms, number of logical CPUs).
The parallel version of IMEX may also be controlled via the command line arguments:
-jacdoms
*JACDOMS 2
and overrides the JACPAR/JACDOMS/DTYPE/DPLANES keywords in the data set,
and
-parasol
*SOLVER *PARASOL
keywords
To use n logical CPUs where n is greater than or equal two, on a machine with two or more
logical CPUs, the command line options:
-jacdoms n parasol n
may be used.
User's Guide IMEX
Using command line arguments over-ride the keywords in the dataset. If jacdoms m and
-parasol n are both used, then the number of threads to be used will be determined by m. If
-pnthrds l is used, then l threads will be used, rather than depending on m or n.
User's Guide IMEX
Target Number of Jacobian Domains based on Parasol

Classes (Optional)
*JACPAR
PURPOSE:
Choose the target number of Jacobian domains. (In order to use *JACPAR, the parallel
FORMAT:
*JACPAR *ON|*OFF
DEFAULTS:
Optional keyword.
CONDITIONS:
EXPLANATION:
This keyword allows the user to control parallel Jacobian building when also using *SOLVER
*PARASOL. *SOLVER *PARASOL is required in order to solve the linear system of
equations in parallel. With this option, the Jacobian domains are based on the Parasol
classes.*JACPAR, *JACDOMS, *DPLANES or *DTYPE is required to solve the Jacobian
building in parallel.
If the number of threads, *PNTHRDS, is not specified, then the number of threads is set to
min (ndoms, number of logical CPUs).
The parallel version of IMEX may also be controlled via the command line arguments:
-jacpar
*JACPAR *ON
and overrides the JACPAR/JACDOMS/DTYPE/DPLANES keywords in the data set,
and
-parasol
which is equivalent to the keywords
*SOLVER *PARASOL
To use n logical CPUs where n is greater than or equal two, on a machine with two or more
logical CPUs, the command line options:
-jacpar parasol n
may be used.
Using command line arguments over-ride the keywords in the dataset. If -pnthrds l is used,
then l threads will be used, rather than depending on n.
User's Guide IMEX
User's Guide IMEX
Notes on Well and Recurrent Data

The *DATE or *TIME keyword indicates a new well change time. Any well keywords that
appears between two *DATE or *TIME cards will be applied between the times indicated. The
first well change must be a *DATE keyword indicating the starting date of the simulation.
Well names are entered via the keyword *WELL. Wellbore geometries can be entered using
the *GEOMETRY keyword and *PERF indicates the grid blocks where the well is
perforated. Refer to Appendix B for the calculation of well index from the well geometries
and other information for the well model used.
The type of well is described by the keywords *PRODUCER or *INJECTOR. They must be
followed by the operating conditions and constraints of the well indicated by the *OPERATE
and *MONITOR keyword. For injectors, the injection fluid composition must also be
specified by the *INCOMP keyword. Wells can be connect to a group using the
*ATTACHTO subkeyword under the *WELL keyword. If a well is not attached to any
group it is attached to a group called 'DEFAULT-GROUP' by IMEX. Any operating
condition and its value can be altered at a later time using a single keyword *TARGET
A well can be shut in at any time by the keyword *SHUTIN and reopened at a later time
using the keyword *OPEN. It can also be placed in an automatic drill queue by specifying its
status as *AUTODRILL. The automatic drilling option is useful during the prediction phase
of a simulation study. It allows the user to meet prespecified targets for a group of wells by
allowing the program to open (drill) new wells to meet a specified target.
The wellbore hydraulics model relates the bottom-hole pressure to the well head pressure of a
well. This model must be activated for wells with well head pressure constraints specified.
This could be achieved by entering the wellbore data using the *PWELLBORE or
*IWELLBORE keywords. These keywords must be entered after the well type specification
keywords (*PRODUCER or *INJECTOR).
The simulation is started by *RUN and terminated by *STOP. The timestep sizes after a well
change is controlled by *DTWELL. All timestep sizes are limited by *DTMIN and *DTMAX.
Initial implicit-explicit pattern of the grid blocks are controlled by the keywords *AIMSET
and *AIMWELL.
User's Guide IMEX
Well and Recurrent Data 663
In this section of the user's manual all *MODSI units are the same as *SI units except for
pressures which are in kg/cm2 for *MODSI.
The use of quotes in well names is supported in a consistent manner. To use a double quote
in a well name, use a single quote as a well string delimiter. To use a single quote in a well
name, use a double quote as a well string delimiter.
664 Well and Recurrent Data
User's Guide IMEX
New Keywords and Options for the 2010.10 Release

1. Completion Lumping Option
2. *WHYSTAB *AUTOWELL Monitor
3. *MXCNRPT Maximum number of Continue Repeats
4. *KRPERF *SORMAX Entry for Trapped Oil Hysteresis
Completion Lumping Option
*LAYERCLUMP provides a way for the user to define a control lump - a set of layers
(perforations, completions) which possesses a name and a set of computed rate, cumulative,
and stream ratio values. To each control lump there is also an associated setting value.
*LAYERRLUMP provides a way for the user to lump well layers in a group called a report
lump. For report lumps, stream rates, cumulatives and production rate ratios are calculated
and the user can design triggers based on these quantities. A given layer can be assigned a
partial membership to report lump and the same layer can belong to more than one group.
*RM-LAYERCLUMP removes membership of a layer from the named control lump.
*RM-LAYERRLUMP removes membership of a layer from the named report lump.
*CLUMPSETTING assigns a control setting or flow restriction parameter to the named
control lumps. The flow restriction parameter is used as a multiplication factor in adjusting
the value of the well index for all the layers belonging to a given control lump.
There is also a set of computed rate, cumulative, and stream ratio values for each control
lump. Triggers may be defined using these quantities specific to named control lumps;
actions can be any valid well data lines, including lines which alter the control lump setting
values. The syntax of the control lump based trigger statement is:
*TRIGGER trigger_name *ON_CTRLLUMP control_lump_name quantity > value
The syntax of the report lump based trigger statement is:
*TRIGGER trigger_name *ON_RPTLUMP report_lump_name quantity > value
See *LAYERCLUMP keyword for a list of quantities which can be used in control and report
lump triggers.
*WHYSTAB *AUTOWELL Monitor
*MONITOR WHYSTAB *AUTOWELL frequency can be used to monitor production
well wellbore hydraulics stability using the *AUTOWELL option, which shuts the well for a
specified time.
User's Guide IMEX
*MXCNRPT Maximum Number of Continue Repeats

and keyword *MXCNRPT lets the user override its default value (of 1).
*KRPERF *SORMAX Entry for Trapped Oil Hysteresis
SORMAX on the *KRPERF keyword allow the user to change this saturation endpoint in
recurrent data. Sormax is the oil saturation in the two phase oil-water system at which oil first
becomes immobile during the boundary/primary imbibition process (the imbibition starting
from oil saturation equal to 1-Swcon). It only has an effect when the Krow hysteresis or
trapped oil hysteresis is modeled for the rock type of the completion.
User's Guide IMEX

1. *ITUBE1/*PTUBE1
2. *GCONI Stream *PMAINT *PMMAXR d1 d2
3. *GCONPMULT *RECYCLE
4. *LAYERHFP
5. *PERFRG keyword removed
*ITUBE1/*PTUBE1
PTUBE has been replaced by the more flexible PTUBE1, in addition ITUBE1 allows for the
definition of injection tables using CMG standard well head pressure input table format.
*PTUBE tables are still supported.
*GCONI Stream *PMAINT *PMMAXR d1 d2
Introduces the maximum surface rate of the specified injection stream that is available for the
pressure maintenance. It is the summation of two sources: a fraction (d1) of the total
produced (from the same group) and a make-up rate (d2). The values must be non-negative
real numbers. The fraction d1 is nondimensional and defaulted to 0. The make-up rate d2 is
(m3/day | bbl/day | cm3/min) for water and (m3/day | scf/day | cm3/min) for gas or solvent,
and is defaulted to 1.0e+20.
*GCONPMULT *RECYCLE
This subkeyword specifies a recycling production target. This indicates that the production
wells connected to this group produce such that the phase injected by the injection wells
connected to this group as specified by *GAS, *WATER or *SOLVENT is reproduced
(recycled) out of the reservoir.
*LAYERHFP
productivity indices which act as proxies for vertical fractures (assuming pseudo radial flow)
on a layer by layer basis.
*PERFRG keyword removed
This keyword is no longer required as the *PERF keyword can be used instead.
User's Guide IMEX

1. Cycling Groups
2. *GAPPOR *AUTODRILL stream identifier
Cycling Groups
Allow the user to specify data for EOR/IOR processes which rely on alternating production
and/or injection cycles. Generally multiple cycles are required and each cycle consists of a
number of well defined parts. Examples include water alternating gas injection (WAG) or
cyclic steam injection.
*GCONCYCLE_START allows the user to begin to specify data for EOR/IOR processes
which rely on alternating production and/or injection cycles.
GAPPOR Stream Specification
User's Guide IMEX

1. Shut in Wells above Formation *MODELSHUT
2. Two New Drawdown Constraint Types *DWA, *DWB
3. Monitoring Well Head Pressure Stability *WHYSTAB
4. Voidage Replacement Production Target
5. Recycling Production Target *RECYCLE
6. Sector Pressure Maintenance Production Target *PMAINT
7. Shutting in Multiple Wells to meet Production Targets/Monitors *SHUTMOWS
8. Voidage Replacement Injection Targets including Makeup of Gas, Water or
Solvent *GMKUP, *WMKUP, *SMKUP
9. Sector Pressure Maintenance Injection Target *PMAINT
10. New Apportionment Methods for Meeting Group Targets *APPOR_METHOD,
*IP, *GUIDE, *INGUIDE, *PRIOR, *PRIOR-FORM
11. On-Time Factors Applying to Groups
Shut in Wells above Formation
Two New Drawdown Constraint Types
*DWA defines the maximum drawdown within all open layers,
Pd = max (Pblock l Pwell l )

l, open
which corresponds to the notion of imposing the maximum drawdown constraint to avoid
formation damage.
*DWB specifies the average drawdown for all open layers:
Pd =
l, open
l,open
weighted by the total product/injectivity (PI) at the reservoir condition.

Monitoring Well Head Pressure Stability
A real well can operate only above a certain rate at which the bottom-hole pressure remains
in the stable region of the well-head pressure curve (a region in which the curve of WHP vs.
BHP has a positive slope). It can only be used for wells for which a method of computing
WHP has been introduced with the *PWELLBORE or *IWELLBORE keyword.
User's Guide IMEX
Voidage Replacement Production Target (*GCONP *VREP)
This indicates that the production wells connected to this group produce an amount of the
bottom-hole fluid in proportion to the total bottom-hole fluid injected into the reservoir by the
injection wells connected to this group.
Recycling Production Target *GCONP *RECYCLE)
This indicates that the production wells connected to this group produce such that the phase
injected by the injection wells connected to this group as specified by *GAS, *WATER or
*SOLVENT is reproduced (recycled) out of the reservoir.
Sector Pressure Maintenance Production Target (*GCONP *PMAINT)
This subkeyword specifies that the group production rates shall be adjusted so as to maintain
Shutting in Multiple Wells to Meet Production Targets/Monitors (Subkeyword
*SHUTMOWS)
Action subkeyword indicating that if a maximum stock tank rate is exceeded for a group, then
a list of prioritized most offending wells (MOWS the wells with the higher rates of the
named surface stream) should be shut.
Voidage Replacement Injection Targets including Makeup of Gas, Water or Solvent
(GCONI *VREP *GMKUP/*WMKUP/*SMKUP)
A make-up stream can be added with *GMKUP, *WMKUP or *SMKUP to meet a total
voidage replacement fraction for each group under voidage replacement.
Sector Pressure Maintenance Injection Target
This subkeyword specifies that the group injection rates shall be adjusted so as to maintain
New Apportionment Methods for Meeting Group Targets
groups and the wells directly attached. . CMGs multi-level well management module
currently supports the following methods:
*PRIOR-FORM defines the priority formulae and numerical control parameters for the priority
ranking apportionment method set by *APPOR-METHOD *PRIOR to meet group targets.
User's Guide IMEX
On-Time Factors Applying to Groups
*ON-TIME specifies the fraction of time during which a well or group operates. Please see
the *ON-TIME keyword for a detailed description of how *ON-TIME operates with Groups.
User's Guide IMEX

1. Seawater Injection/Scale Buildup Option

2. TRIGGER Action Option
3. Well Definition Order Relaxed/Improved
4. *WLISTSHUT/*WLISTOPEN Option
5. *WTMULT/*GCONIMULT/*GCONPMULT Options
6. Improvements to Tubing Head Pressure Calculations
Seawater Injection/Scale Buildup Option
A Seawater Injection/Scale buildup option has been incorporated in IMEX. This feature is
actually made up of two independent options: 1) A Seawater Injection Option and 2) a Scale
Buildup/Well Damage model. The two models can be coupled to make well damage a
function of both water rate and the fraction of seawater in the produced water.
The Seawater Injection option adds a seawater flow equation to the normal Black Oil or Oil
Water Models in IMEX. This equation is solved simultaneously with the oil, water and gas
flow equations. It is an active tracking model, as the injected water may have a viscosity
different from formation water. Thus injection of seawater may alter fluid flow patterns within
the reservoir. This model and the Scale Buildup/Well Damage model are coupled through the
*SCLDPS table (Scale Buildup/Well Damage Option) which relates scale deposition to the
volume fraction of seawater in the produced water. The Seawater Option is controlled by the
following new keywords:
1)
2)
3)
4)
5)
6)
*MODEL *OILWATER_SEAWATER
*OUTPRN/*OUTSRF *GRID *SEAWATFRC
*CONVERGE *MAXRES *SEAWATER
*INCOMP SEAWATER
*ALTERFS
The Scale Buildup/Well Damage Model allows the user to model well (layer) productivity
damage based on the amount of water produced through each layer of a well. Optionally the
damage can be made a function of the fraction of seawater produced along with formation water.
This Option is controlled by the following new keywords:
1)
2)
3)
4)
5)
6)
*SCLDPS
*SCLDMG
*SCLTBL-WELL
*SCLTBL-LAYER
*SCLRMV-WELL
*SCLRMV-LAYER
Rock Fluid Data

Rock Fluid Data
User's Guide IMEX
The Scale Buildup/Damage model can be used with any Property model available in IMEX (for
example: API Tracking, Solvent, and Gas-water). It is only with the *BLACKOIL_SEAWATER
and *OILWATER_SEAWATER models that the active seawater equation can be used to make
scale formation a function of seawater fraction produced.
Trigger Actions (Well and Recurrent Data)
This option allows for certain actions to be implemented when a specified condition or trigger
is satisfied. The actions are specified in the form of a block of valid keywords encapsulated
within the *TRIGGER and *END_TRIGGER keywords.
The trigger condition of each active trigger is tested at the bottom (end) of the timestep (that
is after the timestep is considered to be completed, the cumulatives have been updated and
the simulation is ready to proceed to the next timestep). If the trigger condition is satisfied
then the list of associated actions are processed at the top (beginning) of the next timestep.
triggers are defined. The triggers are not sorted in any other way.
keyword of the outer trigger). Please see the *TRIGGER Keyword in the Well and
Recurrent Data Section for a complete description of the option.
More Flexible Well Definition Ordering (Well and Recurrent Data)
It is now possible to define a well, name it, and specify its group affiliation with a *WELL
become active. Previously it was necessary to define a wells type (injector or producer)
immediately after its initial definition.
*WLISTSHUT, *WLISTOPEN (Well and Recurrent Data)
*WLISTSHUT and *WLISTOPEN provide a means to temporarily shut a large
list of wells and later re-open them without disturbing the pattern of shut and
autodrillable wells that existed when the *WLISTSHUT keyword was read.
*WTMULT (Well and Recurrent Data)
*WTMULT allows modification of any previously specified well constraint
value for well(s) listed by well_numbers or 'well_names' by applying a
multiplying factor.
*GCONIMULT, *GCONPMULT (Well and Recurrent Data)
*GCONIMULT is used to specify multipliers which modify existing group
injection target controls. *GCONPMULT is used to modify existing group
production target controls with the use of multipliers.
Mechanistic/Momentum Wellbore Pressure Drop Model Improvements
The Mechanistic Wellbore pressure drop calculation in IMEX (*HEAD-METHOD

*MODEL-MOMENTUM) has been made more rigorous. The updated Momentum option is
currently a full implementation of the Petalas-Aziz mechanistic model.
User's Guide IMEX

Two major options have been added to IMEXs handling of groups and wells. The first is the
frictional pressure drop option; the second is the manifold group control option.
Keywords associated with frictional pressure drop.
*HEAD-METHOD well_list *GRAV-FRIC (*MODEL-AG |MODEL-HOMOGENEOUS|
*MODEL-MOMENTUM | *MODEL-DRIFTFLUX) (*HEADRROUGH rrough)
(*HEADSURFT surft)
This keyword describes which frictional pressure drop calculation method is
used for wells in the well list.
This keyword specifies the calculation procedure for calculating the frictional
pressure drop. The *ITERATIVE method allows the model to compute
heads which are consistent with completion pressures. The *ONE-STEP
method does not iterate to calculate consistent heads and very often results in
unphysical results.
This keyword when used determines whether well bottom hole pressures as
well as layer heads are made consistent with completion pressures. This
option is not normally required for good results as the HEADITER
*ITERATIVE option is quite robust on its own. Using the *COMBINED
option can result in significantly increased run times.
New keywords associated with the Manifold Model.
Determines which groups are manifold groups.
*PCON-MANIFOLD (*PROD | *GASI | *WATI | *SOLI) group_list (*ON | *OFF)
Determines whether rate constraints are translated into manifold pressure
constraints.
*GPHYDDEP (*PROD | *GASI | *WATI | *SOLI) group_list depths
This keyword adjusts the depth used to calculate the pressure drop between a
surface compressor and a subsea manifold.
This keyword determines which table is used to calculate the pressure drop
between a surface compressor and a subsea manifold.
This keyword specifies the group ALQ property which is used in standard
VFPPROD and VFPINJ tables which are associated with flow from manifold
to compressor.
User's Guide IMEX
*WELLALQ well_list
This keyword specifies the well ALQ property which is used in standard well
VFPPROD and VFPINJ tables.
Other keywords which are modified to account for manifold groups are:
*GCONI, *GCONP, *INCOMP and *IWELLBORE
User's Guide IMEX

*GCONM *MAXGAS
These keywords identifies a maximum surface gas rate monitor for group
with the highest GOR rather than the one with the highest gas rate.
*GCONM *MAXSTW
These keywords identifies a maximum water rate monitor for group
with the highest WCUT rather than the one with the highest water rate.
The maximum number of characters in a well name has been increased from 16 to 40. The
group name number of characters has not been changed from 16.
User's Guide IMEX

*PTUBE *CONDENSATE
This keyword enables the use of condensate tubing head pressure tables.
These tables can be used with the GASWATER_WITH_CONDENSATE
option.
User's Guide IMEX

There are two major additions to the well data for this release: The introduction of reporting
groups (the *REPORTING-GROUP keyword) and the addition of a bottom-hole fluid (BHF)
group production rate target and monitor.
The purpose of the reporting groups is to allow reporting of results for well aggregates which
are very flexibly defined, without the restrictions which apply to the hierarchical groups
through which group controls are imposed.
*REPORTING-GROUP reporting_group_name well_list weight_list
*REPORTING-GROUP allows the user to define a set of wells with
differing membership weights. No Group Control can be specified for a
reporting group, but there are no restrictions placed on well membership in
reporting groups. The weight_list allows the reporting group to handle
element of symmetry output.
The BHF group target allows group production strategies to be set during prediction runs
which may better allow reservoir pressure to be maintained while respecting maximum rate
limits upon injectors when used in combination with the injection voidage replacement target.
*GCONP
*MAX
*BHF
value
action
*GCONP
*TARGET
*BHF
value
action
*GCONP *MAX *BHF and *GCONP *TARGET *BHF : allow the

user to control bottom-hole fluid group production rates.
*GCONM
*MINBHF
value
( *STOP | *SHUTALL)
*GCONM *MINBHF: allows the user to monitor group bottom-hole fluid

group production rates.
*GUIDEP
*BHF
group_or_well_list
guide_rates
*GUIDEP *BHF: allows the user to specify guide rates proportional to

bottom-hole fluid rates
User's Guide IMEX
There is also a minor keyword addition in this section. The keyword *NWFHYS has been
added to give users extra control over some aspects of the numerical method used to compute
the bottom-hole pressure which yields a specified tubing head pressure. In nearly all cases
the default will work well and the user need not be concerned with this keyword. Please see
the relevant manual pages for more detail.
*NWHYFS
well_list
Integer_list
User's Guide IMEX

*GEOA and *KHA
The anisotropic Peaceman well model has been added to the simulator. This
model accounts automatically for permeability anisotropy and gridblock
aspect ratio. It is available on the well model based well index calculations
(*GEO => *GEOA, and *KH => *KHA). The geometry keyword remains
as is, but the geometric factor C (normally 0.37) is ignored as it is calculated
internally.
This feature is especially useful when used with the *LAYERXYZ and
*LAYERIJK deviated well options.
*LAYERIJK
The new *LAYERIJK keyword allows the user to override, on a layer basis,
the well direction input on the *GEOMETRY keyword.
The *LAYERXYZ allows the user to enter trajectory information into the
simulator through Builder. In addition, *LAYERXYZ information can be
synthesized from Builder without requiring trajectory data.
However, for simple well trajectories, the *LAYERIJK keyword allows the
user to model some aspects of well deviation without using builder or
requiring trajectory files.
*GMKUPTO *WMKUPTO *SMKUPTO
*GMKMAXR *WMKMAXR *SMKMAXR
*GRECYMAXR *WRECYMAXR *SRECYMAXR
*GPRODGROUP*WPRODGROUP
*SPRODGROUP *GSALESFRAC *GSALESMAXR
*GFUELFRAC *GFUELMAXR
Enhancements to the group recycling options have been made in this version
of IMEX. In addition to reducing the amount of gas injected due to fuel and
sales gas, options to separately specify maximum recycling rates and total
injection rates has been implemented for gas, water, and solvent recycling.
See the *GCONI keyword in this manual as it describes in detail the recycling calculation. In
addition, see the manual entries describing the recycling options listed above.
*BHPDEPTH *BHPGRAD *LAYERGRAD
These new keywords allow the user to specify a depth to which a wells
bottom hole pressure is referred to.
The use of quotes in well names is now supported in a consistent manner. To use a double
quote in a well name, use a single quote as a well string delimiter. To use a single quote in a
well name, use a double quote as a well string delimiter.
User's Guide IMEX

IMEX now allows the user to perforate null blocks. Previous versions of IMEX stopped and
reported errors when perforations went though null blocks. This new feature allows
perforations in null blocks, the perforations are closed automatically (*NULL-PREF). This
feature is on by default.
IMEX allows the user to enter the layer X, Y and Z co-ordinates of the entrance and exit
location of a well perforation running through a grid block. With this information IMEX will
calculate an accurate deviated well productivity/injectivity index (*LAYERXYZ).
The *KRPERF keyword allows the user more flexible control of perforation relative
permeability arrays.
The *GCONP and *GCONM keywords now allow the user to shut all wells in a group when
group constraints are violated.
Well reference layers can be defined anywhere in a branching well.
The *PERF keyword now can handle all refined grid situations including hybrid grids. The
*PERFRG, *PERFVHY and *PERFHY keywords are valid but no longer required.
The *FULLY-MIXED wellbore crossflow model is now the default for all wells.
User's Guide IMEX

IMEX includes a significant modification to the well module simulator interface. This
modification allows the well module better access to simulator information. This in turn
allows the well module to more accurately calculate well and group operating constraints.
As a consequence of the above, the *CONT *REPEAT subkeywords on operating and
monitor constraints are now operational.
In addition, the well gravitational head calculation has been moved to a position within the
code which ensures the well module and the simulator use identical heads.
The well index printout (available when using *WPRN *WELL *ALL) has been greatly
enhanced. In addition to well index, the actual layer and total well fluid
productivity/injectivity is displayed.
As well as being able to display well productivity the user is now able do directly input fluid
phase or total well productivity/injectivity or productivity/injectivity multipliers using the
*SETPI keyword listed below.
*SETPI
SETPI can be used to alter the well productivity to measured values or by a
multiplicative factor correcting the last (*MULT) or the fundamental
(*MULTO) productivity.
*SETPI
(*OIL)
well_list
(*WAT)
(*GAS)
(*LIQ)
(*TOT)
(*MULT)
(*MULTO)
Each time a *SETPI keyword is read, wells on the well list have their
fundamental productivity corrected to match the user input value. The
correction is only applied when the *SETPI keyword is read (i.e. when the
test was done). Well productivity will normally change as fluids move in the
vicinity of the well. However, productivities can be corrected at any later
time by using the *SETPI keyword again.
User's Guide IMEX

For the 98.05 release the interface between the simulator and well management module has been
improved so that the action of repeating a timestep in response to a well constraint violation works
more smoothly than in past releases. This is now the strategy of choice if, for example, it is
extremely important that a certain well's bottom-hole pressure never fall below a certain value.
Since certain constraint checks and constraint actions may be handled slightly differently in the
new version of IMEX, some data sets may run differently than with earlier versions. The user
should examine output carefully when comparing runs made with earlier versions of IMEX.
The default mode of the WHP well constraint is now *IMPLICIT instead of the former
default of *INITIALIZE; IMEX users have for the most part preferred the *IMPLICIT mode.
Please see the documentation for the *OPERATE keyword for more information.
Two new well-related keywords have been introduced:
*WORKOVER-DATA
This keyword is intended to allow the user more direct control over the parameters
determining the workover action which can be imposed in response to certain well
monitor violations. It is intended to replace the entry of these parameters after the
*WORKOVER action subkeyword of the *MONITOR keyword, which in the
past was the only way to enter these parameters; the old way, however, is still
supported and documented. Please see the manual pages for the *MONITOR and
the *WORKOVER-DATA keywords for more information.
*MRC-RESET
This keyword allows the user to specify on a well- by-well basis whether a well
should be tested for and set to operate upon its most restrictive constraint after a
constraint value has been changed, or whether it should continue with its current
operating constraint and switch only if a constraint is violated at the end of a
timestep. The default is to check and switch to the most restrictive constraint
whenever a constraint value is changed through data (through the *OPERATE,
*ALTER, or *TARGET keywords). This can save the simulator the sometimes
difficult task of trying to cause a timestep to converge with an unrealistic
constraint; however the option of not resetting is allowed in case that the user
wishes to enforce operation upon a certain constraint for the timestep. Please see
the manual page for the *MRC-RESET keyword for more information.
User's Guide IMEX

*RANGE and *REFINE
The capability to define new refined grids in the Recurrent Data section has
been added to IMEX. Thus the keywords *RANGE and *REFINE may
appear in the well (recurrent) data. Please refer to the Reservoir Description
section for information on the usage of these cards.
*PERFRG
The ability to perforate in multilevel refined blocks has been added to the
*PERFRG keyword.
The *PERFRG card now can handle perfs which exist at in layers at different
levels of refinement. For example the following is now possible.
Fundamental
*PERFRG
Refine1
(5x5x1)
j1
Refine2
(3x3x1)
i
2
j
2
k
2
i1
k1
i2
j2
Well Index
k2
1.0
1.0
2
1.0
In addition, wells which are accidentally perforated in pinched out or nulled

out blocks are now flagged in a more informative manner. A list of valid
perfs (with i,j,k locations) is presented to the user, along with a separate list
for invalid perfs. It is now a simple process to produce a valid perf list.
*WELLINIT
The *WELLINIT keyword can now also be defined on a well by well basis.
In addition to the previous usage,
*WELLINIT
(*ITER)
(*CHANGE)
(*TIMESTEP)
the format,
*WELLINIT
well_list
(*ITER)
(*CHANGE)
(*TIMESTEP)
can be used.
*ALTER, *TARGET, *GUIDEI, *GUIDEP, *GLIFT, *ON-TIME, *ALTERCP
These keywords listed allow the user a simpler method for assigning all
values in a well list to one value. Previous versions of IMEX allowed the
user to employ well lists with wildcarding, but values (rates, etc) assigned to
each well had to be entered explicitly (even if one value is used for all wells).
User's Guide IMEX
Example:
*ALTER 'PRODNW*'
14*1000.0
<----- Input List (14 entries)
Thus the number of wells in the wildcard list had to be known, making input
more time consuming.
If the user now specifies a single value in the list input IMEX will now set all
wells in the list to this value. This applies to all types of well lists (not just
those using wildcards).
*ALTER 'PRODNW*'
1000.0
<----- single entry, all wells

in list alter primary
constraint to 1000.0
It is not necessary to know the number of wells in the well list to apply the
single value to the list.
User's Guide IMEX

Multilateral Well Option
*FLOW-FROM
*FLOW-TO
*CLOSED
*REFLAYER
It is now possible to specify well trajectories for multilateral wells under the
*PERF or *PERFRG keywords. For each perforation, following the
location, relative permeability and status information, the optional keyword
*FLOW-TO for producers or *FLOW-FROM for injectors can be used to
assign "parent-child" relationships between perforation layers. This
information serves to define the well trajectory. In addition to open or
automatic status designation for a perforation, a third choice of closed is
introduced with the keyword *CLOSED. Unlike the two previous choices a
"perforation" with a closed status has no flow associated. Finally the
keyword *REFLAYER may be used to designate the reference layer for
bottom hole calculations. Please see the pages for *PERF and *PERFRG for
more information.
Wildcards in Well Lists
A significant new feature has been added to all keywords which acts upon well lists (e.g.
*ALTER 'well1' 'well2'). Wild cards have been incorporated in the following manner.
* replaces any number of characters at the end of a well name or can be used on its own to
represent all wells (e.g. *ALTER '*' or *ALTER 'wel*').
Example: if the wells in the model are 'well1' 'well2' and 'prod1', then, '*' would list all 3, 'p*'
would list 'prod1'. '?ell?' would list 'well1' and 'well2'.
Cross Flow Model
*SOURCE-SINK
*ZERO-FLOW
*FULLY-MIXED
A fully featured implicit cross flow model has been added into IMEX 97.00.
This model when used in *FULLY-MIXED mode allows the user to model
cross flow in a well. *ZERO-FLOW allows the user to shut in backflowing
layers (the default), while *SOURCE-SINK reverts the model to the IMEX
96.00 default which does not properly handle cross flow or backflow.
*XFLOW-MODEL well list
User's Guide IMEX
Misc. Options
*GRAVITY
*ZERO-HEAD
In flowing wells it is appropriate to calculate the density of the fluids in the well
bore based on mobilities of fluids in the well blocks. However when a well is
being pumped, except for a small amount of liquid collected at the bottom of
the well, the head in the well is effectively zero (the head of gas). For this case,
we offer the user the *ZERO-HEAD option. The default *GRAVITY option
calculates well head as was previously done in IMEX 96.00.
*HEAD-METHOD well list
*WGR
*NODAMP
*DAMP
*PHASE2
The ability to use water gas ratio as a penalty has been added along with
different methods for calculating the penalty. The *PHASE2 most accurately
honors the penalty rate but the *NODAMP and *DAMP methods may more
accurately mimic field operations.
*OPERATE *PENALTY
Mobility Weighted Well Pressures
IMEX 97.00 will print out mobility weighted well pressures to aid in history matching. In
order to obtain this report, the user must request detailed well and layer printout.
Autodrilling of Wells
In the 96.00 release the automatic opening of wells possessing the *AUTODRILL status was
triggered as an action taken when a group minimum oil or minimum gas constraint was
violated. This had the unfortunate effect of pre-empting the use, in these situations, of
minimum oil or gas monitors to terminate the simulation. In the 97.00 version, the opening
of wells with *AUTODRILL status is triggered when *AUTODRILL *ON is specified for a
group or groups using the *GAPPOR (group apportionment) keyword and a group listed fails
to meet a target during the target apportionment. The *AUTO-OPEN action for the
*GCONM keyword is no longer supported; this keyword structure must be converted by the
user to the *GAPPOR form. Please see the discussion of the *GAPPOR, *AUTODRILL,
and *DRILLQ keywords for more information.
Guide Rate Apportionment
In earlier releases of IMEX when guide rates were specified to determine target rate
apportionment, production targets other than oil rates were not properly handled. This has been
corrected in version 97.00; guide rate apportionment is done correctly for all allowed target rates.
Transfer of Cumulatives in Hierarchy Redefinition
In earlier releases if a group structure was introduced after a model had been run for some
time with no group structure defined, the FIELD cumulative was reset to zero when the
FIELD became part of the group hierarchy. In version 97.00, when the group structure is
redefined, the newly defined top-level group acquires the cumulative production of the old
top-level group.
User's Guide IMEX
Identification of Reference Layer after Reperforation
In earlier releases when a multilayer was reperforated in such a way that some layers were
reperforated, some eliminated, and some new layers were added, the reference layer (the one
in which the well's bottom-hole pressure is defined) was chosen as the layer first perforated
historically, regardless of its position in the current set of perforated layers. For example, if a
well were first perforated in layers 1, 3, and 4, in that order, and later reperforated in layers 2,
3, and 4, earlier releases would pick layer 3 as the reference layer. Version 97.00 always
picks the first layer perforated in the most recent set of perforation lines as the reference
layer; in the example, version 97.00 would pick layer 2 as the reference layer. Please see the
discussion of the *PERF keyword for more information.
User's Guide IMEX

Water-gas ratios have been added as a quantity which can be monitored both for individual
wells and for groups. This is especially useful for well control in gas fields. Please see the
entries for the *WGR subkeyword under *MONITOR and *GCONM in this chapter for more
information. Also, the water-gas ratio has been added to the gas-oil ratio and the water-oil ratio
as quantities upon which penalty operating constraints can be based. In addition, flexibility has
been added in the ways in which penalty constraints can be imposed with the introduction of the
*NODAMP, *DAMP, and *PHASE2 mode subkeywords of the *OPERATE *PENALTY
keyword. Please see the pages for *OPERATE in this chapter for more information.
User's Guide IMEX

The new ON-TIME keyword allows the input of fractions for all wells or selected wells
which specify the fraction of the simulated time during which the well is on-line and actually
in operation. If a well is assigned an on-time fraction less than one, the rates reported (the
"averaged" rates) are less than the rates which apply when the well is actually in operation
(the "instantaneous" rates). The user should enter well constraints (such as BHP, WHP and
rate constraints) which correspond to INSTANTANEOUS rates; reported BHP and WHP
values will correspond to the instantaneous rates. Reported rates and cumulative volumes
will, however, be the averaged values. If the *ON-TIME keyword does not appear in the data
set, the default value of one is applied for the on-time fraction and the simulation is
unaffected by the introduction of the option.
It is now possible to enter maximum BHP and WHP values as monitored constraints for
producing wells; please see the entry for the *MONITOR keyword.
It is also now possible to specify maximum lift gas rates for wells under the *GLIFT *OPT
option; please see the *GLIFT entry for details.
When entering wellbore hydraulics tables under the *PTUBE keyword, it is now possible to
direct the simulator to supply missing values by interpolation or extrapolation from other
values in the same row of the table by entering *EXTP in the place which the missing number
would normally occupy. Please see the *PTUBE entry in this manual for more details.
User's Guide IMEX
Old Options No Longer Supported by IMEX 97.00

*MONITOR *BACKFLOW *AUTOLAYER for injectors is no longer valid. Use the default
*XFLOW-MODEL *ZERO-FLOW to accomplish this function.
In addition, we recommend that *MONITOR *BACKFLOW *AUTOLAYER not be used for
producers, instead the XFLOW-MODEL should be used.
The *AUTO-OPEN action for the *MINOIL and *MINGAS monitors under the *GCONM
keyword is no longer supported; this keyword structure must be converted by the user to the
*GAPPOR form. Please see the discussion of the *GAPPOR, *AUTODRILL, and
*DRILLQ keywords for more information.
User's Guide IMEX
Well Management and Group Control

There is no limitation on the maximum of group levels in the group hierarchy; i.e., the
hierarchy can consist of one top-level group, second-level groups connected to the top-level
group, and third-level groups connected to second-level groups, and so on. Wells or groups
may be connected to a second-level or lower-level group, but not a mixture of the two.
An example is shown below:
level 1
'FIELD'
level 2
'GNAME-2'
'GNAME-1'
'Default-Group'
'GNAME-3'
level 3
'GNAME-4'
'W4'
'W5'
'GNAME-5'
'W6'
'W7'
'GNAME-6'
'W8'
'W9'
'GNAME-7'
'W10'
'WELL-1'
'WELL-2'
'WELL-3'
'W11'
The highest level group is the 'FIELD'. The highest level is not optional. If *GROUP lines
appear and either no top-level group is specified
Example:
*GROUP 'G1' *ATTACHTO 'G2'
or more than one top-level group is specified

Example:
then an error is generated and simulation terminates.

Wells can be attached to any group except the field. A group to which a well is attached can
have only wells attached to it, and not other groups. Wells and groups cannot be attached to
the same group.
The group control hierarchy is optional. Thus all IMEX data sets from previous versions that
do not have gathering centers can be run unchanged with this version of IMEX. If a group
control hierarchy is used, then not all wells need to be attached explicitly to groups. Those
wells that are not attached to a group by the user will be automatically attached to the
internally-generated group with the name 'Default-Group'.
User's Guide IMEX
An example of an invalid well-management hierarchy is given below:

'FIELD'
'GNAME-4'
'GNAME-5'
'W4'
'W6'
'W5'
User's Guide IMEX
'Default-Group'
'GNAME-2'
'GNAME-1'
'W7'
'GNAME-6'
'W8'
'W9'
'GNAME-7'
'W10'
'W11'
'GNAME-8'
'WELL-2'
'GNAME-3'
'WELL-3'
This is invalid - group and wells

attached to same group.
Specifying the Well and Group Control Hierarchy

The control hierarchy for wells and groups is constructed using the *GROUP and *WELL
keywords. Group controls (injection and production targets) and monitored constraints are
specified by the *GCONP, *GCONI and *GCONM keywords. The injection and production
distribution to wells and groups is specified by using the *GUIDEP and *GUIDEI keywords.
The most offending well can be shut by using the *SHUTIN action under the *GCONM
keyword.
Wells can be drilled automatically to maintain production or injection targets by specifying their
initial status as *AUTODRILL and by specifying *GAPPOR 'group' *AUTODRILL *ON.
Well operating and monitored constraints are specified by using the *OPERATE and
*MONITOR keywords.
Well layers can be initially either specified as open or candidates for recompletion by using the
*OPEN or *AUTO subkeywords under the *PERF, *PERFV, or the PERFVHY keywords.
For describing more complex well trajectories such as multilateral wells, new connection
keywords, FLOW-TO (producers) and *FLOW-FROM (injectors) are now supported under the
*PERF keyword. In addition a layer may be assigned a status of "*CLOSED" to indicate that
the layer is introduced purely to define the well trajectory. Such a layer is not perforated and
cannot produce or inject any fluids. Greater detail and examples are provided in this manual
under the *PERF keyword.
Well injectivities or productivities can be automatically improved by specifying the
*WORKOVER action under the *MONITOR keyword. Other possible actions under the
*MONITOR keyword are to plug the most offending layer (*SHUTLAYER), monitor the
layer for possible reopening (*AUTOLAYER), and monitoring the well for possible
reopening when the GOR or water-cut improves (*AUTOWELL).
Please note that it is possible to accidentally overwrite group properties, constraints and
actions. When all the groups have been defined using *GROUP cards and all group actions,
properties and constraints have been specified using card such as *GCONI, *GCONP and
*GCONM. A further specification of any previous or new group using a *GROUP keyword
requires the user to completely respecify all group actions, properties and constraints.
If the group actions, properties and controls are not respecified, they are lost.
The same is not true for moving wells between groups by using the *ATTACHTO keyword.
This can be done without group action respecification.
Examples:
The first example below has a new group defined after group 1 actions are specified.
Group 1 controls would be lost.
The second example shows the proper way of specifying multiple groups.
The third example below has a new group defined after group 1 actions are specified, but the
lost group controls are respecified.
User's Guide IMEX
Example 1: Group 1 Actions lost

*GROUP 'Group1' *ATTACHTO 'Field'
*GCONP 'Group1'
*MAX
*STG 100000.0
*TARGET *STO
1000.0
*IPP
*GCONP 'Group2'
*MAX
*STG 110000.0
*TARGET *STO
900.0
*IPP
However in the cases below all group controls would be maintained

Example 2: Group definitions grouped together followed by all group control specifications.
*GCONP 'Group1'
*MAX
*STG 100000.0
*TARGET *STO
1000.0
*IPP
*GCONP 'Group2'
*MAX
*STG 110000.0
*TARGET *STO
900.0
*IPP
Example 3: Group 1 group control definitions lost and then respecified.

*GCONP 'Group1'
*MAX
*STG 100000.0
*TARGET *STO
1000.0
*IPP
*GCONP 'Group1'
*MAX
*STG 100000.0
*TARGET *STO
1000.0
*IPP
*GCONP 'Group2'
*MAX
*STG 110000.0
*TARGET *STO
900.0
*IPP
User's Guide IMEX
Introducing Group Control after Start of Simulation

Group control can be instituted after the simulation has started, at a well change time or upon
restart. The simplest example consists in introducing a field target after the simulation has
run for some time. Assume that a run has been started with no group structure referred to, i.e.
with wells defined using *WELL lines with no *ATTACHTO subkeywords defining parent
groups. To establish a group structure, it suffices (at a well change time) to introduce the
single line
*GROUP 'Default-Group' *ATTACHTO 'Field'
The name 'Default-Group' must appear exactly as shown in the above line, since it is an
internally set name. The top-level group is given the name 'Field' in this example but the user
is completely free to choose this name (up to a maximum length of 16 characters); 'Campo' or
'FIELD' would have exactly the same effect. Then to introduce a field target, it suffices to
follow the above line with the line
*GCONP 'Field' *TARGET *STO 400.0
Here it is important that 'Field' match exactly the character string after *ATTACHTO in the
*GROUP line; if 'Campo' had been used above it would have to be used here.
User's Guide IMEX
Data Incompatibilities Between 93.00 and Previous Versions

of IMEX
Although most of the well data are compatible with previous versions of IMEX, there are
some incompatibilities.
1. For the pseudo-miscible option (solvent injection) the previous versions of IMEX
monitored the gas+solvent-oil ratio when *MONITOR *GOR was used. In this
version of IMEX *MONITOR *GOR refers to gas-oil ratio only. If it is desired to
monitor the gas+solvent-oil ratio the *MONITOR *GSOR should be used.
2. Minimum rate operating constraints have been replaced by minimum rate
monitoring constraints. If a minimum rate well constraint is specified, it is
converted automatically to a monitoring constraint and a warning message is
printed.
3. Well names now have to be unique i.e. it is no longer possible to have two or more
wells with the same name. This is because wells can now be referred to by their
names only. Well numbers are optional.
4. *PERFV can no longer be used to perforate wells located in hybrid grids. The
*PERFVHY or *PERFRG keywords should be used to perforate wells located in
hybrid grid blocks.
5. The *GATHER keyword and the related keywords *PTARG and *ITARG are now
obsolete and cannot be used. The *GROUP and *GCONP, *GCONI and
*GCONM keywords must be used instead.
User's Guide IMEX
Limitations of the Well Management and Group Control

Module
The following limitations currently apply to the well management and group control module.
1. The topmost group level (the field) cannot have wells attached to it, but only other
groups.
2. Groups to which wells are attached cannot have other groups attached to them.
3. New wells may be attached to a group at any time; however, a well cannot be
attached to more than one group at a time. Redefining a well's parent group
automatically detaches it from the earlier parent.
4. Group controlled injection fluids are limited to gas, water and solvent only. Oil
injection is not supported.
5. The well management module can automatically shut in and reopen well layers
when GOR or WCUT exceed a certain limit. When this option is used the layers
are sorted according to their depths, in order to open / close layers. Thus if the
well is perforated horizontally, the behavior of this option may be unpredictable.
6. Please see the *GROUP Keyword for further limitations
User's Guide IMEX
Recurrent Data from Other Sections

There are a few keywords from the other sections whose parameters may be changed during
the course of a simulation run. These keywords must be positioned after a *DATE or *TIME
keyword in the Well Data section. At a given time or date the older or existent parameters
will be overridden by the new ones.
From the Input/Output Control section, these keywords may be reassigned:
a) *WRST
b) *WPRN
c) *OUTPRN
d) *WSRF and
e) *OUTSRF
From the Reservoir Description section, the refined grid definition keywords and the
transmissibility modifiers may appear in the Well Data section. It is critical that grid
definition data be grouped together within recurrent data (i.e. not interspersed with well data),
failure to do so may cause unpredictable results:
a) *RANGE
b) *REFINE
c) *TRANSI (*TRANLI)
d) *TRANSJ (*TRANLJ)
e) *TRANSK (*TRANLK)
f) *PERMI
g) *PERMJ
h) *PERMK
From the Rock-Fluid Property section, only the rock type array and the end-point arrays may
appear in the Well Data section:
a) *RTYPE
b) *SWCON
c) *SWCRIT
d) *SORW
e) *SORMAX
f) *SORG
g) *SOIRW
h) *SLCON
i)
*SGCON
j)
*SGCRIT
User's Guide IMEX
k) *KRWIRO
l)
*KROCW
m) *KRGCL
n) *KROGCG
o) *PCWMAX
p) *PCGMAX
q) *JFWMAX
r) *JFGMAX
s) *NDARCYCOR
From the Numerical Methods Control section, these keywords may be changed:
a) *DTMAX and
b) *DTMIN
User's Guide IMEX
Well and Recurrent Data Identifier (Required)
*RUN
PURPOSE:
*RUN identifies the beginning of all well and recurrent data keywords.
FORMAT:
*RUN
DEFAULTS:

CONDITIONS:
The Well and Recurrent Data keyword group follows the Numerical keyword group in the
data file. It is the last keyword group in the input-data-file.
User's Guide IMEX
Well Change Date (Conditional)
*DATE
PURPOSE:
*DATE indicates that the well change will occur at a specified date.
FORMAT:
*DATE yyyy mm dd
DEFINITIONS:
yyyy
An integer specifying the well change year.
mm
An integer specifying the well change month.
dd
A real number specifying the well change day.
DEFAULTS:
Conditional keyword. Defaults for starting date only are:

yyyy:
mm:
dd:
0000
1
1.0
CONDITIONS:
This keyword must be located in the Well and Recurrent Data keyword group. Either
*DATE or *TIME is required.
EXPLANATION:
To indicate when well changes occur you must use this keyword or the *TIME keyword. If
this keyword does not appear immediately following the *RUN keyword, a starting date of
0000 01 01 for the simulation is assumed.
The keyword *DATE is recommended immediately after *RUN.
All well changes are assumed to occur according to the fraction of the day given.
For example, a well change date of August 19, 1988, at noon, is entered as:
*DATE 1988 08 19.5
All well changes and other recurrent data changes which are specified by keywords located
between one *TIME or *DATE keyword and the next *TIME or *DATE keyword are
applied at the time of the first *TIME or *DATE keyword.
User's Guide IMEX
Well Change Time (Conditional)
*TIME
PURPOSE:
*TIME indicates that the well change will occur at a specified time.
FORMAT:
*TIME num
DEFINITIONS:
num
A real number specifying the length of time after the simulation starting date
that the well will be changed (days | days | mins | days).
DEFAULTS:
Conditional keyword. Default for starting time only is: 0.0

CONDITIONS:
This keyword must be located in the Well and Recurrent Data keyword group. Either
*DATE or *TIME is required.
EXPLANATION:
To indicate well changes you must use this keyword or the *DATE keyword. If this keyword
and the *DATE keyword do not appear immediately following the *RUN keyword, the date
0000 01 01 and the time of 0.0 are assumed.
All well changes and other recurrent data changes which are specified by keywords located
between one *TIME or *DATE keyword and the next *TIME or *DATE keyword are
applied at the time of the first *TIME or *DATE keyword.
Example:
*PRODUCER 2
** At the following time, change the
** primary operating constraint.
*TIME 20.0
*ALTER 2
500.0
** The following *TIME forces a simulation
** timestep at 365.0 days to force
** printout and plot information at this
** time.
*TIME 365.0
** Simulation stops at 900.0 days.
*TIME 900.0
*STOP
User's Guide IMEX
Well Change First Time-Step Size (Optional)
*DTWELL
PURPOSE:
*DTWELL identifies the first time-step size to be used immediately after the current well
change.
FORMAT:
*DTWELL time_size
DEFINITIONS:
time_size
A real number to specify the time-step size (days | days | min | days).
DEFAULTS:
Optional. Default is 1.0 days for Cartesian or variable depth/variable thickness grid. Default
is 0.025 days for radial grids.
CONDITIONS:
This keyword must be located in the Well and Recurrent Data keyword group. *DTWELL
must be larger than *DTMIN.
EXPLANATION:
This keyword is required to specify the size of the first timestep in the simulation, and may
optionally appear for subsequent well changes.
This keyword is part of recurrent data, and may be changed during the simulation run.
User's Guide IMEX
Maximum Time-Step Size (Optional)
*DTMAX
PURPOSE:
*DTMAX identifies the maximum time-step size.

FORMAT:
*DTMAX max_time_size
DEFINITIONS:
max_time_size
A real number to specify the maximum time-step size allowed (days | days |
min | days).
DEFAULTS:
Optional keyword. Default: The maximum time-step size is 365 days.

CONDITIONS:
This keyword may be located in the Well and Recurrent Data keyword group, and may also
be located as part of Numerical Methods Control.
EXPLANATION:
If the time-step size calculated by the automatic time-step selector is larger than
max_time_size, it is set to max_time_size.
The time-step size is always automatically adjusted so that a timestep will end exactly at the
time specified by a *TIME or *DATE keyword in recurrent data.
If the maximum time-step size is once defined by a *DTMAX keyword, the maximum timestep size remains the same until changed by a subsequent *DTMAX keyword.
The acceptable range of values for maximum time-step size is:
min
max
User's Guide IMEX
SI
Days
1.0E-10
1.0E+20
Field
Days
1.0E-10
1.0E+20
Lab
Min
1.44E-07
1.44E+23
Minimum Time-Step Size (Optional)
*DTMIN
PURPOSE:
*DTMIN identifies the minimum time-step size.

FORMAT:
*DTMIN min_time_size
DEFINITIONS:
min_time_size
A real number to specify the minimum time-step size allowed (days | days |
mins | days).
DEFAULTS:
Optional keyword. Default: The minimum time-step size is 0.001 days

CONDITIONS:
This keyword may be located in the Well and Recurrent Data keyword group, and may also
be located as part of Numerical Methods Control.
EXPLANATION:
If the automatic time-step selector cuts the time-step size to smaller than min_time_size
specified by *DTMIN, the time-step size is set to min_time_size. If the timestep does not
converge, the simulation is stopped.
If the minimum time-step size is once defined by a *DTMIN keyword, the minimum timestep size remains the same until changed by a subsequent *DTMIN keyword.
*DTMIN must be smaller than *DTWELL.
The acceptable range of values for minimum time-step size is:
min
max
SI
Days
1.0E-10
1.0E+20
Field
Days
1.0E-10
1.0E+20
Lab
Min
1.44E-07
1.44E+23
User's Guide IMEX
Setting Grid Blocks to Implicit or Explicit (Optional)
*AIMSET
PURPOSE:
*AIMSET initializes grid blocks to implicit or explicit formulations.

ARRAY:
*AIMSET
DEFINITIONS:
values
A number indicating a given block is explicit or implicit:
0 explicit
1 implicit
DEFAULTS:
Optional keyword. Default: 0 (explicit)

CONDITIONS:
This keyword must be located in the Well and Recurrent Data keyword group.
EXPLANATION:
This keyword sets blocks to an initial explicit or implicit pattern. After the blocks are set, the
active well blocks are set to implicit. Optionally the neighbours or the neighbours plus their
neighbours (see diagram in *AIMWELL) are set to implicit by using *AIMWELL *WELLN
(or *WELLNN).
If the keyword is left out at subsequent *DATE or *TIME keywords, then the implicitness of
the blocks does not change except for those well blocks which are beginning operation.
These blocks are set to implicit.
For example,
*AIMSET *CON 0
sets all blocks to explicit (except for the active well blocks). This is the default.
During a simulation run, the default would result in blocks being switched to implicit as required.
*AIMSET *CON 1
sets all blocks to implicit.

The *AIMSET keyword is applied first. If either keyword strings, *AIMWELL *WELLN or
*AIMWELL *WELLNN, appear then these are used next.
In general *AIMSET and *AIMWELL should be omitted. If you are doing a field study and
wish to adjust only the rates as part of a history match, then omit the keyword. But on the first
well opening time use a small time-step size (*DTWELL) such as 1.0 days. For subsequent
well changes, time-step sizes, typically as large or larger than thirty days, may be used.
User's Guide IMEX
If you are modelling a process where you inject into the well and then produce from it, then a
small time-step size should be used at the start of each cycle. Then the combination *AIMSET
*CON (with, optionally, *AIMWELL) may be used to set most of the blocks to explicit at the
beginning of each cycle, in order to reduce execution time.
User's Guide IMEX
Setting Well Blocks and Neighbours to Implicit (Optional)

*AIMWELL
PURPOSE:
*AIMWELL is used to set blocks near wells to an implicit formulation.

FORMAT:
*AIMWELL
(*WELLN)
(*WELLNN)
DEFINITIONS:
*WELLN
This subkeyword sets well blocks and neighbours of active well blocks and
well blocks to implicit.
*WELLNN
This subkeyword sets well blocks, the neighbours of active well blocks and
the neighbours of the neighbours to implicit.
DEFAULTS:
Optional keyword. Default: Only well blocks are set to implicit.

CONDITIONS:
EXPLANATION:
Observing the diagram, *AIMWELL on its own sets the block marked W to implicit. W
denotes the well block.
Observing the diagram, *AIMWELL *WELLN sets the blocks marked W and N to implicit.
W denotes the well block. N denotes neighbour blocks.
Observing the diagram, *AIMWELL *WELLNN sets the blocks marked W,N and NN to
implicit. W denotes the well block. N denotes neighbour blocks and NN denotes neighbours
of neighbours.
NN
J=5
J=4
J=3
NN
J=2
NN
NN
NN
NN
NN
NN
J=1
I=1
User's Guide IMEX
I=2
I=3
I=4
I=5
In three dimensional systems, the neighbouring grid blocks in the third direction are also set
to implicit.
The *AIMSET keyword is applied first. If either keyword strings, *AIMWELL *WELLN or
*AIMWELL *WELLNN, appear then these are used next.
In general *AIMSET should be omitted (i.e. defaults to *CON 0). For all but the largest field
studies, where the number of total well perforations rival the number of grid blocks,
*AIMWELL *WELLNN should be used.
User's Guide IMEX
Set Frequency of Initialization of Bottom-Hole Pressure

(Optional)
*WELLINIT
PURPOSE:
*WELLINIT allows the user to specify that bottom hole pressure values for all, or for specified,
wells running on rate, drawdown, or implicitly imposed well head pressure constraints should
be reinitialized before every timestep or before every Newtonian iteration. Both particular well
initialization frequencies, which are entered for a particular well or set of wells named in a list,
and a global initialization frequency, which applies to wells for which no particular frequency
has been set, can be set using *WELLINIT.
FORMAT:
1. To set the global well initialization frequency, i.e. to apply to all wells, enter
*WELLINIT
(*CHANGE)
(*TIMESTEP)
(*ITER)
2. To set all wells in a user-chosen list to a particular well initialization frequency,

enter
*WELLINIT well_list
(*CHANGE)
(*TIMESTEP)
(*ITER)
DEFINITIONS:
CHANGE
Indicates that bottom hole pressures of rate or implicit well head pressureconstrained wells are to be reinitialized only after significant changes in well
operating conditions.
TIMESTEP
Indicates that bottom hole pressures of rate- or implicit well head pressureconstrained wells are to be reinitialized at the beginning of each timestep.
ITER
Indicates that bottom hole pressures of rate- or implicit well head pressureconstrained wells are to be reinitialized before each Newtonian iteration.
well_list
A set of 'well_names' or well_numbers; see below. The presence or absence
of the well_list identifies to the simulator whether the global initialization
frequency is being set or particular frequencies are being set.
User's Guide IMEX
well_names
Any number of well names (in quotes) to specify the wells to which this
alteration of initialization frequency applies. Limited wildcarding is
available for the list of well names; please see the manual page for the
*SHUTIN keyword for an explanation of the wildcard facility.
well_numbers
Any number of integers, or ranges of integers to specify the well numbers to
which this alteration of initialization frequency applies.
DEFAULTS:
Optional keyword. If *WELLINIT does not appear in the data set, *WELLINIT *ITER is the
default global well initialization frequency in IMEX.
The global initialization frequency may be reset by entering *WELLINIT in the first format
above without the well list; the global frequency may be overridden for particular wells by
using the second format above with the well list.
If *WELLINIT appears alone on a line with no following subkeyword or well list, then
*WELLINIT *ITER is set as the global well initialization frequency.
CONDITIONS:
If it appears, this keyword must be located in the Well and Recurrent Data keyword group.
It must appear AFTER (but not necessarily immediately after) the first *DATE line. If a well
list is included in the *WELLINIT line, then the *WELLINIT line must follow all of the
*WELL lines which define the wells in the list.
EXPLANATION:
*WELLINIT *CHANGE may give a sufficiently accurate initial bottom hole pressure that
the Newtonian iterations for the well constraint equation converge rapidly. In difficult cases,
however, the well equations may converge slowly; in such cases, invoking *WELLINIT
*TIMESTEP or *WELLINIT *ITER may facilitate convergence of the Newtonian iterations.
The *WELLINIT *ITER option requires more CPU time per Newtonian iteration, but may
result in an overall CPU time saving if the number of Newtonian iterations is reduced
significantly, as is frequently the case in typical IMEX runs. *WELLINIT *TIMESTEP is
intermediate between the other two frequencies.
Sometimes only a few wells in a large field require the *WELLINIT *ITER or *WELLINIT
*TIMESTEP treatment; in such cases the global treatment may be set to *CHANGE using
the first format above and the problem wells may be flagged for special treatment by using
the second format above, including a well list.
In IMEX the global initialization frequency applies to all wells.
Initialization frequencies specified for a well under the second format above (i.e. by inclusion of
the well's name or number in a well list following *WELLINIT) are always honored, regardless of
the well's current operating constraint and the current setting of the global initialization frequency.
User's Guide IMEX
Example:
If the *TIMESTEP frequency suffices for most wells in the field, but *CHANGE suffices for
'WELL1' and *ITER is necessary for 'WELL2', the following sequence is appropriate:
*WELLINIT *TIMESTEP
*WELLINIT 'WELL1' *CHANGE
*WELLINIT 'WELL2' *ITER
in the Well and Recurrent Data section.
User's Guide IMEX
Group Identification (Optional)
*GROUP
PURPOSE:
*GROUP is used to identify gathering centers, groups and platforms. This keyword is useful
for building a tree structure of groups.
FORMAT:
*GROUP 'child_1' ... 'child_n' *ATTACHTO 'parent'

DEFINITIONS:
'child_1', child_2 .. , child_n

Names of child groups that are attached to the 'parent' group. Each group is
identified by a unique name up to 16 characters long and enclosed in single
quotes. The *ATTACHTO keyword is not optional and must be present.
*ATTACHTO
Defines the parent group of all groups named in the list following *GROUP.
'parent'
Name of the parent group.
DEFAULTS:
Optional keyword. If no *GROUP line appears in the data, no group structure exists in the
simulation and well rates and cumulatives are summed directly into a field cumulative. This
is reported as the FIELD cumulative in output, but no group called 'FIELD' actually exists in
this case and no group control of production or injection is possible.
When a *GROUP line is encountered in the data a group structure is established, which
always consists of at least two groups : a top-level group and the 'Default-Group'.
The top-level group has no default name; its identity is determined by finding the unique
group which appears in the list of parent groups but not in the list of child groups. If there is
no such group or more than one, an error is generated and simulation terminates.
Wells can only be attached to groups other than the field (top-level) group. Any wells which are
not explicitly associated with a parent group are automatically attached to an internally-created
group which has the name 'Default-Group' and which has the top-level group as its parent group.
For example if no reference is made to group hierarchy in the data, the following conceptual
structure (not truly a group structure) will exist by default:
FIELD
'WELL-1'
'WELL-2'
...
'WELL-n'
User's Guide IMEX
Quotes were purposely omitted around FIELD above to emphasize that no group with the
name 'FIELD' actually exists in this case. No group control is possible in this case.
If the single line
is added to the data, then the following group structure is established :

'Field'
'Default-Group'
'WELL-1'
'WELL-2'
...
'WELL-n'
and group controls could then be imposed upon 'Field'.

CONDITIONS:
EXPLANATION:
This keyword identifies the group, by name, and assigns it to a parent group. There is no
limitation on the levels of groups in the group hierarchy:
1. Top-level. Only one group is allowed at this level; it has no default name and can be
assigned any name not exceeding 16 characters in length by the user. This group
cannot have wells attached directly to it. This group represents the whole field. The
name of this group is entered after *ATTACHTO in a *GROUP line. The top-level
group is identified as the only group whose name appears after *ATTACHTO in at
least one *GROUP line but whose name never appears in a list immediately
following *GROUP in a *GROUP line.
2. Level 2. These groups have the top-level group as their parent. When a group
structure exists, there is always at least one group in this category, with the name
'Default-Group'. 'Default-Group' has connected to it any wells which do not
explicitly have a parent group assigned. Level 2 groups can have either wells or
groups attached to them, but not a mixture of the two; i.e. if a level 2 group is named
after the *ATTACHTO subkeyword in a *WELL line, then that group must not
appear after the *ATTACHTO subkeyword in any *GROUP line, and vice versa.
3. Level n. These groups have level n1 groups as their parents. Level n groups can
have either wells or groups attached to them, but not a mixture of the two.
User's Guide IMEX
Examples of valid and invalid well control trees are given below.
Valid example of a well control hierarchy
'FIELD'
Level 1
Level 2
Level 3
'GNAME-1'
'GNAME-2'
'GNAME-4'
'GNAME-5'
'GNAME-6'
'W4'
'W6'
'W8'
'W5'
'W7'
'W9'
'GNAME-3'
'Default-Group'
'GNAME-7'
'W10'
'W11'
'W-1'
'W-2'
'W-3'
This above example was obtained by using the following keywords:

*GROUP
*GROUP
*GROUP
*WELL
*WELL
*WELL
*WELL
*WELL
*WELL
*WELL
*WELL
*WELL
*WELL
*WELL
GNAME-1 GNAME-2 GNAME-3 *ATTACHTO

GNAME-4 GNAME-5
*ATTACHTO
GNAME-6 GNAME-7
*ATTACHTO
W-1
*ATTACHTO
W-2
*ATTACHTO
W-3
*ATTACHTO
W4
*ATTACHTO
W5
*ATTACHTO
W6
*ATTACHTO
W7
*ATTACHTO
W8
*ATTACHTO
W9
*ATTACHTO
W10
*ATTACHTO
W11
*ATTACHTO
FIELD
GNAME-1
GNAME-2
GNAME-3
GNAME-3
GNAME-3
GNAME-4
GNAME-4
GNAME-5
GNAME-5
GNAME-6
GNAME-6
GNAME-7
GNAME-7
Invalid example of a well control hierarchy:

'FIELD'
'GNAME-1'
'GNAME-4'
'W4'
'W5'
'GNAME-2'
'GNAME-5'
'W6'
'W7'
'GNAME-6'
'W8'
'W9'
'Default-Group'
'GNAME-7'
'W10'
'W11'
'GNAME-8'
'W1'
'GNAME-3'
'WELL-2'
'WELL-3'
This is invalid - wells and

group with same parent.
User's Guide IMEX
Please note that it is possible to accidentally overwrite group properties, constraints and
actions. When all the groups have been defined using *GROUP cards and all group actions,
properties and constraints have been specified using card such as *GCONI, *GCONP and
*GCONM. A further specification of any previous or new group using a *GROUP keyword
requires the user to completely respecify all group actions, properties and constraints.
If the group actions, properties and controls are not respecified, they are lost.
The same is not true for moving wells between groups by using the *ATTACHTO keyword.
This can be done without group action respecification.
Examples:
The first example below has a new group defined after group 1 actions are specified. Group 1
controls would be lost.
The second example shows the proper way of specifying multiple groups.
The third example below has a new group defined after group 1 actions are specified, but the
lost group controls are respecified.
Example 1: Group 1 Actions lost
*GCONP 'Group1'
*MAX
*STG 100000.0
*TARGET *STO
1000.0
*IPP
*GCONP 'Group2'
*MAX
*STG 110000.0
*TARGET *STO
900.0
*IPP
However in the cases below all group controls would be maintained

Example 2: Group definitions grouped together followed by all group control specifications.
*GCONP 'Group1'
*MAX
*STG 100000.0
*TARGET *STO
1000.0
*IPP
*GCONP 'Group2'
*MAX
*STG 110000.0
*TARGET *STO
900.0
*IPP
Example 3: Group 1 group control definitions lost and then respecified.

*GCONP 'Group1'
*MAX
*STG 100000.0
*TARGET *STO
1000.0
*IPP
User's Guide IMEX

*GCONP 'Group1'
*MAX
*STG 100000.0
*TARGET *STO
1000.0
*IPP
*GCONP 'Group2'
*MAX
*STG 110000.0
*TARGET *STO
900.0
*IPP
User's Guide IMEX
Well Identification (Required)
*WELL
PURPOSE:
*WELL is used to identify wells.

FORMAT:
*WELL
wnum well_name
(*VERT ibl jbl)

(*ATTACHTO group_name)
DEFINITIONS:
wnum
An integer representing the well sequence number. Well sequence numbers
must start at 1 and increase in increments of 1 as additional wells are defined.
Either the well sequence number or the well name must be specified. If only
the well sequence number is specified without the well name, then an internal
well name is generated automatically by the program. The program
generated name is: 'WELL-wnum'. Here wnum is the well sequence number.
'well_name'
Any alphanumeric character string (40 characters maximum). Enclose the
string in single quotes. Either the well sequence number or the well name
must be specified. If only the well name is specified without the well
sequence number, then an internal well sequence number is assigned to the
well automatically by the program. This sequence number is simply the
number of *WELL keywords in the data set up to and including the current
*WELL keyword. The use of quotes in well names is supported in a
consistent manner. To use a double quote in a well name, use a single quote
as a well string delimiter. To use a single quote in a well name, use a double
quote as a well string delimiter.
*VERT
This keyword indicates that the well is vertical and all completion layers
have the same I and J grid block indices. The completion layers are defined
by the *PERFV or *PERFVHY keyword. (Optional)
ibl
I direction grid block index for the vertical well. (Required with *VERT.)
There is no default value.
jbl
J direction grid block index for the vertical well. (Required with *VERT.) There
is no default value.
User's Guide IMEX
*ATTACHTO
Introduces the name of the parent group to which the well is attached.
Specified by the name 'group_name'. (Optional)
'group_name'
Name of the group to which the well is connected. If a group name is not
specified then the well is connected to an internally-generated group named
'Default-Group' by default. (Optional)
DEFAULTS:
Required keyword. No defaults. Minimum required is:

*WELL 'well_name'
-or*WELL wnum
Wells cannot be attached to the top-level group (the group which represents the whole field).
For information on how to specify the name of the top-level group, please see the manual
page for the *GROUP keyword. If there are no second-level groups defined in a data set then
all wells are attached to a group called 'Default-Group' automatically. For example if the line
appears in the data and a target rate for 'Field' is specified using the *GCONP keyword, the
following tree structure will exist by default.
FIELD
'WELL-1'
'WELL-2'
...
'WELL-n'
CONDITIONS:
EXPLANATION:
This keyword identifies the well number and name, and optionally assigns it to a well group.
It also identifies vertical wells.
'wnum' must be an integer number. If well numbers are used instead of well names, they
should start with 1 and increase in sequence for subsequent *WELL keywords.
User's Guide IMEX
Example:
** Both well sequence number and name
** are specified.
*WELL 1 'Producer1'
** Only well sequence number is specified.
** In this example the program will assign
** the name 'WELL-2' to the well.
*WELL 2 *VERT 12 14
** Only the well name has been specified.
** In this example the program will assign
** well number = 3 to this well.
*WELL 'Prod. 54-233a'
User's Guide IMEX
Well Backflow Model (Optional)
*XFLOW-MODEL
PURPOSE:
*XFLOW-MODEL is used to identify the well backflow, crossflow model to be used.

FORMAT:
*XFLOW-MODEL well_name wnum

(*FULLY-MIXED | *ZERO-FLOW | *SOURCE-SINK)
DEFINITIONS:
'well_name'
A well name or a list of well names to specify the wells to which this
definition applies (See explanation).
Note: wildcards may be used in the 'well_name' string as follows:
* replaces any number of characters at the end of a well name or can be used
on its own to replace all wells (e.g. *ALTER '*' or *ALTER 'wel*').
? replaces any single character anywhere in the well name (e.g. *ALTER
'?ell1').
The two wild cards can be combined on any list and when wild cards are
used the well list generated is printed out for the user to check.
The @ symbol in quotes may be used as a place holder for a list of wells or
groups that have satisfied a *TRIGGER condition. This list will be a subset
of the list of wells or groups specified as part of the *TRIGGER definition.
(See *TRIGGER for details).
wellnum
Any number of integers, or a range of integers to specify the well numbers to
which this definition applies. (See explanation).
*FULLY-MIXED
This keyword indicates that if backflow or crossflow occurs in the wellbore,
then it is modelled using a fully mixed assumption. This option is based on
one from Phillips Petroleum (cf. Coats et al, SPE 29111, 1995).
*ZERO-FLOW
This option will zero the flow of a backflowing layer at the Newton iteration
level as soon as it is detected. It will reopen the layer at the Newton iteration
level, as soon as it will no longer backflow.
*SOURCE-SINK
This option uses the block phase mobilities to calculate rates for backflowing
wells. This approach may result in unphysical results and is not recommended.
It is provided only to allow comparison with IMEX 96.11 and earlier runs.
User's Guide IMEX
DEFAULTS:
The default cross flow option is *FULLY-MIXED for all wells except injectors which do not
use mobility weighting (i.e. use the *UNWEIGHT) option. Unweighted injectors are
converted to use the *ZERO-FLOW cross flow option.
CONDITIONS:
This keyword must be located in the Well and Recurrent Data keyword group. For the
*FULLY-MIXED and the *ZERO-FLOW options, *WELLINIT *ITER is recommended.
*UNWEIGHT injectors are automatically converted to use the *ZERO-FLOW option
regardless of the option specified.
EXPLANATION:
This keyword identifies the well crossflow, backflow model to be used as a function of well
number and/or name.
The wellnum may be a single integer, a series of integers delimited by spaces or commas, or a
range of integers defined in the form: imin:imax.
Example:
*WELL 1 'Producer1'
*WELL 2 'Injector1' *VERT 12 14
*WELL 3 'Prod. 54-233a'
*XFLOW-MODEL 1:2 *FULLY-MIXED
*XFLOW-MODEL 'Prod. 54-233a' *ZERO-FLOW
User's Guide IMEX
Well Head Method (Optional)
*HEAD-METHOD
PURPOSE:
*HEAD-METHOD is used to identify the head method to be used.

FORMAT:
*HEAD-METHOD
well_list
(*GRAVITY | *ZERO-HEAD | GRAV-FRIC (*MODEL-AG | *MODELHOMOGENEOUS | *MODEL-MOMENTUM | *MODEL-DRIFTFLUX))

(*HEADRROUGH rrough) (*HEADSURFT surft)
DEFINITIONS:
well_list
( well_names | wellnums )
'well_names'
definition applies (See explanation).
Note: wildcard characters may be used in the 'well_name' string as follows:
'?ell1').
The two wildcard characters can be combined on any list and when wild
cards are used the well list generated is printed out for the user to check.
wellnums
which this definition applies. (See explanation).
*GRAVITY
The head between well layers is calculated based on mobility weighted
densities.
*ZERO-HEAD
The head between well layers is set to zero. This may be used, for instance,
to model a submersible pump where the liquid column height in the annulus
is small compared to the perforated length of the well and the gas density at
reservoir conditions is small.
*GRAV-FRIC
The head between well layers is calculated using a correlation which takes
account of fluid densities, frictional effects, and kinetic energy effects.
User's Guide IMEX
*MODEL-HOMOGENEOUS
Homogeneous well friction model (*GRAV-FRIC sub-keyword).
*MODEL-AG
The Aziz-Govier well friction model (*GRAV-FRIC sub-keyword)
*MODEL-MOMENTUM
The momentum equation well friction model (*GRAV-FRIC sub-keyword).
*MODEL-DRIFTFLUX
The drift flux well friction model (*GRAV-FRIC sub-keyword).
*HEADRROUGH
Subkeyword of *GRAV-FRIC introducing input of the value of relative
roughness to be used for the wellbore frictional head calculation.
rrough
Value of the relative roughness to be used in the frictional head calculation
for the listed wells. Default value is 0.0001 if *HEADRROUGH does not
appear in the line. Dimensionless.
*HEADSURFT
Sub-keyword of *GRAV-FRIC introducing input of the value of wellbore
gas-liquid surface tension used for the wellbore frictional head calculation.
The default value is 30 dynes/cm.
surft
Value of the gas-liquid surface tension to be used in the frictional head
calculation for the listed wells (dynes/cm).
DEFAULTS:
For IMEX, the default head method is *GRAVITY.

The default *GRAV-FRIC method is *MODEL-AG.
The default for *HEAD-RROUGH is 1.0e-4.
The default for *HEAD-SURFT is 30.0 dynes/cm.
CONDITIONS:
This keyword must be located in the WELL AND RECURRENT DATA keyword group.
EXPLANATION:
This keyword identifies the well head calculation method to be used for a given list of wells
specified by name or number.
User's Guide IMEX
Example:
*WELL 1 'Producer1'
*WELL 2 'Injector1' *VERT 12 14
*WELL 3 'Prod. 54-233a'
*HEAD-METHOD 1 3 *ZERO-HEAD
*HEAD-METHOD 2 *GRAVITY
User's Guide IMEX
Well Head Iteration (Optional)
*HEADITER
PURPOSE:
*HEADITER specifies that certain wells are to have the wellbore pressure differences
between completions (heads) computed iteratively at the beginning of a timestep to achieve
consistency between completion pressures and the computed heads. See also the manual
page for the related keyword *BHPHEADINIT.
FORMAT:

DEFINITIONS:
well_list
( well_names| wellnums )
'well_names'
A well name or a list of well names to specify the wells to which this definition
applies (See explanation). The specified wells may be producers or injectors.
'?ell1').
wellnums
which this definition applies. (See explanation). They specified wells may be
producers or injectors.
*ONE-STEP
The heads between well layers are computed without iteration to consistency.
This is the default and was the only method available before the introduction of
the *HEADITER keyword (*1-STEP is a permitted variant of this keyword).
*ITERATIVE
The heads between well layers are computed iteratively to achieve consistency
between the calculated pressure drops in the wellbore and the completion
pressures. This method may be useful in eliminating rate/pressure oscillations
from timestep to timestep. It is recommended that the default *XFLOWMODEL *FULLY-MIXED is used when the ITERATIVE option is specified.
User's Guide IMEX
DEFAULTS:
For IMEX, the default setting is *ONE-STEP.

CONDITIONS:
EXPLANATION:
*HEADITER allows the user to direct that the head calculation done at the beginning of a
timestep should be done iteratively to achieve consistency between the completion pressures
and the computed pressure differences between completions. Lack of consistency may be
manifested as oscillations of non-controlled rates from timestep to timestep.
Our experience has been that *HEADITER *ITERATIVE suffices to eliminate oscillations in
most models without the use of BHPHEADINIT.
If the well bore pressure in completion k is denoted as pwk and the surrounding gridblock
pressure as pbk, then the flow rate in the well bore depends upon the difference (pbk pwk). If
a full set of pwk values are chosen for a well, then the flow rates in all segments of the well are
determined. From these flows, heads may be determined (see the manual page for the
*HEAD_METHOD keyword); the head hk for completion k is the difference computed
between the wells BHP and the pressure in completion k. In the absence of a procedure to
enforce the equality of pwk (the pressure value used to compute the completion flow rates) and
BHP + hk (the pressure value resulting from these flow rates), there is no guarantee of
consistency between these two values for the completion pressure. The *HEADINIT
procedure applies a Newtonian iteration to determine a set of hk values for which pwk(BHP+hk) = 0 is satisfied in all completions, for a given value of BHP.
The application of this method requires the solution of dense linear systems with order equal
to the number of active completions in the well; for wells with 50 or more completions this
iteration may become excessively time-consuming.
The iterations may not converge; in this case the heads resulting from the first step of the
iteration (those that would have been obtained without invoking *HEADITER) are used.
Example:
*HEADITER Well1 *ITERATIVE
User's Guide IMEX
Simultaneous BHP and Well Head Iteration (Optional)

*BHPHEADINIT
PURPOSE:
*BHPHEADINIT specifies that certain wells are to have the bottom-hole pressure set to meet
the wells rate or WHP target and simultaneously have the wellbore pressure differences
between completions (heads) computed iteratively to achieve consistency between
completion pressures and the computed heads. See also the manual page for the related
keyword *HEADITER.. It is recommended that the default *WELLINIT *ITER option is
used when specifying *BHPHEADINIT *COMBINED
FORMAT:

DEFINITIONS:
well_list
( well_names| wellnums )
'well_names'
definition applies (See explanation). The specified wells may be producers
or injectors.
'?ell1').
wellnums
which this definition applies. (See explanation). They specified wells may be
producers or injectors.
*SEPARATE
The heads are set as directed by the *HEAD-METHOD and *HEADITER
keywords at the beginning of each timestep and left fixed throughout the
timestep; the BHP is set to meet the wells rate or WHP target with a frequency
dictated by the *WELLINIT keyword. This is the default and was the only
method available before the introduction of the *BHPHEADINIT keyword.
User's Guide IMEX
*COMBINED
A simultaneous iteration sets the BHP to meet the wells rate or WHP target
and computes heads between well layers to achieve consistency between the
calculated pressure drops in the wellbore and the completion pressures. This
method accounts for the effects of varying reservoir properties during the
timestep.
DEFAULTS:
For IMEX, the default setting is *SEPARATE.

CONDITIONS:
EXPLANATION:
*BHPHEADINIT allows the user to direct that the well initialization done to set the BHP to a
value meeting the wells rate or WHP target should simultaneously set the heads to achieve
consistency between the completion pressures and the computed pressure differences between
completions. Lack of consistency may be manifested as oscillations of non-controlled rates or
of layer rates.
When *SEPARATE is in effect, a wells BHP is adjusted iteratively until T(calculated)T(specified) = 0, where T is the value of the wells target rate or WHP. When *COMBINED
is in effect, the BHP and completion heads (see the discussion in the manual page for
*HEADITER) are iterated until T(calculated)-T(specified) = 0 and pwk-(BHP+hk) = 0 in all
completions k.
The application of this method requires the solution of dense linear systems with order equal
to the number of active completions in the well plus 1; for wells with 50 or more completions
this iteration may become excessively time-consuming.
The iterations may not converge; in this case the BHP and heads resulting from the first step
of the iteration are used.
Example:
*BHPHEADINIT Well1 *COMBINED
User's Guide IMEX
Perforations in Inactive Blocks (Optional)
*NULL-PERF
PURPOSE:
*NULL-PERF specifies how perforations in inactive (null or pinched out) grid blocks are to
be handled.
FORMAT:
*NULL-PERF well_list
(*STOP-SIM | *CLOSED)
DEFINITIONS:
well_list
A list of one or more well names or well numbers to specify the wells to
which this definition is to be applied. Well names must be enclosed in single
quotes. Well names and well numbers cannot be mixed in the same list. (See
explanation).
Note: When the list consists of well names, wildcards may be used in the
'well_name' string as follows:
'?ell1').
When the list consists of well numbers, ranges may be indicated with a
colon, e.g. 1:4.
*STOP-SIM
If the simulator detects an attempt to perforate a named well in an inactive
grid block, an error message identifying the well and block is printed and the
simulation is terminated.
*CLOSED
If the simulator detects an attempt to perforate a named well in an inactive
grid block, a warning message is printed and the perforation is given
CLOSED status. With CLOSED status, no fluids flow from or to the
reservoir in the layer, but the layer is retained in the well and enters the
wells head calculation.
User's Guide IMEX
DEFAULTS:
The default is *CLOSED.

CONDITIONS:
EXPLANATION:
This keyword specifies the treatment of attempts to perforate a well in an inactive (null or
pinched out) grid block. The attempt may either be considered an error, resulting in the
termination of the simulation, or may result in the layers being perforated with a status of
*CLOSED. A layer with closed status is allowed to be the reference layer for the well (the
layer in which the wells bottom hole pressure is defined).
Example:
*WELL 1 'Producer1'
*WELL 2 'Injector1'
*WELL 3 'Prod. 54-233a'
*NULL-PERF 1:2 *CLOSED
*NULL-PERF 3 *STOP-SIM
User's Guide IMEX
Well Type Definition (Required)
*PRODUCER, *INJECTOR
PURPOSE:
*PRODUCER indicates that the well identified by wellnum is a producer.

*INJECTOR indicates that the well identified by wellnum is an injector.
FORMAT:
*PRODUCER
-or*INJECTOR
('well_name')
(wellnum)
(*MOBWEIGHT) (well_name)
(*UNWEIGHT) (wellnum)
DEFINITIONS:
'well_name'
definition and subsequent operating constraint applies (See explanation).
wellnum
which this definition and subsequent operating constraint applies. (See
explanation).
*MOBWEIGHT
This subkeyword defines a total mobility weighted injector. (valid for
*INJECTOR only). Default.
*UNWEIGHT
This subkeyword defines an unweighted injector. (valid for *INJECTOR only).
DEFAULTS:
Required keywords. For *INJECTOR, *MOBWEIGHT is the default.
User's Guide IMEX
CONDITIONS:
This keyword must be located in the Well and Recurrent Data keyword group. The well type
specification must follow the well definition and must immediately precede the
*PWELLBORE or *IWELLBORE lines if present. Perforation data may precede the well
type specification *GEOMETRY cannot be used with *INJECTOR *UNWEIGHT.
The *FULLY-MIXED cross flow option cannot be used with *INJECTOR *UNWEIGHT
option, the *ZERO-FLOW option will be used instead.
A given well may be declared *SHUTIN any time after it has been defined with a *WELL
statement. Wells are initialized to the *SHUTIN status when *WELL is issued for the well;
*SHUTIN may be entered for the well but need not be.
EXPLANATION:
These required keywords are used for well definition. A well may be defined at one time using
*WELL, have its completions specified at a later time with *PERF, and at a still later time have
its type defined with *PRODUCER or *INJECTOR and go into operation at this time.
Example:
*WELL 1 'Injector'
*INJECTOR *MOBWEIGHT 1 ** Define the type of well 1.
*INCOMP *GAS
** Injecting gas.
** Operating on a rate specification.
** If backflow occurs, stop the simulation
*PERF 1
** if
jf
kf
WI
1
1
3 1.0E+5
*WELL 2 'Producer'
*PRODUCER 'Producer' ** Define the type of well 2.
** Operate a rate constraint first.
** If the BHP falls below 1000 psi, switch to a
** minimum pressure constraint.
** When the oil rate falls below 1000 bbl/D
** stop the simulation.
*MONITOR *BACKFLOW
*STOP
*MONITOR *GOR 20000.0
*STOP
*PERF 2
** if jf kf
WI
10 10
1 1.0E+5
** At this point you may shut in any of these wells.
** Shut in the injector
*TIME 900.0
*SHUTIN 1
User's Guide IMEX
Shut in Wells above Formation (Optional)
*MODELSHUT,
*EQUILIBRATE
PURPOSE:
FORMAT:
*MODELSHUT
*EQUILIBRATE
well_list
epsmds
(*ON | *OFF)
DEFINITIONS:
well_list
A list of 'well_names' or well_numbers to which this status applies. A limited

wildcard facility is available for lists of well names. Please see the manual
page for the *SHUTIN keyword for an explanation of the wildcard facility.
epsmds
Fluid equilibration criterion epsmds (m3/day | bbl/day | cm3/min) must be a

non-negative, real number.
DEFAULTS:
Optional keyword. Default is *MODELSHUT *OFF for all wells. Use of *MODELSHUT
with no occurrence of *ON or *OFF has the effect of setting *MODELSHUT *ON for the
wells listed.
If keyword *EQUILIBRATE is absent then epsmds is assumed to be 0.1.
CONDITIONS:
This keyword must be located in the WELL AND RECURRENT DATA keyword group after
wells have been defined with the keyword *WELL.
The *MODELSHUT option cannot be applied to a well whose type does not support the
*FULLY-MIXED back/cross flow model, such as the unweighted injectors.
EXPLANATION:
Various actions can cause a well to become shut in: *SHUTIN, effective *WLISTSHUT,
*ALTER or *OPERATE with a zero rate value, or violation of operation constraints. The
default method for shutting in a well (fully shut) is to close immediately all of its active
layers. However, keyword *MODELSHUT allows fluid to equilibrate for some time in the
wellbore before all the layers are closed (model-shut).
Keyword *MODELSHUT itself does not cause a well to shut in. Instead, it gives a well the
permission to go through an equilibration stage when the well is shut in for the reasons listed
above. During this equilibration stage the well is operated on a total rate constraint of zero,
which allows cross-flow in the wellbore to redistribute fluid between layers in the reservoir
(*FULLY-MIXED cross flow model). This occurs even if a different *XFLOW-MODEL
User's Guide IMEX
option was specified for that well. In the text output file, a well operating in an equilibration
stage is identified by the well status MSHT (as opposed to SHUT).
As the equilibration stage continues, pressure drops and layer flow rates may gradually
decrease. Depending on the reservoir, well completion and operation conditions, such a fluid
equilibration process may last months or even much longer in cases with persistent back flow.
When each layers rate falls below the criterion defined by *EQUILIBRATE the well leaves
the equilibration stage to become fully shut, that is, all the wells active layers are closed
immediately. A message is issued when the equilibration criterion has been satisfied.
Once a well has switched from the equilibration stage to fully shut, fluid equilibration is no
longer checked unless it is activated by another *MODELSHUT action. As a shutin well, a
*MODELSHUT well will not be involved in any group target apportionment. The keyword
*AUTODRILL will not put an autodrillable well in an equilibration stage unless such a well
has been drilled previously.
*MODELSHUT is not effective for a single-perforation well, which will equilibrate instantly.
Various actions can cause a shut-in well to open: *OPEN, effective *WLISTOPEN,
*ALTER or *OPERATE with a non-zero rate value. If an open action is encountered while
the well is in an equilibration stage, the well is immediately put back on its usual (most
restrictive) operating constraint with its assigned or defaulted backflow model. However, the
well retains its *MODELSHUT status, that is, its ability to use an equilibration stage upon the
next shut-in action. If you want to remove this ability, use *MODELSHUT *OFF explicitly.
Example:
** Wells #1 and #2 will experience fluid equilibration
** when they encounter a shut-in action. Well #3 will
** be fully shut in immediately.
*TIME ...
*MODELSHUT 1:2 *ON
*EQUILIBRATE 1d-2
** Larger equilibration criterion
...
*TIME ...
*SHUTIN 1:3
User's Guide IMEX
Well Status Definition (Optional)
*SHUTIN, *OPEN, *AUTODRILL
PURPOSE:
*SHUTIN indicates that the well identified by wellnum is shut in. *SHUTIN may be issued
for a well any time after it has been defined with a *WELL statement. A producer or an
injector must be fully defined, including the constraints before *OPEN or *AUTODRILL
may be issued for the well. A well with a monitor of *AUTOWELL or *AUTOLAYER can
reopen a shut in well unless the *AUTOWELL of *AUTOLAYER monitor is removed by
refining the well operating conditions.
*OPEN indicates that a previously shut-in well is reopened.
*AUTODRILL indicates that the well is currently not drilled (is shut in) but will be drilled
(opened) automatically to meet the target rates of its group, if the group has *AUTODRILL
*ON specified under the *GAPPOR keyword. Please see the manual entries for the *DRILLQ
and *GAPPOR keywords for more information.
FORMAT:
*SHUTIN
-or*OPEN
('well_name')
(wellnum)
('well_name')
(wellnum)
-or*AUTODRILL
('well_name')
(wellnum)
DEFINITIONS:
'well_name'
A well name or a list of well names to specify the wells to which this status
applies.
wellnum
which this status applies (See explanation).
User's Guide IMEX
DEFAULTS:
Optional keywords. No defaults.

CONDITIONS:
This keyword must be located in the Well and Recurrent Data keyword group, and must be
located after the well type (*PRODUCER or *INJECTOR) as well as the operating
constraints have been defined.
EXPLANATION:
These required keywords are used for well status. If a status of *AUTODRILL is specified,
the well must be defined completely, i.e. all its perforations and operating constraints must be
specified. When *WELL is first issued for a well, its status is initialized as *SHUTIN, and
*SHUTIN may be explicitly issued for a well any time after that, though it need not be.
Example:
*WELL 1 'Injector' *ATTACHTO 'Group-1'
*INJECTOR *MOBWEIGHT 1 ** Define the type of well 1.
*INCOMP *GAS ** Injecting gas.
** If backflow occurs, stop the simulation
*PERF 1
** if jf kf WI
1 1 3 1.0E+5
*WELL 2 'Producer' *ATTACHTO 'Group-1'
*PRODUCER 2 ** Define the type of well 2.
** Operate a rate constraint first.
** If the BHP falls below 1000 psi, switch to a
** minimum pressure constraint.
** When the oil rate falls below 1000 bbl/D
** stop the simulation.
*MONITOR *BACKFLOW
*STOP
*MONITOR *GOR
20000.0 *STOP
*PERF 2
** if jf kf WI
10 10 1 1.0E+5
*SHUTIN 1
*AUTODRILL 'Producer'
** At this point you may open any of these wells.
** Open the injector
*TIME 900.0
*OPEN 1
User's Guide IMEX
Shut and Reopen a List of Wells (Optional)
*WLISTSHUT,
*WLISTOPEN
PURPOSE:
*WLISTSHUT and *WLISTOPEN provide a means to temporarily shut a large list of wells
and later re-open them without disturbing the pattern of shut and autodrillable wells that
existed when the *WLISTSHUT keyword was read.
*WLISTSHUT indicates that the listed wells identified by well_numbers or well_names are
shut in. The difference between this keyword and the keyword *SHUTIN is that
*WLISTSHUT has a dynamic well list that a paired keyword *WLISTOPEN can operate
upon to reopen the listed wells without disturbing the pattern of shut and open wells that
existed when the *WLISTSHUT keyword was applied (see explanation).
*WLISTOPEN reopens wells or restores the auto-drillable status to wells in the common set
of the following well list (optional) and the effective well list from previous *WLISTSHUT.
All of the *WLISTSHUT list will be opened if no well list follows *WLISTOPEN.
FORMAT:
*WLISTSHUT
*WLISTOPEN
well_list
(well_list)
DEFINITIONS:
well_list
(well_names | well_numbers); numbers and names cannot be mixed in the
same list.
The well list is required for the *WLISTSHUT keyword and optional for
*WLISTOPEN.
well_names
A well name or a list of well names to specify the wells to which this status
applies. If a list of well names is given, a limited wild-card facility is
available to generate well lists automatically; please see the *SHUTIN and
*OPEN keywords for more information. Each well name must be enclosed
in single quotes.
well_numbers
which this status applies (See the explanation on the manual page for the
*SHUTIN and *OPEN keywords).
DEFAULTS:

CONDITIONS:
These keywords must be located in, but may appear anywhere in the WELL AND
RECURRENT DATA keyword group.
User's Guide IMEX
EXPLANATION:
*WLISTSHUT is used to pause the operations of a list of wells within the period ended by
the issuance of a paired keyword *WLISTOPEN. All wells on the list will be shut-in. Autodrillable status, if applicable, will be removed temporarily. Wells that have already been shutin or whose operational types have not yet been defined prior to the entry of the keyword are
automatically excluded from the list. Entering a second *WLISTSHUT list before
*WLISTOPEN appends the second list to the original one. If there is an explicit action that
could potentially modify the status for a particular well during this period (i.e. *OPEN,
*SHUTIN, *AUTODRILL, *PRODUCER, *INJECTOR, *CYCLPROD, *ALTER,
*TARGET, *ALTERCP), the change is made and that well is removed from the well list.
Wells being reopened or having their auto-drillable status re-acquired by *WLISTOPEN are
taken out of the remaining well list. The entire well list entered with *WLISTSHUT will be
processed if there is no well_list following the keyword *WLISTOPEN.
Example:
*TIME 0.
*PRODUCER 1:3
*OPERATE *MAX *STO 500.
*INJECTOR 4:5
*INCOMP *WATER
*OPERATE *MAX *STW 150.
*SHUTIN 1
*AUTODRILL 3
*TIME 100.
*WLISTSHUT 1:5
**Shutting a list of wells. Since Well 1 has already been shut in
before entry of the keyword, it will be automatically excluded from
the list. Well 3 is temporally disallowed to be auto-drilled.
*TIME 200.
*ALTER 2
400.
**Well 2 is explicitly altered and is thus removed from the
*WLISTSHUT list.
*TIME 300.
*WLISTOPEN
**Reopening wells on the remaining list issued by *WLISTSHUT, i.e.
Well 4 and 5. Meanwhile, Well 3 re-acquires the auto-drillable
candidacy.
User's Guide IMEX
Tubing Data for Injectors (Conditional)
*IWELLBORE
PURPOSE:
*IWELLBORE specifies that the wellbore pressure drops (reservoir to surface) will be
computed for an injector. Tubing data is required if a wellbore model is used. The pressure
drop can also be interpolated from a designated hydraulics table.
FORMAT:
*IWELLBORE
wdepth wlength rel_rough whtemp bhtemp wradius
or
*IWELLBORE *MODEL
or
*IWELLBORE *TABLE
wdepth table_number
DEFINITIONS:
wdepth
A real number specifying the well depth of an injector well (m | ft | cm | m).
When tabular interpolation is used, wdepth is used to scale the pressure drop
linearly relative to the depth entered with the designated table.
wlength
A real number specifying the well length of an injector well (m | ft | cm | m).
rel_rough
A real number specifying the relative well roughness. Dimensionless.
whtemp
A real number specifying the well head temperature (deg. C | deg. F | deg. C |
deg. C).
bhtemp
A real number specifying the reservoir temperature (deg. C | deg. F | deg. C |
deg. C).
wradius
A real number specifying the tubing radius (m | ft | cm | m).
MODEL
Indicates that the tubing pressure drop should be computed using a
correlation.
User's Guide IMEX
TABLE
Indicates that the tubing pressure drop should be interpolated from a
hydraulics table (input using the *ITUBE1 keyword).
table_number
*ITUBE1 table number.
DEFAULTS:

CONDITIONS:
This keyword must be located in the Well and Recurrent Data keyword group immediately
after *INJECTOR if the wellbore option is used and must precede the well completion
keywords.
When the first format above is used (with neither *MODEL nor *TABLE entered) the model
is used with the parameters supplied on the next line. *MODEL has exactly the same effect
as omitting the (*MODEL|*TABLE) subkeyword.
Required with *INJECTOR when *WHP is an operating constraint or when the wellbore
model is desired.
EXPLANATION:
This keyword is required if one of the operating constraints is *WHP or when the wellbore
model is desired for injectors.
Well depth is the vertical distance from the well head to the center of the grid block in which
the well's reference layer is perforated.
The length is the length of the well from the well head to the center of the grid block in which
the well's reference layer is completed.
Relative well roughness is defined as:
maximum roughness size
well diameter
The acceptable range of values for well depth is:
min
max
SI
m
-1.0E+4
1.0E+4
Field
ft
-32,808.0
32,808.0
Lab
cm
-1.0E+6
1.0E+6
User's Guide IMEX
The acceptable range of values for well length is:
min
max
SI
m
-1.0E+4
1.0E+4
Field
ft
-32,808.0
32,808.0
Lab
cm
-1.0E+6
1.0E+6
The acceptable range of values for the relative roughness is:
min
max
SI
m
-1.0E+4
1.0E+4
Field
ft
-32,808.0
32,808.0
Lab
cm
-1.0E+6
1.0E+6
The acceptable range of values for wellbore temperature is:

SI
C
-1.0E+4
1.0E+4
o
min
max
Field
o
F
-18032.0
18032.0
Lab
o
C
-1.0E+6
1.0E+6

SI
C
-1.0E+4
1.0E+4
o
min
max
Field
o
F
-18032.0
18032.0
Lab
o
C
-1.0E+6
1.0E+6
min
max
User's Guide IMEX
SI
m
-1.0E+4
1.0E+4
Field
ft
-32,808.0
32,808.0
Lab
cm
-1.0E+6
1.0E+6
Tubing Data for Producers (Conditional)
*PWELLBORE
PURPOSE:
*PWELLBORE specifies that the wellbore model will be used for this producer.
FORMAT:
*PWELLBORE *TABLE wdepth table_number

-or*PWELLBORE *MODEL gas_composition
DEFINITIONS:
*TABLE
This subkeyword specifies that the well-bore hydraulic pressure loss table
(input using the *PTUBE1) will be used for pressure loss calculations.
wdepth
A real number specifying the well depth of an producer well (m | ft | cm).
table_number
*PTUBE1 table number.
*MODEL
This subkeyword specifies that the well-bore hydraulic pressure loss FOR
SINGLE PHASE GAS PRODUCERS will be calculated by using the friction
factor correlation and the equation of state.
wlength
A real number specifying the well length of an producer well (m | ft | cm).
rel_rough
A real number specifying the relative well roughness. Dimensionless.
whtemp
A real number specifying the well head temperature (deg. C | deg. F | deg. C).
bhtemp
A real number specifying the reservoir temperature (deg. C | deg. F | deg. C).
wradius
A real number specifying the tubing radius (m | ft | cm).
gas_composition
Mole fractions of produced gas components in the following order. C1 C2
C3 IC4 NC4 IC5 NC5 C6 CO2. Nine values are required, one for each of the
above components.
User's Guide IMEX
DEFAULTS:
Conditional keyword. No defaults. If both *TABLE and *MODEL are missing then the
default is *TABLE.
CONDITIONS:
This keyword must be located in the Well and Recurrent Data keyword group immediately
following *PRODUCER if the wellbore model is used and before the well completion
keywords.
*PWELLBORE is required when the operating constraint *WHP is used or when the
wellbore model is desired.
EXPLANATION:
This keyword is required if one of the operating constraints is *WHP or the wellbore model is
desired, for producers only. Injectors use the *IWELLBORE keyword to enter the well depth.
The *MODEL option for calculating the pressure loss in the wellbore is applicable for single
phase GAS PRODUCERS ONLY. Use the pressure loss tables (*TABLE) for multi phase
flow in producers.
The composition of a gas producer is NOT used elsewhere in the simulator.
In IMEX the global composition of the gas must be entered in terms of the following
components in the order:
C1
C2
C3
IC4
NC4
IC5
NC5
C6
CO2
This is because the properties of the above components are built into the program.
Well depth is the vertical distance from the well head to the center of the grid block in which
the well's reference layer is perforated.
The length is the length of the well from the well head to the center of the top grid block that
is perforated by the well.
Relative well roughness is defined as:
maximum roughness size
well diameter
Example # 1
** In the following example *MODEL is used to calculate

** Wellbore hydraulic pressure loss for a gas producer.
** The units being used are *MODSI.
*DATE 1988 01 01
*DTWELL 0.003
*DTMIN 1.0E-06
*WELL 1 'Water Injector'
*INJECTOR 1
User's Guide IMEX
*INCOMP *WATER
**
rad
geofac wfrac skin
*GEOMETRY *K 0.0572 0.34
1.0
0.0
*PERF *GEO 1
** if jf kf
ff
4 4 4:4 1.0
*WELL 2 'Producer'
*PRODUCER 2
** Gas producer; uses analytical model.
** Composition
C1
C2 C3 IC4 NC4 IC5 NC5 C6 CO2
*PWELLBORE *MODEL 0.77 0.0 0.20 0.0 0.0 0.0 0.0 0.03 0.0
** wdepth wlength rel_rough whtemp bhtemp wradius
8400. 8400.
0.0001
60.0
160.0 0.25
*OPERATE *MIN *WHP 50.986
**
rad
geofac wfrac skin
*GEOMETRY *K 0.0572 0.34
1.0
0.0
*PERF *GEO 2
** if jf kf ff
2 2 1:4 1.0
*TIME 100.0
*STOP
Example #2
** In the following example *TABLE is used to calculate

** Wellbore hydraulic pressure loss for a producer.
** The units being used are *FIELD.
*DATE 1988 01 01
*DTWELL 0.003
*DTMIN 1.0E-06
*WELL 1 'Water Injector'
*INJECTOR 1
*INCOMP *WATER
*OPERATE *MAX *BHP 3138.
**
rad
geofac wfrac skin
*GEOMETRY *K
0.0572 0.34
1.0
0.0
*PERF *GEO 1
** if jf kf
ff
4 4
4:4 1.0
*WELL 2 'Producer'
*PRODUCER 2
** For producer use hydraulic table # 1.
*PWELLBORE *TABLE 8400.0 1
**
rad
geofac wfrac skin
*GEOMETRY *K 0.0572 0.34
1.0
0.0
*PERF *GEO 2
** if jf kf
ff
2
2 1:4 1.0
** Define the pressure loss table
User's Guide IMEX
*PTUBE1 1
*DEPTH 5000.0
*OIL **flo(1)
0.0
*GOR **gfr(1)
500.0
*WCUT **wfr(1)
0.00
*ALQ **add(1)
0.0
*WHP **whp(1)
200.0
*BHP
**iflo igfr iwfr
1
1
1
1
2
1
1
1
2
1
2
2
2
1
1
2
2
1
2
1
2
2
2
2
3
1
1
3
2
1
3
1
2
3
2
2
4
1
1
4
2
1
4
1
2
4
2
2
*TIME 100.0
*STOP
flo(2)
4000.0
gfr(2)
1000.0
wfr(2)
0.50
flo(3)
8000.0
whp(2)
900.0
whp(3)
1600.0
iadd
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
bhp(1)
2176.1
1997.7
2646.7
2447.7
2135.5
1682.7
2618.0
2189.0
2133.6
1463.1
2630.9
2022.0
2160.1
1425.7
2696.4
2080.0
flo(4)
16000.0
bhp(2)
2873.7
2670.9
3344.7
3124.7
2876.6
2674.6
3351.2
3132.3
2884.2
2684.5
3368.4
3152.8
2912.5
2721.3
3433.4
3231.0
bhp(3)
3573.7
3370.9
4044.7
3824.7
3576.6
3374.6
4051.2
3832.3
3584.2
3384.5
4068.4
3852.8
3612.5
3421.3
4133.4
3931.0
The acceptable range of values for well depth is:
min
max
SI
m
-1.0E+4
1.0E+4
Field
ft
-32,808.0
32,808.0
Lab
cm
-1.0E+6
1.0E+6
The acceptable range of values for well length is:
min
max
SI
m
-1.0E+4
1.0E+4
Field
ft
-32,808.0
32,808.0
Lab
cm
-1.0E+6
1.0E+6
The acceptable range of values for the relative roughness is:
min
max
User's Guide IMEX
SI
m
-1.0E+4
1.0E+4
Field
ft
-32,808.0
32,808.0
Lab
cm
-1.0E+6
1.0E+6
The acceptable range of values for wellbore temperature is:

SI
C
-1.0E+4
1.0E+4
o
min
max
Field
o
F
-18032.0
18032.0
Lab
o
C
-1.0E+6
1.0E+6

SI
C
-1.0E+4
1.0E+4
o
min
max
Field
o
F
-18032.0
18032.0
Lab
o
C
-1.0E+6
1.0E+6
min
max
SI
m
-1.0E+4
1.0E+4
Field
ft
-32,808.0
32,808.0
Lab
cm
-1.0E+6
1.0E+6
User's Guide IMEX
Composition of Injected Water, Gas, and Oil Phases

(Conditional)
*INCOMP
PURPOSE:
*INCOMP specifies the compositions of injected oil, water and gas phases.
ARRAY:
*INCOMP
(*OIL)
(*WATER)
(*GAS)
(*SOLVENT)
(*SEAWATER)
(cpvalue)
(*GLOBAL)
values
(*GLOBAL)
values
(fsvalue)
DEFINITIONS:
*OIL
This subkeyword identifies an oil injector.
*WATER
This subkeyword identifies a water injector.
*GAS
This subkeyword identifies a gas injector.
*SOLVENT
This subkeyword identifies a solvent injector. Use only with *MODEL
*MISCG or *MODEL *MISNCG.
*SEAWATER
This subkeyword identifies a seawater injector. Use only with *MODEL
*BLACKOIL_SEAWATER or *MODEL *OILWATER_SEAWATER.
*GLOBAL
This subkeyword identifies the global composition in mole fractions. Use
with *GAS or *SOLVENT.
cpvalue
Polymer concentration to be injected in the water phase. Use only with
*MODEL *POLYOW or *MODEL POLY (kg/m3 | lb/STB | g/cm3).
fsvalue
Volume fraction of seawater injected. Default value is 1.0. Use only with
*MODEL * BLACKOIL_SEAWATER or *MODEL
*OILWATER_SEAWATER (fraction).
User's Guide IMEX
values
Mole fractions of components in the order as defined in the Explanation
section.
DEFAULTS:

CONDITIONS:
This keyword must be located in the Well and Recurrent Data keyword group. *INCOMP must
be located after *INJECTOR and *IWELLBORE and before the well completion keywords.
Note: The use of the *GLOBAL subkeyword to model wellbore effects depends upon the
prior definition of *IWELLBORE for this well. Failure to define *IWELLBORE before
*INCOMP *GAS (or *SOLVENT) *GLOBAL is read will cause a read error.
Required for all injection wells.
EXPLANATION:
The keyword *INCOMP identifies the composition of injector fluid.

In IMEX, *INCOMP *GAS (or *SOLVENT) *GLOBAL is required for wellbore calculations
(*WHP constraint or when the wellbore model is desired) for gas (or solvent) injectors ONLY.
When *IWELLBORE *TABLE is specified, *GLOBAL and the following mole fraction
values need not (but may) be entered.
The composition of a gas or solvent injector is NOT used elsewhere in the simulator.
In IMEX the global composition of the gas (or solvent) must be entered in terms of the
following components in the order:
C1
C2
C3
IC4
NC4
IC5
NC5
C6
CO2
This is because the properties of the above components are built into the program.
Usually most of the injected gas should be C1 (i.e. greater than 80% of the total composition)
with trace amounts of C2, C3, C4, C5+ and CO2 unless the pseudo-miscible option is used.
Typically, solvent is entirely CO2, or predominantly C1, C2, C3 and trace amounts of C4+.
Let us assume the composition is as follows:
77% C1, 20% C3, 3% C6
*MODEL *MISCG
:
*RUN
:
*INJECTOR
*IWELLBORE
** wdepth wlength rough whtemp bhtemp rw
2305.0
2617.0 0.002
215.0
215.0 0.25
*INCOMP *SOLVENT *GLOBAL
** C1
C2
C3
IC4 NC4 IC5 NC5 NC6
CO2
0.77 0.0 0.2 0.0 0.0 0.0 0.0 0.03 0.0
User's Guide IMEX
Well Operating Constraints (Required)
*OPERATE
PURPOSE:
*OPERATE defines the well operating constraints, and the remedial action to be taken when
a constraint is violated.
FORMAT:
*OPERATE
-or*OPERATE
(*MAX or *MIN) constraint_type value (action)

(*PENALTY) penalty_type (mode) (value) (action)
where constraint_type is one of:

(*STO) value
(*STG)
(*STW)
(*STS)
(*STL)
(*BHP)
(*WHP (*INITIALIZE|*IMPLICIT))
(*BHW)
(*BHG)
(*BHF)
(*BHS)
(*DWN, *DWA, *DWB)
penalty_type is one of:
(*GOR)
(*WOR)
(*WGR)
( value is base_gor )
( value is base_wor )
( value is base_wgr )
mode is one of:

(*NODAMP)
(*DAMP)
(*PHASE2)
action is one of:
(*STOP)
(*CONT)
(*CONT *REPEAT)
(*SHUTIN)
(*SHUTIN *REPEAT)
DEFINITIONS:
*MAX
This subkeyword specifies that the specified value is a maximum constraint.
This is the default when neither *MAX nor *MIN is present.
User's Guide IMEX
*MIN
This subkeyword specifies that the specified value is a minimum constraint.
This subkeyword can be used only for the *WHP and *BHP constraints for
producers.
*PENALTY
This subkeyword specifies that the well is to be penalized for overproduction
resulting in a high GOR, WOR or WGR based on government regulations.
*PENALTY cannot be the only operating constraint. A maximum oil rate
constraint MUST be specified in order to use GOR or WOR penalization. A
maximum gas rate constraint MUST be specified in order to use WGR
penalization.
*STO
This subkeyword identifies a surface oil rate (m3/day | STB/day | cm3/min)
constraint. Specified rate values of zero are allowed and have the same
effect as the *SHUTIN keyword.
*STG
This subkeyword identifies a surface gas rate (m3/day | scf/day | cm3/min)
constraint. Specified rate values of zero are allowed and have the same
effect as the *SHUTIN keyword.
*STW
This subkeyword identifies a surface water rate (m3/day | STB/day |
cm3/min) constraint. Specified rate values of zero are allowed and have the
same effect as the *SHUTIN keyword.
*STS
This subkeyword identifies a surface solvent rate (m3/day | scf/day |
cm3/min) constraint. Specified rate values of zero are allowed and have the
*STL
This subkeyword identifies a total surface liquid rate (oil + water) (m3/day |
STB/day | cm3/min) constraint. Specified rate values of zero are allowed and
have the same effect as the *SHUTIN keyword. This keyword is applicable
to producers only.
*BHF
This subkeyword identifies a total reservoir fluid rate (oil + water + gas +
solvent) (m3/day | bbl/day | cm3/min) constraint. Specified rate values of
zero are allowed and have the same effect as the *SHUTIN keyword. This
keyword is applicable to producers only.
User's Guide IMEX
*BHW
This subkeyword identifies a reservoir water rate (m3/day | bbl/day |
cm3/min) constraint.
Specified rate values of zero are allowed and have the same effect as the
*SHUTIN keyword.
*BHG
This subkeyword identifies a reservoir gas rate (m3/day | ft3/day | cm3/min)
constraint. Specified rate values are allowed and have the same effect as the
*SHUTIN keyword. This keyword is applicable to injectors only.
*BHS
This subkeyword identifies a reservoir solvent rate (m3/day | ft3/day | cm3/min)
constraint. Specified rate values of zero are allowed and have the same effect as
the *SHUTIN keyword. This keyword is applicable to injectors only.
*GOR
This subkeyword indicates that penalization for the well is determined from
the produced gas-oil ratio. The produced GOR is determined from all the
layers of the well.
*WOR
the produced water-oil ratio. The produced WOR is determined from all the
layers of the well.
*WGR
the produced water-gas ratio. The produced WGR is determined from all the
layers of the well.
*BHP
This subkeyword identifies a bottom hole pressure (kPa | psi | kPa | kg/cm2)
constraint. All producers must have a minimum bottom hole pressure
constraint. If *MIN *BHP is missing for a producer then a default minimum
bottom hole pressure of 101.3 kPa or 14.7 psia will be assigned. Maximum
BHP constraints for producers are not allowed under *OPERATE, but may be
imposed using *MONITOR. Similarly, minimum BHP constraints for injectors
are not allowed under *OPERATE, but may be imposed using *MONITOR.
*WHP
This subkeyword identifies a well-head pressure (kPa | psi | kPa | kg/cm2)
constraint. To use this constraint, well tubing data must be entered via the
*IWELLBORE or *PWELLBORE keyword. The *WHP constraint can be
applied in two different ways, which are identified by the subkeywords:
User's Guide IMEX
*WHP *INITIALIZE : well bottom-hole pressure is initialized iteratively at

the beginning of a timestep, taking account of changes in bottom-hole
pressure but not of pressure changes in surrounding grid blocks. Bottom-hole
pressure is not adjusted during the Newtonian iterations.
*WHP *IMPLICIT: well bottom-hole pressure is initialized as in *WHP
*INITIALIZE, but is adjusted during the Newtonian iterations. This option
may require more Newtonian iterations than does *WHP *INITIALIZE, but
will honor the constraint value more accurately. This is the default and is
used if *WHP is specified without a subkeyword.
Maximum WHP constraints for producers are not allowed under *OPERATE,
but may be imposed using *MONITOR. Similarly, minimum WHP constraints
for injectors are not allowed under *OPERATE, but may be imposed using
*MONITOR.
*DWN, *DWA, *DWB
This subkeyword identifies a drawdown (kPa | psi | kPa | kg/cm2) constraint
the difference between the wellbore pressure and the grid block pressure.
Only *MAX is allowed for both producers and injectors. Specified
drawdown values of zero are allowed and have the same effect as the
*SHUTIN keyword. There are three definitions of drawdown that the user
can choose from. *DWN refers to the drawdown in the wells reference
layer; symbolically,
Pd = (Pblock ref Pwell ref )
where the sign is used to designate the producer (+) or injector (). It
should be noted that when the reference layer is perforated in a null/pinched
block, the simulation will either be terminated if *DWN is set as the primary
(first) operating constraint at a well change time; or continue with *DWN
constraint taking no effect if it is other than the primary one. *DWA defines
the maximum drawdown within all open layers,
Pd = max (Pblock l Pwell l )
l, open
which corresponds to the notion of imposing the maximum drawdown

constraint to avoid formation damage. *DWB specifies the average
drawdown for all open layers:
Pd =
l, open
l,open
weighted by the total product/injectivity (PI) at the reservoir condition. This

shall have the same effect as running the well on an equivalent, timedependent BHF rate constraint. In situations where the pressure differences
between completions in the wellbore differ considerably from the
corresponding differences in reservoir pressures, the actual drawdown in
each layer may differ considerably from the average value.
User's Guide IMEX
value
A real number specifying the constraint value.
*STOP
This subkeyword specifies that the simulation is to be terminated if the
constraint is violated.
*CONT
This subkeyword specifies that if the constraint is violated, the well should
switch to operating on the violated constraint, and the simulation continue.
This is the default remedial action for a constraint violation.
*SHUTIN
be shut in.
*NODAMP
Calculation mode for penalty constraints in which the reference phase target
rate (oil rate for GOR and WOR penalties, gas rate for WGR penalty) is
computed by literal application of the penalty formula. For further details
please see the EXPLANATION section below. In some cases this
calculation mode may cause oscillations in rate.
*DAMP
Calculation mode for penalty constraints in which the geometric mean of the
current reference phase rate and the literally computed target rate (that
calculated under *NODAMP) is assigned as the reference phase target rate.
For further details please see the EXPLANATION section below. This mode
is less subject to rate oscillations than is the *NODAMP mode, but when
reference phase mobilities are small near the well, this mode (and *NODAMP)
may encounter convergence difficulties.
*PHASE2
Calculation mode for penalty constraints in which a rate target is assigned for
the second phase (gas for GOR penalty, water for WOR and WGR penalties).
This mode is not subject to rate oscillations and does not encounter convergence
difficulties when the reference phase mobility is small, but may result in smaller
recovery of the reference phase than do the other two modes. For further details
please see the EXPLANATION section below. This mode is the default.
base_gor
A real number specifying the base value of GOR to be used for determining
the GOR penalty factor (m3/m3 | SCF/STB | cm3/cm3). If no value is entered,
base_gor defaults to the value 100 m3/m3 (561 SCF/STB, 100 cm3/cm3).
User's Guide IMEX
base_wor
A real number specifying the base value of WOR to be used for determining the
WOR penalty factor (m3/m3 | STB/STB | cm3/cm3. Unless otherwise specified,
the default value of base_wor is 100 m3/m3 (100 STB/STB, 100 cm3/cm3).
base_wgr
A real number specifying the base value of WGR to be used for determining
the WGR penalty factor (m3/m3 | STB/SCF | cm3/cm3). If no value is
entered, base_wgr defaults to the value 100 m3/m3 (17.8 STB/SCF).
*REPEAT
This subkeyword specifies that if the constraint is violated, the timestep
should be repeated after switching to operating on the violated constraint.
Every well with this option is entitled to repeat once with in one time step.
However, the maximum number of total repeats for this time step is limited
by user input (see keyword *MXCNRPT).
DEFAULTS:
Required keywords. No defaults. At least one operating constraint must appear in the data set.
All producers must have a minimum bottom hole pressure constraint. If *MIN *BHP is
missing for a producer then a default minimum bottom hole pressure of 101.3 kPa or 14.7
psia will be assigned.
All injectors must have a maximum bottom hole pressure constraint. If *MAX *BHP is
missing for an injector then a default maximum bottom hole pressure of 1,000,000 kPa or
147,000 psia will be assigned.
The default action is *CONT.
CONDITIONS:
This keyword must be located in the Well and Recurrent Data keyword group, and must
follow the *PRODUCER or *INJECTOR keywords.
EXPLANATION:
The keyword *OPERATE defines well operating constraints. The remedial action when a
constraint is violated is also chosen.
The first constraint entered is called the primary operating constraint. The simulator will
initially attempt to operate the well on this constraint (i.e. to enforce this constraint as an
equality) and monitor all other constraints regardless of whether the other constraints are
operating constraints or monitored constraints.
If one of the other specified operating constraints is violated while being monitored and
*CONT has been specified, then this violated constraint becomes the operating constraint,
and is enforced as an equality constraint (target constraint) instead of being monitored as an
inequality constraint.
If more than one constraint is violated and the most serious action is *CONT, the constraints
are checked to determine which constraint is the most restrictive, and the most restrictive
constraint becomes the well's target constraint.
User's Guide IMEX
The hierarchy of actions from most serious to least serious is as shown:

most serious
*STOP
*SHUTIN
*CONT
least serious
The value of the target constraint is changed easily at a subsequent simulation time for a
number of wells with the respecification of the *TARGET keyword.
A value for each constraint, regardless of the type of constraint, is required.
The following table shows the constraints allowed for each type of well:
Water injector
BHP
WHP
------STW
----BHW
DWN, DWA, DWB
-----
Solvent (gas) injector
Producer
BHP
WHP
--STG
STS
----BHF
--DWN, DWA, DWB
BHG
BHS
BHP
WHP
STO
STG
STS
STW
STL
BHF
BHW
DWN, DWA, DWB
-----
For example, to produce a well at a maximum rate of 500 m3/D, with a minimum bottom
hole pressure at 2500 kPa and with monitoring constraints of GOR at 2000 m3/m3, water-cut
at 98% and monitoring for backflow, the input looks like this:
*OPERATE
*OPERATE
*MONITOR
*MONTIOR
*MONITOR
*MAX
*MIN
*GOR
*WCUT
*BACKFLOW
*STO
*BHP
500.00
2500.00
2000.00
0.98
*CONT
*STOP
*STOP
*STOP
*REPEAT
If, while the primary constraint is active, the bottom hole pressure falls below 2500 kPa then
the bottom hole pressure constraint becomes the operating constraint.
If more than one operating or monitoring constraint is violated, then the most serious action
associated with those constraints is taken.
User's Guide IMEX
PENALTY CONSTRAINTS
Some regulatory agencies require that a phase ratio, such as GOR, be measured periodically
(e.g. at the end of each month) and that if this measured value exceeds a prescribed base
value, the well's production in the coming period (e.g. the next month) must not exceed a
specified maximum rate times the ratio of the base and measured values, for example
Allowed rate = (Max rate)*(Base GOR)/(Measured GOR)
where the allowed rate is for the coming month and the GOR was measured near the end of
the past month.
When the *PENALTY *GOR, *WOR, or *WGR constraints are specified they must be
preceded by a maximum reference phase rate constraint entered using the *MAX *STO or
the *MAX *STG subkeywords of *OPERATE. This rate is used as the maximum rate in the
simulator's computation of the allowed rate. The "reference" phase is oil for the GOR and WOR
penalties and gas for the WGR penalty.
The *PENALTY subkeyword applies only to production wells. If *OPERATE *PENALTY
is entered for an injector, the program will terminate with an error message.
When *PENALTY and its subkeywords are used, the well rate during the coming timestep may
be reduced by a penalty factor calculated using the GOR which prevails at the beginning of the
timestep. If the GOR, WOR, or WGR at the beginning of the timestep exceeds the specified
base value, then the production rate in the coming timestep is decreased in a manner which
depends on the calculation mode chosen for the penalty constraint, as described below. The
maximum oil or gas rate constraints are still honored. In applying the *PENALTY keyword to
simulate regulatory rate limitations, care should be taken to have timestep lengths equal to the
regulatory period. This can be achieved with *TIME or *DATE data lines.
The GOR, WOR, and WGR penalty factors are defined by the following equations :
PF(gor) = Base_GOR / Produced GOR
PF(wor) = Base_WOR / Produced WOR
PF(wgr) = Base_WGR / Produced WGR
Here Produced GOR, WOR, and WGR are the values at the beginning of the timestep for
which the penalty factor is being calculated. If PF exceeds or equals one, then the
*PENALTY constraint imposes no limit upon the rate. If PF is less than one, the
*PENALTY constraint limits the well rate in the coming timestep as follows.
When the *NODAMP calculation mode is in effect the penalized rate for the coming timestep
becomes:
Qo_new = Qo_max * PF
for GOR and WOR penalties and
Qg_new = Qg_max * PF
for a WGR penalty. Qo_max and Qg_max are the maximum stock-tank oil and gas rates
entered for the penalized well using the *OPERATE *MAX *STO or *OPERATE *MAX
*STG keywords. If the timestep size is equal to the regulatory period this corresponds to a
literal application of the expression above for Allowed rate. The *NODAMP mode is subject
to oscillations in the assigned rates.
User's Guide IMEX
If the *DAMP mode is chosen, then the geometric mean of Qo_new or Qg_new as computed
above and the current rate (the rate assigned in the timestep just finished) is assigned as the
rate in the coming timestep. In this case,
Qo_new = sqrt(Qo_base * PF * Qo_old)
for GOR and WOR penalties and
Qg_new = sqrt(Qg_base * PF * Qg_old)
for a WGR penalty. This mode is less subject to severe oscillations in assigned rate than is
the *DAMP mode, but both *DAMP and *NODAMP may give rise to simulator convergence
problems when the reference phase mobility is small near the well.
If the *PHASE2 mode is chosen, and PF is less than one, then a target rate is assigned for the
second phase (gas for GOR, water for WOR and WGR penalties) equal to
Qg_target = Qo_base * Base_GOR
for the GOR penalty or
Qw_target = Qo_base * Base_WOR
for the WOR penalty or
Qw_target = Qg_base * Base_WGR
for the WGR penalty. This mode is not subject to oscillations in assigned rate and is not
affected by small reference phase mobilities near the well. However, in situations in which
production penalties are imposed by regulation, the measured phase ratio (e.g. the GOR)
tends to rise over the course of a timestep. Since the *DAMP and *NODAMP modes assign
rates based on GOR values at the start of the timestep, the *DAMP and *NODAMP modes
usually result in higher rates for the second phase (and hence also the reference phase) than
are allowed by the *PHASE2 mode.
If the well is operating under a constraint other than a reference phase (for *NODAMP and
*DAMP modes) or second phase (for *PHASE2) rate constraint with a *PENALTY
constraint specified, and the calculated penalty factor is less than one, then the well's target
constraint is switched to a rate constraint for the proper phase.
Examples:
(Usage of *OPERATE with *PENALTY subkeyword)

*OPERATE *MAX *STO 3000.0 *CONT
*OPERATE *MIN *BHP 1500.0 *CONT *REPEAT
*MONITOR *BACKFLOW
*STOP
*MONITOR *GOR
50000.0 *STOP
*OPERATE *PENALTY *GOR *PHASE2 393.0
*OPERATE *PENALTY *WOR *DAMP
50.0
- - - - - - - - - - - - - - - - - - - - - - *OPERATE *MAX *STG 3000000.0 *CONT
*OPERATE *MIN *BHP
1200.0 *CONT *REPEAT
*MONITOR *WGR
1600. *STOP
*OPERATE *PENALTY *WGR *PHASE2 150.0
User's Guide IMEX
Maximum Number of Continue-Repeat (Optional)
*MXCNRPT
PURPOSE:
*MXCNRPT specifies maximum number of continue-repeat allowed in each time step.

FORMAT:
*MXCNRPT nmxrpt
DEFINITIONS:
nmxrpt
A non-negative integer defining the maximum allowed number of time step

repeats due to well constraint switches.
DEFAULTS:
The default is 1. At most, one time step repeat due to well constraint switch is allowed per
time step, no matter which well(s) initiates the constraint switch.
CONDITIONS:
This keyword must be located in the WELL AND RECURRENT DATA keyword group. It
has no effect if none of the wells has *CONT *REPEAT specified through *OPERATE.
EXPLANATION:

and keyword *MXCNRPT lets the user override its default value.
User's Guide IMEX
Monitored Well Constraints (Optional)
*MONITOR
PURPOSE
*MONITOR defines the monitored well constraints, and the remedial action to be taken when
a monitored constraint is violated. Monitored constraints differ from operating constraints in
that a monitored constraint can never be imposed directly as a well target. The constraints are
checked and the actions are applied only after a timestep has been completed; actions are not
applied part way through a timestep.
FORMAT:
*MONITOR monitored_constraint value action

Where:
monitored_constraint for producers is one of:
(*GOR)
(*SOR)
(*GSOR)
(*WCUT)
(*WGR)
(*BACKFLOW)
(*WHYSTAB)
(*MIN)
(*MIN)
(*MAX)
(*MAX)
(*MAX)
(*MAX)
(*MAX)
(*MAX)
(*STO)
(*STG)
(*STG)
(*STW)
(*STS)
(*STL)
(*WHP)
(*BHP)
monitored_constraint for injectors is one of:

(*BACKFLOW)
(*WHYSTAB)
(*MIN)
(*MIN)
(*MIN)
(*MIN)
(*MIN)
User's Guide IMEX
(*STG)
(*STW)
(*STS)
(*WHP)
(*BHP)
action for producers is one of:

(*STOP)
(*SHUTIN)
(*SHUTIN *REPEAT)
(*AUTOWELL frequency)
(*RECOMPLETE *UP)
(*RECOMPLETE *DOWN)
(*RECOMPLETE *UP_OR_DOWN)
(*SHUTLAYER)
(*AUTOLAYER frequency)
(*WORKOVER (pi_factor ntimes wo_action))
action for injectors is one of:
(*STOP)
(*SHUTIN)
(*SHUTIN *REPEAT)
action when *BACKFLOW is monitored is one of:
(*STOP)
(*SHUTIN)
(*SHUTIN *REPEAT)
(*AUTOLAYER frequency)
action when *WHYSTAB is monitored is one of:
(*STOP)
(*SHUTIN)
(*SHUTIN *REPEAT)
action for producers when *MAX WHP or *MAX BHP is monitored is one of:
(*STOP)
(*SHUTIN)
(*SHUTIN *REPEAT)
No other actions are permitted.
DEFINITIONS:
*GOR
This subkeyword identifies a maximum gas-oil ratio (m3/m3 | scf/STB |
cm3/cm3) monitor for a producer.
*SOR
This subkeyword identifies a maximum solvent-oil ratio (m3/m3 | scf/STB |
*GSOR
This subkeyword identifies a maximum (gas+solvent)/oil ratio (m3/m3 |
scf/STB | cm3/cm3) monitor for a producer.
User's Guide IMEX
*WCUT
This subkeyword identifies a maximum water-cut (fraction) monitor for a
producer.
*WGR
This subkeyword identifies a maximum water-gas ratio (m3/m3 | STB/scf |
*BACKFLOW
This subkeyword identifies backflow monitoring. No value is required for
this monitor; only an action is required. The default action is to print a
message with no further action taken.
*WHYSTAB
This subkeyword identifies wellbore hydraulics stability monitoring. A real
well can operate only above a certain rate at which the bottom-hole pressure
remains in the stable region of the well-head pressure curve (a region in
which the curve of WHP vs. BHP has a positive slope). It can only be used
for wells for which a method of computing WHP has been introduced with
the *PWELLBORE or *IWELLBORE keyword. No value is required for this
monitor, but an action is required. The default action is to print a message
with no further action taken.
*MIN
This subkeyword indicates that a rate, BHP, or WHP constraint is to be
monitored as a minimum.
*MAX
This subkeyword indicates that a rate, BHP, or WHP constraint is to be
monitored as a maximum.
*STO
constraint for a producer.
*STG
constraint.
*STW
*STS
User's Guide IMEX
*STL
This subkeyword identifies a total surface liquid (oil + water) rate (m3/day |
STB/day | cm3/min) constraint for a producer.
*WHP
This subkeyword identifies a maximum (producers) or minimum (injectors)
well head pressure (psi | kPa | kPa | kg/cm2) monitor. It can only be used for
wells for which a method for computing WHP has been introduced with the
*PWELLBORE or *IWELLBORE keywords.
*BHP
This subkeyword identifies a maximum (producers) or minimum (injectors)
bottom hole pressure (psi | kPa | kPa | kg/cm2) monitor.
value
A real number specifying the monitored value.
*STOP
This subkeyword specifies that the simulation is to be terminated if the
constraint is violated.
*SHUTIN
be shut in. This is the default remedial action for a constraint violation.
*REPEAT
This subkeyword of the *SHUTIN action specifies that the timestep should
be repeated after the indicated *SHUTIN action is taken.
*AUTOWELL
Similar to *SHUTIN except that well productivity is checked periodically
and the well is reopened automatically if the previously violated monitored
constraint ceases to be violated. The *AUTOWELL action is valid for the
*MIN *STO, *MIN *STG, *GOR, *SOR, *WCUT, *GSOR, *WGR, *MAX
*WHP, *MAX *BHP, and *WHYSTAB monitors for producers only.
*AUTOWELL can reopen explicitly shut in wells. To avoid this, redefine
the wells monitors to exclude the *AUTOWELL condition, the well can be
then shut using the *SHUTIN keyword without fear of reopening.
frequency
This is the frequency in days at which the checking which is part of the
*AUTOWELL or *AUTOLAYER actions should be performed in an attempt
to reopen the well or layer.
User's Guide IMEX
*RECOMPLETE
Plug the most offending layer (the one with the largest value of a maximum
constraint or the smallest value of a minimum constraint) and, if possible,
open a currently plugged layer in the well. This layer may have been
plugged through earlier constraint violation action or through specification of
the *AUTO layer status with the *PERF, *PERFV, or *PERFVHY
keywords. *RECOMPLETE is a valid action for all allowed monitored
constraints for producers.
*UP
Keyword specifying that the recompletion must be done in the upward
direction from the most offending layer.
*DOWN
Keyword specifying that the recompletion must be done in the downward
direction from the most offending layer.
*UP-OR-DOWN
Keyword specifying that the recompletion will be done upward from the
most offending layer for WCUT, MIN STO, MIN STG, MAX STW, MAX
STL, and WGR violations, and downward from the most offending layer for
GOR, SOR, GSOR, MAX STG, MAX STS, and BACKFLOW violations.
*SHUTLAYER
Plug the most offending layer and continue the simulation.
*AUTOLAYER
Similar to *SHUTLAYER except that the plugged layer is checked
periodically and if the previously violated monitored constraint ceases to be
violated, the layer is reopened automatically. *AUTOLAYER is valid only
for the *GOR, *WCUT, *WGR, and *BACKFLOW monitored constraints.
When *AUTOLAYER is the action for the *BACKFLOW monitor, a layer
is opened at the checking time only if at least one layer of the well has
remained open; i.e., for *BACKFLOW the *AUTOLAYER action will not
reopen a well which has been shut because all of its layers have been
plugged. For the other monitors, *AUTOLAYER may reopen a well.
*AUTOLAYER may be applied only to producers -- a backflowing injector
can not have the *AUTOLAYER action specified.
*WORKOVER
Increases the well productivity indices for all layers of the well by a factor
pi_factor and continues the simulation. This workover is applied a maximum
of ntimes in the simulation. The *WORKOVER action is valid only for the
*MIN *STO and *MIN *STG monitored constraints.
User's Guide IMEX
pi_factor
A real number specifying the improvement in all the layer productivities for
the well:
New_PI = Old_PI * pi_factor.
The entry of pi_factor, ntimes, and wo_action after WORKOVER is
optional; the recommended procedure is to use the *WORKOVER-DATA
keyword to enter the workover parameters; please consult the manual page
for *WORKOVER-DATA.
ntimes
The maximum number of workovers allowed before wo_action is taken
instead of the productivity index increment. The entry of pi_factor, ntimes,
and wo_action after *WORKOVER is optional; the recommended procedure
is to use the *WORKOVER-DATA keyword to enter the workover
parameters; please consult the manual page for *WORKOVER-DATA.
wo_action
The action to be taken if the monitor is violated after the maximum number
of workovers has been exhausted. wo_action can be one of the following:
*STOP, *SHUTIN or *SHUTIN *REPEAT. The default is *SHUTIN.
The entry of pi_factor, ntimes, and wo_action after WORKOVER is
optional; the recommended procedure is to use the *WORKOVER-DATA
keyword to enter the workover parameters; please consult the manual page
for *WORKOVER-DATA.
DEFAULTS:

The default action is *SHUTIN. See the manual page for *WORKOVER-DATA for the
default values of pi_factor, ntimes, and wo_action.
CONDITIONS:
This keyword must be located in the Well and Recurrent Data keyword group, and must follow
the *PRODUCER or *INJECTOR.
EXPLANATION:
The keyword *MONITOR defines monitored well constraints, i.e. constraints which are not
to be set as targets if violated but which cause a specified remedial action to be taken.
Some quantities can be monitored only for certain well types. For instance, *GOR, *SOR,
*GSOR, and *WCUT are used only for producing wells, while the *MIN *BHP monitor can
be used only for injectors.
*BACKFLOW can be monitored for either producers or injectors.
A value for each constraint, except for *BACKFLOW, is required.
User's Guide IMEX
For example, to produce a well at a maximum rate of 500 m3/day, with a minimum bottom
hole pressure at 500 kPa and with monitoring constraints of GOR at 25000 m3/m3, water-cut
at 98% and monitoring for backflow, the input looks like this:
*OPERATE
*OPERATE
*MONITOR
*MONITOR
*MONITOR
*MAX *STO
500.00
*MIN *BHP
500.00
*GOR
25000.00
*WCUT
0 .98
*BACKFLOW
*CONT *REPEAT
*STOP
*STOP
*AUTOLAYER 10.0
If, while the primary constraint is active, the bottom hole pressure falls below 500 kPa then
the bottom hole pressure constraint becomes the operating constraint.
If, while the primary constraint is active, the bottom hole pressure constraint, the GOR
constraint, and the backflow monitor are violated simultaneously, then the most serious of the
specified actions for the three violated monitors will be taken. In this case, the simulation is
terminated, since *STOP is specified for a GOR violation.
A well is considered to be operating normally if, in all layers,
for an injector Pbh + head > P(block)
for a producer Pbh + head < P(block)
If the above do not hold in all layers, then the well is considered to backflow.
The acceptable range of values for gas-oil ratios is:
min
max
SI
m3/m3
0.0
1.0E+20
Field
scf/STB
0.0
5.615E+20
Lab
cm3/cm3
0.0
1.0E+20
The acceptable range of values for solvent-oil ratios is:
min
max
SI
m3/m3
0.0
1.0E+10
Field
scf/STB
0.0
56,165
Lab
cm3/cm3
0.0
1.0E+10
The acceptable range of values for water-gas ratios is:
min
max
User's Guide IMEX
SI
m3/m3
0.0
1.0E+20
Field
STB/scf
0.0
1.781E+19
Alter Primary Well Operating Constraint Value (Optional)

*ALTER
PURPOSE:
*ALTER allows modification of only the primary operating constraint value the well(s)
whose names or numbers are listed. The primary operating constraint is the FIRST constraint
entered for the well using the *OPERATE keyword.
FORMAT:
*ALTER ('well_names') or (well_numbers) values

DEFINITIONS:
well_names
alteration of the primary operating constraint applies. These names must be
on the same line as the *ALTER keyword.
Note: wildcards may be used in the 'well_names' string as follows:
'?ell1').
well_numbers
which this alteration of primary operating constraint applies. These well
numbers must be on the same line as the *ALTER keyword. If more than
one line is required to specify all the wells then the *ALTER keyword must
be repeated.
values
One number for each well identified by well_numbers or 'well_names' (or a
single value for all wells) specifying the new value of the primary operating
constraint. Values must appear on one or more new lines immediately
following the *ALTER line.
If only one value of 'values' is input, all wells defined in the well list will be
set to that single value.
User's Guide IMEX
DEFAULTS:

CONDITIONS:
*ALTER must be located in the Well and Recurrent Data keyword group, and may appear
anywhere in this keyword group following the initial *OPERATE declarations for all of the
wells in the *ALTER list.
EXPLANATION:
This optional keyword is used to alter the primary operating constraint for a well or set of
wells without having to redefine all of the additional operating constraints. It is an effective
method of altering values when performing a history match.
*ALTER, followed by a non-zero value, also opens a well if the well has been shut in by a
previous action or if the well has been initially defined as a shut in well.
When *ALTER is encountered in a data set, the simulator checks if the primary constraint
with the new value becomes the most restrictive well constraint. If so, the well is switched to
the primary constraint. If not, the new value is entered for the primary constraint but the well
is switched to (or continues to run on) the currently most restrictive constraint.
If a primary constraint value of zero is specified using *ALTER for an open well then that
well is shut in. If a non-zero value is specified for a shut in well then that well is opened.
*ALTER has essentially the same effect as *TARGET, except that *TARGET can apply to
constraints other than the primary constraint.
The *TARGET keyword is a more flexible version of the *ALTER keyword. The *ALTER
keyword allows for the modification of only the primary operating constraint (the FIRST
operating constraint specified for a well). The *TARGET keyword, on the other hand, allows
for the specification of any operating constraint.
Examples:
*PRODUCER 1
*OPERATE *MAX *OIL 500.00
:
*ALTER 1
750
The *ALTER keyword may also look like this when
several wells have been defined:
*WELL 1 'Producer 1'
*WELL 4 'Injector 1'
:
*PRODUCER 1
*PRODUCER 2
User's Guide IMEX
*PRODUCER 3
*INJECTOR 4
*MONITOR *MAX *WATER 100.0
:
*TIME 1200.
** At a later date, want to adjust the operating
** constraint values.
** well_numbers
*ALTER 1:2
3
4
** values
2*1000.0 800.0 50.0
User's Guide IMEX
Alter Well Constraint Value (Optional)
*TARGET
PURPOSE:
*TARGET allows modification of any previously specified well constraint value or the
specification of a new constraint type and value for well(s) listed by well_numbers or
'well_names'.
FORMAT:
*TARGET
(*STO) (well_numbers)
(*STG) ('well_names')
(*STW)
(*STL)
(*STS)
(*BHP)
(*WHP)
(*DWN, *DWA, *DWB)
(*BHF)
(*BHG)
(*BHS)
values
DEFINITIONS:
*STO
constraint. Zero rates are allowed and will have the same effect as the
*SHUTIN keyword.
*STG
*SHUTIN keyword.
*STW
cm3/min) constraint. Zero rates are allowed and will have the same effect as
the *SHUTIN keyword.
*STS
*STL
STB/day | cm3/min) constraint. Zero rates are allowed and will have the
User's Guide IMEX
*BHF
solvent) (m3/day | STB/day | cm3/min) constraint. Zero rates are allowed
and will have the same effect as the *SHUTIN keyword.
*BHW
This subkeyword identifies a reservoir water rate (m3/day | STB/day |
*BHP
This subkeyword identifies a bottom hole pressure (kPa | psi | kPa | kg/cm2)
*WHP
This subkeyword identifies a well-head pressure (kPa | psi | kPa | kg/cm2)
constraint.
*DWN, *DWA, *DWB
This subkeyword identifies a draw-down pressure (kPa | psi | kPa | kg/cm2)
constraint.
*BHG
This subkeyword identifies a reservoir gas rate (m3/day | ft3/day | cm3/min)
*SHUTIN keyword. This keyword is applicable to injectors only.
*BHS
This subkeyword identifies a reservoir solvent rate (m3/day | ft3/day |
the *SHUTIN keyword. This keyword is applicable to injectors only.
well_numbers
which this target applies. These well numbers must be on the same line as
the *TARGET keyword. If more than one line is required to specify all the
wells, then the *TARGET keyword must be repeated.
well_names
Any number of 'well_names' in quotes to specify the wells to which this
target applies. These names must be on the same line as the *TARGET
keyword. If more than one line is required for the well list, then the
*TARGET keyword must be repeated.
User's Guide IMEX
values
single value for all wells) specifying the new value of the target. Values must
appear on one or more new lines immediately following the *TARGET line.
Values must NOT appear on the same line as the list of well names or numbers.
DEFAULTS:

CONDITIONS:
*TARGET must be located in the Well and Recurrent Data keyword group, and may appear
anywhere in this keyword group following the initial *OPERATE declaration.
EXPLANATION:
This optional keyword is used to alter the target constraint for wells without having to
redefine all of the additional operating constraints. It is an effective method of altering
targets when performing a history match.
*TARGET also changes a target constraint with a new input value if the well is operating
with a different target or if the well has been shut in by a previous action or if the well has
been initially defined as a shut in well.
The target type is specified by using one of *STO, *STG, *STW, *STL, *STS, *BHP,
*WHP, *DWN, *DWA, *DWB, *BHF, *BHW, *BHG or *BHS. One of these qualifiers
must be present.
The *TARGET keyword is a more flexible version of the *ALTER keyword. The *ALTER
keyword allows for the modification of only the primary operating constraint (the FIRST
operating constraint specified for a well). The *TARGET keyword on the other-hand allows
for the specification of any operating constraint.
User's Guide IMEX
Examples:
*PRODUCER 1
*MONITOR *MIN *STO 10.00 *SHUTIN
:
*TARGET *STO ** This alters the target constraint in
1
** the *TARGET keyword above, i.e.,
750
** *STO.
The *TARGET keyword may also look like this when
several wells have been defined:
:
*PRODUCER 1
*PRODUCER 2
*PRODUCER 3
*INJECTOR 4
*MONITOR *MIN *STW 10.00 *SHUTIN
:
*TIME 1200.
** At a later date, want to adjust the target
*TARGET *STO
** well_numbers
1:2
** values
2*1000.0
*TARGET *BHP
** well_numbers
3
** values
800.0
*TARGET *STW
** well_numbers
4
** values
50.0
User's Guide IMEX
Sets/Alters a Wells Phase Productivity (Optional)
*SETPI
PURPOSE:
*SETPI sets/multiplies a wells specified phase productivity/injectivity to/by a user defined

value.
FORMAT:
*SETPI
(*OIL) ('well_names') or (well_numbers) values

(*WAT)
(*LIQ)
(*GAS)
(*TOT)
(*MULT)
(*MULTO)
Values of Productivity, Injectivity Index or Multiplication Factor

DEFINITIONS:
*OIL
This subkeyword identifies that the PI of oil will be set to the values
specified. Units are (m3/day/kPa | STB/day/psi | cm3/min/kPa)
*WAT
This subkeyword identifies that the PI of water will be set to the values
specified. Units are (m3/day/kPa | STB/day/psi | cm3/min/kPa)
*GAS
This subkeyword identifies that the PI of gas will be set to the values
specified. Units are (m3/day/kPa | scf/day/psi | cm3/min/kPa)
*LIQ
This subkeyword identifies that the PI of the sum of oil and water will be set
to the values specified. Units are (m3/day/kPa | STB/day/psi | cm3/min/kPa)
*TOT
This subkeyword identifies that the PI of the sum of oil, water and gas will
be set to the values specified. Units are (m3/day/kPa | scf/day/psi |
cm3/min/kPa). Note the units of *TOT is assumed to be scf/day/psi even if
only liquid flow is encountered (not STB/day/psi as is usual for liquids).
*MULT
This subkeyword identifies that the current well indices and phase PIs will
be multiplied by the values which follow this keyword. Existing well indices
that may have been modified by previous uses of *SETPI will be multiplied.
User's Guide IMEX
*MULTO
This subkeyword identifies that unmodified/original well indices and phase
PIs will be multiplied by the values that follow this keyword. Existing well
indices that may have been modified/multiplied by previous uses of *SETPI
will be ignored. The well indices that are multiplied by this keyword are
directly obtained from actual model/reservoir properties.
A *MULTO factor of 1.00 is useful to return to the actual well productivity
after several applications of the *SETPI keyword.
well_names
on the same line as the *SETPI keyword.
on its own to replace all wells (e.g. *SETPI *LIQ '*' or *SETPI *WAT 'wel*').
? replaces any single character anywhere in the well name (e.g. *SETPI
*GAS '?ell1').
well_numbers
numbers must be on the same line as the *SETPI keyword. If more than one line
is required to specify all the wells then the *SETPI keyword must be repeated.
values
single value for all wells) specifying the new value of the primary operating
constraint. Values must appear on one or more new lines immediately
following the *SETPI line.
DEFAULTS:
User's Guide IMEX
CONDITIONS:
*SETPI must be located in the Well and Recurrent Data keyword group, and may appear
anywhere in this keyword group following the initial specification and perforation of the
wells in the *SETPI list.
The use of *SETPI is only valid for producers and total mobility weighted injectors.
*SETPI calculates layer productivity ignoring the effects of both the *TURB and *QUAD
keywords. That is: turbulence and quadratic pressure dependence is ignored in the
productivity index calculation used to generate the wells modified well index.
*SETPI does not account for crossflow when working out productivity indices. If backflow
does occur, the wells actual productivity will differ from the *SETPI value.
When PI is set for phases which currently do not exist at the well perforations, the following
structure of defaults is used:
*GAS
*WAT
*OIL
*LIQ
reverts to *TOT
reverts to *LIQ
reverts to *LIQ
reverts to *TOT
A warning is issued when this occurs.

If total mobility was found to be zero (an unlikely event), then the *SETPI value will be
ignored for that well.
EXPLANATION:
A wells productivity index is defined as the ratio of a production rate divided by a drawdown.
The production rate may be gas production, liquid production, oil, water production or total
production, hence *SETPI can be used to set any one of these.
The essence of the *SETPI calculation is to allow the simulator to first generate its estimate
of well productivity based on the well index calculated and the mobility of each phase. The
wells basic productivity with respect to any of the phases is then available.
The new desired productivity (oil, water, gas, liquid, or total) is used to scale the well index
to produce the desired value.
The wells productivity will naturally change with time due to changes in the fluids near the
wellbore. *SETPI can be entered whenever a well test was performed to account for known
changes in well productivity.
The *SETPI calculation does not include solution gas when an oil productivity is specified, it
is assumed the gas is separated from the oil before oil flow is measured. The PIs which are
printed in the detailed well summary using *WPRN *WELL *ALL are calculated in reservoir
conditions and use each phases formation volume factor to bring the result to surface
conditions. Gas which comes out of solution is ignored.
User's Guide IMEX
Example 1:
*INUNIT
*FIELD
*WELL 1 PRODUCER
*WELL 2 INJECTOR
**
rad
*GEOMETRY *K 0.375
geofac
0.2488
wfrac
1.0
skin
-1.0
*PERF *GEO 1
** if
jf
kf
ff
(setn)
12
6
2:4
1.0
1
13
6
5
.5
2
*PERF *KH 2
** if
jf
kf
kh
di
(setn)
1
7
1:5
20
0.00002
2
2
7
1:5
30
0.00004
1
*SETPI *LIQ 1
5.0
*SETPI *GAS INJECTOR
1.0d05
In the above example, the SETPI would force the liquid well productivity for WELL NO. 1 to
be 5.0 bbl/day/psi. However, the original well index is retained and can be used if MULTO
(multiply original) is used. For WELL NO. 2, the gas injectivity index will be set to 1.0e5
ft3/day/psi, without taking the (TURB) or (QUAD) into effect.
Example 2:
*INUNIT *SI
*WELL 1 PRODUCER
*WELL 2 INJECTOR
**
rad
geofac
wfrac
skin
*GEOMETRY *K 0.375
0.2488
1.0
-1.0
*PERF *GEO 1
** if
jf kf
ff
(setn)
12
6
2:4
1.0
1
*PERF *KH 2
** if
jf kf
kh
di
(setn)
1
7 1:5
20
0.00002
2
** This will set the liquid well productivity for
** WELL No. 1 to 5.0 m3/day/kPa.
*SETPI *LIQ 1
5.0
** This will set the water well injectivity for WELL
** No. 2 to 8.0 m3/day/kPa.
*SETPI *WAT 2
8.0
DATE 1990 1 1
** This will set the total well injectivity for both
** wells to 7.0 m3/day/kPa.
*SETPI *TOT 1:2
7.0
User's Guide IMEX
*DATE 1990 2 1
** This will reset the well index to its original value
*SETPI *MULTO 1:2
1.0
*DATE 1990 3 1
** This will multiply the latest well index for both
**wells by 2.3
*SETPI *MULT *
2.3
*DATE 1990 4 1
**This will multiply the latest well index for WELL No.
**1 by 1.5 and by 2.3 for WELL No. 2. For WELL No. 1
**the new well index is 3.45 ( 2.3*1.5) times the
**original , while for WELL No. 2 the current well
**index is 7.36 (2.3*3.2) times the original.
*SETPI *MULT 1:2
1.5 3.2
*INJECTOR 4
*MONITOR *MAX *WATER 100.0
:
*TIME 1200.
** At a later date, want to adjust the operating
** well_numbers
*ALTER 1:2
3
4
** values
2*1000.0 800.0 50.0
User's Guide IMEX
User-Specified Reference Depth for Well BHP (Optional)

*BHPDEPTH
PURPOSE:
*BHPDEPTH allows the user to specify a depth to which a wells bottom hole pressure is
referred. When *BHPDEPTH has been specified for a well, the BHP for that well is in
general not one of the wellbore completion pressures. The related keyword *BHPGRAD can
be used to specify a pressure gradient used to determine the pressure difference between the
reference depth and the reference completion.
FORMAT:
*BHPDEPTH well_list
depth_values
DEFINITIONS:
well_list
A set of 'well_names' or well_numbers; see below.
well_names
specification of reference depth applies. Limited wildcarding is available for
the list of well names; please see the manual page for the *SHUTIN keyword
for an explanation of the wildcard facility.
well_numbers
Any number of integers, or integer ranges, to specify the wells to which this
specification of reference depth applies.
depth_values
A list consisting of either non-negative real numbers ( m | ft) or the
subkeyword *DEFAULT. If the depth_values list contains only a single
entry then this entry will be applied to all wells in well_list; if there are more
than a single entry in the depth_values list, the number of entries must match
the number of wells in well_list and the first depth value is applied to the first
well, etc. The numbers representing depths must contain a decimal point to
identify them as real numbers and distinguish them from well numbers.
*DEFAULT
When *DEFAULT occurs in the depth_value list, it restores the well to the
default state in which the bottom hole pressure is the wellbore pressure in the
reference completion.
DEFAULTS:
Optional keyword. If BHPDEPTH does not appear in the data set, all wells have bottom hole
pressure equal to the wellbore pressure in the reference completion. The occurrence of
BHPDEPTH does not affect the operation of any well not named in a well list following
BHPDEPTH.
User's Guide IMEX
CONDITIONS:
If it appears, this keyword must be located in the WELL AND RECURRENT DATA
keyword group. It must appear AFTER (but not necessarily immediately after) the first
*DATE line. All wells appearing in the well list following *BHPDEPTH must already have
been defined in *WELL lines. The effect of multiple *BHPDEPTH lines is cumulative; that
is, if *BHPDEPTH appears first followed by one well list and later followed by another, at
the end of the second occurrence the wells in both lists will have the specified BHP reference
depths. A well which has previously had a reference depth defined can be restored to having
the BHP be the wellbore pressure in the reference completion by entering *DEFAULT for
that well in another *BHPDEPTH line. If a well has one reference depth set in an earlier
*BHPDEPTH line and then appears in a later *BHPDEPTH list with a different depth value,
the more recently specified value overwrites the earlier value and is used in BHP calculations
after its entry. The depth values must contain a decimal point to identify them as real
numbers and distinguish them from well numbers. There is no restriction on the distribution
of the well list or depth value list over different lines in the data set; the depth values may
begin on the same line as the last well identifier in the list and continue over as many lines as
necessary. The well list too may be spread over more than a single line.
EXPLANATION:
When a BHP reference depth is entered for a well using *BHPDEPTH, the BHP differs from
the wellbore pressure in the wells reference completion. The pressure difference is equal to
delp = g*rho*(ref_depth completion_depth)
Here g is the gravitational acceleration and rho is a mobility-weighted fluid density
characteristic of the reference completion. Using the *BHPGRAD keyword the user may
enter a pressure gradient which replaces g*rho in the above formula. See the manual entry
for the *PERF keyword for details of how the reference layer is specified.
Example: The sequence
*BHPDEPTH Prod1 Prod2 Inj1 Inj2
1500. 1500. 1000. 1000.
Assigns BHP reference depths of 1500 units (feet or meters) to the two producers and of 1000
units to the two injectors. In this example the depth value list begins (but need not have
begun) a new line of the data set.
User's Guide IMEX
User-Specified Pressure Gradient For Reference Depth for

Well BHP (Optional)
*BHPGRAD
PURPOSE:
*BHPGRAD allows the user to specify a pressure gradient which is used to compute the
pressure difference between a wells reference completion and the reference depth to which
the wells bottom hole pressure is referred. The entered gradient has no effect unless a
reference depth for the well has been specified using *BHPDEPTH.
FORMAT:
*BHPGRAD well_list
gradient_values
DEFINITIONS:
well_list
well_names
specification of pressure gradient applies. Limited wildcarding is available
for the list of well names; please see the manual page for the *SHUTIN
keyword for an explanation of the wildcard facility.
well_numbers
Any number of integers, or integer ranges, to specify the wells to which this
specification of pressure gradient applies.
gradient_values
A list consisting of either non-negative real numbers (kPa/m | psi/ft | kPa/cm |
kg/cm2/m) or the subkeyword *DEFAULT. If the gradient_values list
contains only a single entry then this entry will be applied to all wells in
well_list. If there are more than a single entry in the depth_values list, the
number of entries must match the number of wells in well_list and the first
gradient value is applied to the first well, etc. The numbers representing
gradients must contain a decimal point to identify them as real numbers and
distinguish them from well numbers.
*DEFAULT
When *DEFAULT occurs in the gradient_value list, it restores the well to the
default state in which a mobility-weighted fluid density in the reference
completion is used to determine the pressure difference between reference
depth and the reference completion.
User's Guide IMEX
DEFAULTS:
Optional keyword. If BHPGRAD does not appear in the data set, all wells specified under a
*BHPDEPTH list have the pressure difference between reference completion and reference
depth calculated using a fluid density characteristic of the reference completion. The
occurrence of BHPGRAD does not affect the operation of any well not named in a well list
following BHPGRAD, and has no effect unless the well also occurs in a list following the
*BHPDEPTH keyword.
CONDITIONS:
*DATE line. All wells appearing in the well list following *BHPGRAD must already have
been defined in *WELL lines. The effect of multiple *BHPGRAD lines is cumulative; that
is, if *BHPGRAD appears first followed by one well list and later followed by another, at the
end of the second occurrence the wells in both lists will have the specified BHP reference
depths. A well which has previously had a reference gradient defined can be restored to the
default calculation using a weighted density by entering *DEFAULT for that well in another
*BHPGRAD line. If a well has one gradient set in an earlier *BHPGRAD line and then
appears in a later *BHPGRAD list with a different gradient value, the more recently specified
value overwrites the earlier value and is used in BHP calculations after its entry. The
gradient values must contain a decimal point to identify them as real numbers and distinguish
them from well numbers. There is no restriction upon the distribution of the well list or
gradient value list over different lines in the data set; the gradient values may begin on the
same line as the last well identifier in the list and continue over as many lines as necessary.
The well list too may be spread over more than a single line.
EXPLANATION:
When a BHP pressure gradient is entered for a well using *BHPGRAD, the pressure
difference between the reference completion and the reference depth is calculated as
delp = p_grad*(ref_depth completion_depth)
See the manual entry for the *PERF keyword for details of how the reference layer is specified.
Example: The sequence
*BHPGRAD Prod1 Prod2 Inj1 Inj2
0.5 0.5 0.4 0.4
assigns reference pressure gradients of 0.5 psi/ft to the two producers and of 0.4 psi/ft to the
two injectors. In this example the gradient value list begins (but need not have begun) a new
line of the data set.
User's Guide IMEX
Pressure Gradients for Calculation of Pressure Differences

Between Completions (Conditional)
*LAYERGRAD
PURPOSE:
*LAYERGRAD allows the user to specify pressure gradients to be used in a static calculation
of the pressure difference between adjacent completions within a wellbore. This keyword
can be used, for example, to model pumped-off producing wells.
FORMAT:
*LAYERGRAD wname
{location}
:
(pressure_gradient_value|*DEFAULT)
:
DEFINITIONS:
wname
Single well name within single quotes specifying the well to which the layer
pressure gradient specifications apply. No wild-carding is supported in this
situation.
{location}
if jf kf / ir1 jr1 kr1 { / { / irn jrn krn} }
These triples of integers specify the layers of well wname to be assigned userspecified head pressure gradients. It is valid to name some of well wnames
layers and not others in the *LAYERGRAD lines, but layers not named do not
acquire a special gradient value and have heads calculated in the normal way.
Any layer named under *LAYERGRAD must already have been defined for
the well using a *PERF, or *PERFV statement. The gradient specified will be
used for the head calculation between the named layer and the next heel-ward
completion (the completion to which the named completion flows see the
discussion in the manual page for the *PERF keyword).
pressure_gradient_value
Non-negative real value numbers (kPa/m | psi/ft | kPa/cm | kg/cm2/m)
specifying the gradient to be used for the pressure difference calculation.
*DEFAULT
Specifies that the pressure difference between the named layer and the next
heel-ward layer should be calculated normally, not using a specified pressure
gradient value. The head calculation method in place can be specified using
the *HEAD-METHOD keyword.
if
A single integer specifying the I direction grid block index of the
fundamental grid (see explanation).
User's Guide IMEX
jf
A single integer specifying the J direction grid block index of the
fundamental grid (See explanation).
kf
A single integer specifying the K direction grid block index of the refined
grid (See explanation).
ir1 jr1 kr1
Three integers specifying the I, J, K direction grid block index of the first
level of refined grid that the well is completed through.
irn jrn krn
Three integers specifying the I, J, K direction grid block index of the n'th
DEFAULTS:
Layers named under *LAYERGRAD with a gradient value (as opposed to the *DEFAULT
subkeyword) are flagged as receiving special treatment during the layer head calculation.
The default status is assigned to all layers at the beginning of the run. The default status can
be re-imposed through the *DEFAULT subkeyword.
CONDITIONS:
The named layers must all have been previously created for well wname with *PERF or
*PERFV lines. Either a non-negative real number or the subkeyword *DEFAULT must follow
the layer identification or an error is generated. Not all of a wells layers need to be named under
*LAYERGRAD; those omitted have heads calculated in the default manner.
EXPLANATION:
When no pressure gradient is specified for a layer, the pressure difference in the wellbore
between one layer and the layer to which it flows is calculated either statically, using a
pressure gradient depending on local fluid densities, or using a correlation which includes
frictional effects. See the manual page for the *HEAD-METHOD keyword for a description
of the head calculation methods available. When a head pressure gradient hgrad is specified
with *LAYERGRAD, the pressure difference between the named layer and the next heelward completion in the well (which is unique) is calculated as
Delp = (P_next P_named) = hgrad*(depth_next depth_named)
See the manual entry for the *PERF keyword for an explanation of how the identity of the
next heel-ward layer (the layer to which the named layer flows) is determined. The layer
named first in the *PERF.. lines flows to the surface and no layer head is associated with the
corresponding pressure difference; hence any specification of a head gradient for such a
layer is ignored (but is valid).
The *LAYERGRAD keyword can be used to model pumped-off producing wells by
specifying zero or gas-like pressure gradients between completions which are above the
pump level.
User's Guide IMEX
EXAMPLES:
Example 1:
*LAYERGRAD
65 23 5
65 23 6
WELL-NNE17
0.5
0.4
Completions for well WELL-NNE17 in blocks 65 23 5 and 65 23 6 must already have been
created with a *PERF.. line for the above to be valid.
Example 2:
*LAYERGRAD
16 48 11
16 48 12
16 48 13
16 48 14
PUMPED-WELL
0.
0.
0.
0.
If well PUMPED-WELL has completions in blocks 16 48 10:15, with layer 15 deepest, the
above represent a pump located just above the perforation in block 16 48 14. The pressure
difference in the wellbore between blocks 16 48 15 and 16 48 14 is calculated by the usual
head method for the run, but all heads above this are set to zero, to simulate the absence of
liquids above the completion in layer 14. Note that no gradient specification for the
completion in block 16 48 10 is necessary because this layer flows to the surface.
User's Guide IMEX
Alter Polymer Concentration (Optional)
*ALTERCP
PURPOSE:
*ALTERCP allows modification of the polymer concentration given originally with

*INCOMP *WATER (cpvalue).
FORMAT:
*ALTERCP
(well_numbers)
('well_names')
cpvalues
DEFINITIONS:
well_numbers
which this alteration of polymer concentration applies. These well numbers
must be on the same line as the *ALTERCP keyword. If more than one line
is required to specify all the wells, then the *ALTERCP keyword must be
repeated.
well_names
alteration of polymer concentration applies. These names must be on the
same line as the *ALTERCP keyword. If more than one line is required for
the well list, then the *ALTERCP keyword must be repeated.
'?ell1').
cpvalues
One number for each well entered in the list of well names or well numbers
(or one value for all wells), specifying the new value of the polymer
concentration (kg/m3 | lb/STB | g/cm3). Values must appear on one or more
new lines immediately following the *ALTERCP line.
If only one value of 'values' is input all wells defined in the well list will be
User's Guide IMEX
DEFAULTS:

CONDITIONS:
*ALTERCP must be located in the Well and Recurrent Data keyword group, and may appear
anywhere in this keyword group following the well definitions.
EXPLANATION:
This optional keyword is used to alter the polymer concentration for an injection well or a
group of injection wells. *ALTERCP requires the same format as *ALTER.
Examples:
*MODEL *POLY
:
*DATE
*WELL 1
:
*INJECTOR 1
*INCOMP *WATER 0.7
:
*DATE 1989 09 01
*ALTERCP 1
0.0
User's Guide IMEX
Alter Seawater Injection Volume Fraction (Optional)
*ALTERFS
PURPOSE:
*ALTERFS allows modification of the seawater injection volume fractions given originally
with *INCOMP *SEAWATER (fsvalue).
FORMAT:
*ALTERFS
(well_numbers)
('well_names')
fsvalues
DEFINITIONS:
well_numbers
which this alteration of seawater fraction applies. These well numbers must be
on the same line as the *ALTERFS keyword. If more than one line is required
to specify all the wells, then the *ALTERFS keyword must be repeated.
well_names
alteration of seawater fraction applies. These names must be on the same line
as the *ALTERFS keyword. If more than one line is required for the well
list, then the *ALTERFS keyword must be repeated.
(See *TRIGGER for details)
fsvalues
One number for each well entered in the list of well names or well numbers
(or one value for all wells), specifying the new value of the seawater volume
fraction. Values must appear on one or more new lines immediately
following the *ALTERFS line.
If only one value of 'fsvalue' is input all wells defined in the well list will be
DEFAULTS:

User's Guide IMEX
CONDITIONS:
*ALTERFS must be located in the Well and Recurrent Data keyword group, and may appear
anywhere in this keyword group following the well definitions.
EXPLANATION:
This optional keyword is used to alter the seawater injection volume fraction for an injection
well or a group of injection wells. *ALTERFS requires the same format as *ALTER.
Examples:
:
*DATE 1979 01 01
*WELL 1
:
*INJECTOR 1
*INCOMP *SEAWATER 0.95
:
*DATE 1989 09 01
*ALTERFS 1
1.0
User's Guide IMEX
Well Scale Table Assignment (Optional)
*SCLTBL-WELL
PURPOSE:
*SCLTBL-WELL assigns deposition and damage tables (defined in the Rock-Fluid data
section to wells.
FORMAT:
*SCLTBL-WELL ('well_names' | well_numbers)

Deposition Table Set Number
Damage Table Set Number
DEFINITIONS:
well_names
Any number of well names (in quotes) to specify the wells to which the Scale
table definitions applies. These names must be on the same line as the
*SCLTBL-WELL keyword. If more than one line is required for the well
list, then the *SCLTBL-WELL keyword must be repeated.
well_numbers
Any number of integers, or range of integers to specify the wells to which
*SCLTBL-WELL applies. These well numbers must be on the same line as
the *SCLTBL-WELL keyword. If more than one line is required to specify
all the wells, then the *SCLTBL-WELL keyword must be repeated.
*SCLTBL-WELL
Keyword used to enter both a Deposition Table Set Number and a Damage
Table Set Number for the list of wells defined either by well_names and/or
by well numbers. Only two integers are entered on the second line of this
keyword.
The set number of the deposition table used for this well. New perforations
defined after this keyword is set will also use the defined deposition table
number.
User's Guide IMEX

The set number of the damage table used for this well. New perforations
defined after this keyword is encountered will also use the defined damage
table number for the new perforations.
DEFAULTS:
Conditional keyword. There are no defaults. If not defined the well will not experience
scaling damage.
CONDITIONS:
This keyword must be in the Recurrent Data keyword group.

EXPLANATION:
The deposition and damage tables defined in the rock-fluid data section using *SCLDPS and
*SCLDMG are applied to the wells listed on the *SCLTBL-WELL keyword. Each
combination of deposition and damage table entry would require a separate *SCLTBLWELL keyword. After the reading of *SCLTBL-WELL, if the well is reperforated, new
perforations would also use the deposition and damage tables defined on *SCLTBL-WELL.
*SCLTBL-WELL deposition and damage table values can be overwritten by either a
subsequent *SCLTBL-LAYER keyword (for specified layers) or by a subsequent *SCLTBLWELL keyword.
User's Guide IMEX
Layer Scale Table Assignment (Optional)
*SCLTBL-LAYER
PURPOSE:
*SCLTBL-LAYER assigns deposition and damage tables (defined in the Rock-Fluid data
section to multiple specified layers of a single well.
FORMAT:
*SCLTBL-LAYER ('well_name' | well_number)

{Location}
:

:

:
DEFINITIONS:
well_name
A single well name (in quotes)
well_number
A single well number.
{location}
*SCLTBL-LAYER
Keyword used to enter both a Deposition Table Set Number and a Damage
Table Set Number for each specified layer defined in {Location} of the well
defined by well_name or well_number.
The set number of the deposition table used for this layer.
The set number of the damage table used for this layer.
DEFAULTS:
Conditional keyword. There are no defaults. Any number of layers can be entered per
instance of this keyword.
CONDITIONS:
This keyword must be in the Recurrent Data keyword group. It is not necessary to enter all
layers.
User's Guide IMEX
EXPLANATION:
The deposition and damage tables defined in the rock-fluid data section using *SCLDPS and
*SCLDMG are applied to the layers of the well listed on the *SCLTBL-LAYER keyword.
*SCLTBL-LAYER need not be specified for each layer. Unspecified layers will either not
experience scale damage (if *SCLTBL-WELL was not specified) or will use the deposition
and damage tables specified on a previous *SCLTBL-WELL keyword.
*SCLTBL-LAYER deposition and damage table values can be overwritten by either a
subsequent *SCLTBL-LAYER keyword (for specified layers) or by a subsequent *SCLTBLWELL keyword.
User's Guide IMEX
Well Scale Removal (Optional)
*SCLRMV-WELL
PURPOSE:
*SCLRMV-WELL removes a fraction of the accumulated deposited scale from a well.

FORMAT:
*SCLRMV-WELL ('well_names' | well_numbers)

value
DEFINITIONS:
well_names
Any number of well names (in quotes) to specify the wells to which the scale
removal applies. These names must be on the same line as the *SCLRMVWELL keyword. If more than one line is required for the well list, then the
*SCLRMV-WELL keyword must be repeated.
'?ell1').
well_numbers
Any number of integers, or range of integers to specify the wells to which
*SCLRMV-WELL applies. These well numbers must be on the same line as
the *SCLRMV-WELL keyword. If more than one line is required to specify
all the wells, then the *SCLRMV-WELL keyword must be repeated.
*SCLRMV-WELL
Keyword used to enter a well_name or well_number and a single value
which represents the fraction of scale to be removed.
value
The fraction of cumulative scale deposited which is removed A single value
is entered for all listed wells. A value of 1.0 implies complete removal of
accumulated scale.
User's Guide IMEX
DEFAULTS:
Conditional keyword. If value is omitted a value of 1.0 (i.e. complete scale removal is
assumed).
CONDITIONS:
This keyword must be in the Recurrent Data keyword group. Value must lie between 0.00
and 1.00. There is only one value of the fraction removed entered per instance of the
SCLRMV-WELL keyword
EXPLANATION:
This keyword allows the user to completely or partially remove the scale built up on a well
over time. This will completely or partially reverse the damage to the wells productivity.
User's Guide IMEX
Layer Scale Removal (Optional)
*SCLRMV-LAYER
PURPOSE:
*SCLRMV-LAYER removes a fraction of the accumulated deposited scale from multiple

specified perforations of a single well.
FORMAT:
*SCLRMV-LAYER ('well_name' | well_number)

{Location}
Value
:
:
DEFINITIONS:
well_name
A single well name (in quotes)
well_numbers
A single well number.
{location}
*SCLRMV-LAYER
Keyword used to enter a fraction of the scale removed for multiple specified
layers defined in {Location} of the well defined by well_name or
well_number.
value
The fraction of cumulative scale deposited which is removed. A value of 1.0
implies complete removal of accumulated scale.
DEFAULTS:
Conditional keyword. If value is omitted, a value of 1.0 (i.e. complete scale removal) is
assumed.
CONDITIONS:
This keyword must be in the Recurrent Data keyword group. Values must lie between 0.00
and 1.00. Any number of layers can be entered per instance of this keyword.
EXPLANATION:
This keyword allows the user to completely or partially remove the scale built up on
perforations over time. This will completely or partially reverse the damage to the
perforations productivity.
User's Guide IMEX
Data for Workover Action for Wells (Optional)

*WORKOVER-DATA
PURPOSE
*WORKOVER-DATA introduces the input of parameters which control the *WORKOVER

action taken when a producing well violates a minimum rate monitored constraint which has
*WORKOVER specified as the action to be taken in the event that the constraint is violated.
The *WORKOVER action is specified using the *MONITOR keyword. The workover
parameters which can be entered using *WORKOVER-DATA can also be entered after the
*WORKOVER action subkeyword of the *MONITOR keyword; *WORKOVER-DATA is
somewhat more convenient to use and is meant to supplant parameter entry after
*WORKOVER, which however is still supported.
FORMAT:
*WORKOVER-DATA well_list
pi_factor ntimes wo_action
DEFINITIONS:
well_list
A set of 'well_names' or well_numbers to which this specification of
*WORKOVER parameters applies; see below. No error is flagged if an
injector is included in the well list, but the *WORKOVER-DATA entries can
be used only for producers, since no *WORKOVER action is available for
injectors.
well_names
specification of *WORKOVER parameters applies. Limited wildcarding is
well_numbers
which this specification of *WORKOVER parameters applies.
pi_factor
A real number specifying the factor by which all the layer productivities for
the well are multiplied: New_PI = Old_PI * pi_factor. This value must
appear on the first line after the *WORKOVER-DATA line.
ntimes
An integer specifying the maximum number of workovers allowed before
wo_action is taken instead of the productivity index increment. This value
must appear on the first line after the *WORKOVER-DATA line.
User's Guide IMEX
wo_action
Subkeyword specifying the action to be taken if the monitor is violated after
the maximum number of workovers has been exhausted. wo_action can be
one of the following: *STOP, *SHUTIN or *SHUTIN *REPEAT. The
default is *SHUTIN. This subkeyword must appear on the first line after the
*WORKOVER-DATA line.
*STOP
This subkeyword specifies that if a monitored constraint with the
*WORKOVER action is violated, and the maximum number of workovers
for the well has been reached, the simulation should be terminated.
*SHUTIN
This subkeyword specifies that if a monitored constraint with the
*WORKOVER action is violated, and the maximum number of workovers
for the well has been reached, the well should be shut and the simulation
should advance to the next timestep. This is the default remedial action for a
*WORKOVER constraint violation after the maximum number of workovers
has been applied.
*REPEAT
This subkeyword of the *SHUTIN action specifies that if a monitored
constraint with the *WORKOVER action is violated, and the maximum
number of workovers for the well has been reached, the well should be shut
and the timestep should be repeated.
DEFAULTS:
If a *WORKOVER action is specified for a violation of a monitored producer constraint under

the *MONITOR keyword, but pi_factor, ntimes, and wo_action have not been specified for the
well either after the *WORKOVER subkeyword or with the *WORKOVER-DATA keyword,
the defaults are pi_factor = 1.2, ntimes = 10, wo_action = *SHUTIN.
If pi_factor, ntimes, and wo_action have been specified for a well and later new values are
entered for the well either using *WORKOVER-DATA of the *WORKOVER subkeyword
of *MONITOR, the later values replace the earlier.
CONDITIONS:
follow the definition (via the *WELL keyword) of the wells listed in well_list.
Only a single set of pi_factor ntimes wo_action may be entered with a single
*WORKOVER-DATA line; these will be applied to all wells in the list.
All well names or well numbers in well_list must appear on the same line as *WORKOVERDATA; if more wells must be listed than can fit on the line, a second *WORKOVER-DATA
line must be entered.
User's Guide IMEX
EXPLANATION:
The *WORKOVER action can be specified for violations of monitored minimum rate
producing well constraints. Three parameters are used in applying a workover: the workover
factor, the maximum number of workovers to be applied to the well, and the alternative
action to be taken if the constraint is violated but the well's maximum number of workovers
has already been applied.
The number of workovers for a well is counted independently of the identity of the constraint
whose violation has called for the workover. For example, if a producer has the workover
action specified both for a *MIN *STO monitor and for a *MIN *STG monitor, then
violations of either of these constraints increases the number of workovers applied to the
well. However, no more than one workover is applied during a single timestep; for example,
if a producer has *WORKOVER specified for both *MIN *STO and *MIN STG and both are
violated during the same timestep, the productivity indices for the well's completions will be
multiplied by pi_factor only once and the counter numbering the workovers applied to the
well will be incremented only by one. However if the *MIN *STO constraint is violated in
one timestep and the *MIN *STG constraint in the next, two workovers will be performed
and counted toward the maximum ntimes.
Example:
*WORKOVER-DATA 'PROD*'
1.3 12 *STOP
The above pair of lines assigns a productivity index factor of 1.3, a maximum number of
workovers of 12, and an action after maximum workovers of termination of simulation for all
wells whose names begin with the four characters PROD. The workover will only be applied
if *WORKOVER is specified as a producer minimum rate monitored constraint violation
action under the *MONITOR keyword.
User's Guide IMEX
Resetting Well Operating Constraint after Value Change

(Optional)
*MRC-RESET
PURPOSE:
*MRC-RESET allows the user to specify on a well-by- well basis that a well should or
should not beset to its Most Restrictive Constraint after a change in the value of an operating
constraint (through data read under the *OPERATE, *ALTER, or *TARGET keywords).
FORMAT:
*MRC-RESET well_list
(*RESET)
(*NO-RESET)
DEFINITIONS:
well_list
well_names
alteration of initialization frequency applies. Limited wildcarding is
well_numbers
which this alteration of initialization frequency applies.
RESET
Specifies that a determination of the most restrictive well operating
constraint should be done for all wells in well_list after a constraint value is
changed through data and that the currently most restrictive constraint should
be set as the current operating constraint for the wells before the next
timestep is carried out.
NO-RESET
Specifies that no operating constraint change should be made for the wells in
well_list after a change in constraint values through data entry. Changes in
operating constraint may occur after the next timestep as a result of
constraint violations.
DEFAULTS:
Optional keyword. All wells are set in the *RESET mode at the beginning of the simulation
by default; any changes entered under *MRC-RESET are cumulative. If no subkeyword
appears after the well list, *RESET is assumed.
User's Guide IMEX
CONDITIONS:
If it appears, this keyword must be located in the Well and Recurrent Data keyword group. It
must appear AFTER (but not necessarily immediately after) the first *DATE line. The wells
in the list included on the *MRC-RESET line must have already been defined using the
*WELL keyword.
EXPLANATION:
When operating constraint values are changed the type of the most restrictive constraint may
also change; for example, when a rate constraint value is increased, the bottom hole pressure
may become the most restrictive constraint. If no checking is done for this shift in the most
restrictive constraint, the simulator must converge a timestep on the original constraint and let
the constraint switch be accomplished as a result of constraint violation. If this next timestep
must, for example, be run at a very high rate, converging the timestep may be quite difficult.
To circumvent this potential difficulty, the default is to check all wells which have undergone
a change in constraint value through use of the *OPERATE, *ALTER, or *TARGET
keywords to determine the most restrictive constraint and set the well on this most restrictive
constraint before the next timestep. The *NO-RESET option is provided if, for some reason,
the user wishes to over-ride this automatic constraint switch.
Example:
The sequence
*MRC-RESET
'SPECIAL-PRODUCER' *NO-RESET
*TARGET *STO 'SPECIAL-PRODUCER'
500.0
would have the oil rate constraint value of well SPECIAL-PRODUCER set to 500 units per
day and the operating constraint would stay as it is (presumably on the oil rate constraint),
regardless of whether the oil rate constraint is the most restrictive constraint at the beginning
of a timestep.
User's Guide IMEX
Gas Lift Option (Optional)
*GLIFT
PURPOSE:
*GLIFT specifies lift gas rates for a given set of production wells, or the method by which
the lift gas rate is to be calculated.
FORMAT:
(*RATE) ('well_names' | well_numbers)

(*GLR)
(*OPT)
*GLIFT
values
DEFINITIONS:
well_names
on the same line as the *GLIFT keyword. If more than one line is required
for the well list, then the *GLIFT keyword must be repeated.
'?ell1').
well_numbers
numbers must be on the same line as the *GLIFT keyword. If more than one
line is required to specify all the wells, then the *GLIFT keyword must be
repeated.
*RATE
When this subkeyword is specified, gas lift injection rates must be specified
directly. When all of *RATE, *GLR and *OPT are missing this is the default.
For this option the units of 'values' are (m3/day | SCF/day | cm3/min | m3/day).
User's Guide IMEX
*GLR
When this subkeyword is specified, gas lift injection rates are calculated using
produced gas-liquid ratios from the previous timestep (explicitly). When all of
*RATE, *GLR and *OPT are missing *RATE is the default. For this option
the units of 'values' are (m3/m3 | SCF/STB | cm3/cm3 | m3/m3).
*OPT
When this subkeyword is specified, gas lift injection rates are calculated
automatically according to the specifications listed under the *GLOPT
keyword. An error will occur if the *OPT subkeyword is specified for a well
and the *GLOPT keyword is missing. When all of *RATE, *GLR and *OPT
are missing, *RATE is the default. For this option 'values' are not required,
but MAY be entered, in which case they are interpreted as maximum lift gas
rates for the wells specified by well_names or well_numbers. The units of
the values are (m3/day | SCF/day | cm3/min | m3/day) when used with *OPT.
When no values are entered, very large maxima (1.0d+15) are assigned as the
maximum lift gas rates for the listed wells.
values
One number for each well identified by well_numbers or 'well_names' (or one
value for all wells on the list) specifying the new value of the lift gas rate, or gasliquid ratio. Values must appear on one or more new lines immediately
following the *GLIFT keyword. Values must NOT appear on the same line as
the *GLIFT keyword. If the *OPT option is specified, then these values are not
required, but may be entered as described above. See *RATE, *GLR, and *OPT
above for the units of the values. If the angular well fraction wfrac entered by
the user under the *GEOMETRY keyword is less than one, it is assumed that lift
gas rates also correspond to the well fraction values used; i.e., if the well fraction
is 0.25, then the lift gas rate input by the user (or value obtained by the internal
optimization algorithm) is 0.25 the rate for the whole well.
If only one value of `values' is input all wells defined in the well list will be
DEFAULTS:
Optional keyword. If the keyword is missing then gas lift rates are zero. The default
suboption is *RATE; if the well list follows *GLIFT immediately, values following the list
are interpreted as lift gas rates.
CONDITIONS:
This keyword must be located in the Well and Recurrent Data keyword group. It may appear
anywhere in this keyword group following the *PRODUCER keyword(s) for the well_numbers. If
*GLIFT is specified for a particular well then *PWELLBORE must also be specified for that well.
A well must have its type defined as *PRODUCER before it appears in a well list following
*GLIFT; otherwise an error message is printed and the simulation is terminated.
User's Guide IMEX
EXPLANATION:
*GLIFT allows for the specification of a gas lift injection rate (or the specification of the
method by which this rate is computed) for any producing well. These rates can be modified at
different *DATE or *TIME keywords by using the *GLIFT keyword again. The lift gas rate is
added to the producer's formation gas rate before calculating the wellbore pressure drop and
well liquid production rates. Each well's tubing hydraulics table specified by the
*PWELLBORE keyword and entered using the *PTUBE1 keyword should be constructed by
the user to include large enough ranges of gas-oil ratios to model the effects of gas lift
operations.
Gas lift rates can be specified in three ways:
1. Input lift gas rates directly using the *GLIFT *RATE keywords.
2. Input a gas-liquid producing ratio using the *GLIFT *GLR keywords. The lift gas
injection rates will be calculated from the current timestep's production data.
However, the lift gas rates will not be used for the calculation of production rates
until the next timestep.
Lift_gas_inj = GLR_input * Total_liquid_est - GOR_est * Oil_rate_est
3. If *GLIFT *OPT is specified, gas injection rates will be calculated automatically
using parameters specified using the *GLOPT keyword. The values entered under
*GLIFT *OPT, if any, are maximum lift gas rates allowed for the wells.
When *GLIFT is in effect, printouts of total well and field gas rates and gas cumulatives do
not include the lift gas. Individual well layer rates also do not include lift gas. A separate
printout of lift gas rates and cumulatives as well as field totals for all wells producing under
gas lift is given.
Please note that gas rates and cumulatives output to the simulation results file (SRF) do not
include lift gas.
This option has the following assumptions:
1. The lift gas enters the production stream at or above the top reservoir grid block.
2. The wellbore gas flow rate is the sum of the formation gas flow rate and the lift gas
flow rate.
Example:
*GLIFT
1000.0
-or*GLIFT
2500.
-or*GLIFT
1 3:5 7
500.0 700.0
1 3:5 7
1500. 1800.
600.0
900.
1200.0
1600.
*OPT 1 3:5 7
SEE ALSO : *GLCONTROL, *GLOPT
User's Guide IMEX
Gas Lift Control (Optional)
*GLCONTROL
PURPOSE:
*GLCONTROL specifies gas lift controls for a given set of production wells.
FORMAT:
*GLCONTROL ('well_names' | well_numbers)

*WHP val *BHP val *WCUT val *STO val *TABLE val
DEFINITIONS:
well_names
Any number of well names (in quotes) to specify the wells to which the
*GLCONTROL applies. These names must be on the same line as the
*GLCONTROL keyword. If more than one line is required for the well list,
then the *GLCONTROL keyword must be repeated.
'?ell1').
well_numbers
Any number of integers, or range of integers to specify the wells to which the
*GLCONTROL applies. These well numbers must be on the same line as
the *GLCONTROL keyword. If more than one line is required to specify all
the wells, then the *GLCONTROL keyword must be repeated.
*WHP
When this subkeyword is specified, the well will be put on gas lift automatically
when the well-head pressure drops below 'val' (kPa | psi | kPa | kg/cm2).
*BHP
When this subkeyword is specified, the well will be put on gas lift
automatically when the bottom-hole pressure drops below 'val' (kPa | psi |
kPa | kg/cm2).
User's Guide IMEX
*WCUT
automatically when the water-cut exceeds 'val' (fraction).
*STO
automatically when the oil production rate drops below 'val' (m3/day |
STB/day | cm3/min | m3/day).
*TABLE
With the specification of this subkeyword gas lift calculations will be done
using the specified wellbore hydraulics table number (entered under
*PTUBE1), when the well is switched to gas lift automatically. If the well is
assigned zero gas lift injection rates, then the *PTUBE1 table number to be
used will revert to the table number specified with the *PWELLBORE
keyword.
val
Specified minimum or maximum switching value, or *PTUBE1 table
number.
DEFAULTS:
Optional keyword. If the keyword is missing the gas-lift wells will be put on gas lift
immediately following the *GLIFT keyword. The default wellbore hydraulics table is the
table number specified with *PWELLBORE.
CONDITIONS:
anywhere in this keyword group following the *GLIFT keyword(s) for the well_numbers. If
*GLCONTROL is specified for a particular well then *GLIFT and *PWELLBORE must also
be specified for that well. A well can appear in the list following *GLCONTROL only if it
has already had its type defined with a *PRODUCER statement; otherwise an error message
will be printed and the run will be terminated.
EXPLANATION:
*GLCONTROL allows for the specification of operating constraints for particular gas lift wells
which will be checked at every timestep for violation. If at any time one of the constraints is
violated, the well will be switched to gas lift. If a wellbore hydraulics table number is specified
using *GLCONTROL, then this table will be used for wellbore hydraulics calculations when
the well is switched to gas lift. These constraints can be modified at different *DATE or
*TIME keywords by using the *GLCONTROL keyword again.
Once a well begins receiving lift gas through a control violation, it continues to receive lift
gas even if the control condition ceases to be violated, unless the limit is reset by re-entering
the *GLCONTROL keyword.
User's Guide IMEX
No checking is done to ensure that the bottom-hole injection pressure is high enough to open the
gas lift valves. It is left to the user to determine where the lift gas will be injected into the
wellbore. Lift gas injection locations must be built into the tubing performance tables input with
the *PTUBE1 keyword. A different table must be constructed for each gas lift valve elevation.
Example:
*GLCONTROL 1 3:5 7
*WHP 100. *WCUT .75 *STO 10. *TABLE 2
SEE ALSO : *GLIFT, *GLOPT
User's Guide IMEX
Gas Lift Optimization (Optional)
*GLOPT
PURPOSE:
*GLOPT indicates the use of an option to optimize oil production by automatic allocation of
lift gas.
FORMAT:
*GLOPT gc_fact consum_rate

(*MAXGAS max_gas_rate)
(*RECYCLE 'group_name' recyc_frac make_up_vol max_rate)
(*GLOPT-TIME time_interval)
DEFINITIONS:
gc_fact
A minimum gas-cost factor, below which gas lift injection is uneconomic.
This value is actually the derivative of the oil rate vs. lift gas injection. A
realistic input value will ensure that large volumes of lift gas are not allocated
for small incremental oil volumes. (m3/m3 | STB/SCF | cm3/cm3 | m3/m3)
consum_rate
The compressor gas consumption rate per unit gas injected. (m3/m3 |
SCF/SCF | cm3/cm3 | m3/m3)
*MAXGAS
This keyword specifies the maximum amount of gas available for gas lift
injection. This total amount of gas will be automatically allocated to each
well on gas lift production. If this value is very large and a realistic value of
gc_fact has been input, then all of the available lift gas will not be used and
the values reported for total lift gas will be the maximum optimized values
for the field.
max_gas_rate
Total amount of gas available for gas lift injection. (m3/day | SCF/day |
cm3/min | m3/day)
*RECYCLE
When this subkeyword is specified, the total produced gas will be recycled
and used for gas lift injection. The total produced gas is the gas produced
from the reservoir plus the gas produced from previous gas lift injection
averaged over the previous day (assuming that it takes one day to process and
re-inject gas). The effect is cumulative, and if the compressor consumption
is small and the recycle fraction is close to one, the available lift gas will
increase with time.
'group_name'
Group name from which the gas to be injected is produced.
User's Guide IMEX
recyc_frac
Recycle fraction of gas production from a group of wells.
make_up_vol
Make up volume of gas available for injection from some source outside the
field being modelled. (m3/day | SCF/day | cm3/min | m3/day)
max_rate
Compressor maximum gas injection rate. (m3/day | SCF/day | cm3/min |
m3/day)
*GLOPT-TIME
This subkeyword specifies the time-interval between gas-lift optimization
calculations. In the time interval specified by *GLOPT-TIME the allocated
amounts of lift- gas remain at their previous values.
time_interval
The time interval between optimizations specified using the *GLOPT-TIME
keyword. The default is 182.5 days (six months). A well is put on gas lift
only at the times specified by the *GLOPT-TIME keyword. Thus if a well
requires gas lift after 10 days of simulation and the *GLOPT-TIME time
interval is 182.5 days, an additional 172.5 must pass before the well is put on
gas lift. Use of smaller time intervals are recommended if the wells are to be
put onto gas lift more rapidly.
DEFAULTS:
Optional keyword. If the keyword is missing the gas lift rates must be specified on the
*GLIFT keyword. If gc_fact is zero or missing, then the default value is 0.00005 STB/SCF,
or 0.00028m3/m3. The default value for consum_rate is 0.0 SCF/day (m3/day). The default
gas availability is *MAXGAS 1.0E+20, and there is no default for the compressor maximum
injection rates. The default *GLOPT-TIME is 182.5 days (six months).
CONDITIONS:
anywhere in this group after the relevant 'group_name' has been defined. It will only be
operational for wells which are currently on gas lift and have been targeted for optimization
using the *GLIFT *OPT option.
EXPLANATION:
When this option is specified, gas lift injection rates will be automatically optimized to
provide maximum oil production rates for the wells specified. This optimization is an
iterative procedure which uses tubing flow performance curves given by the *PTUBE1
keyword. The values in the *GLOPT option can be modified at different *DATE and *TIME
keywords by using the *GLOPT keyword again.
User's Guide IMEX
The optimization begins with the determination of the total lift gas available (TLGA) for
injection (total gas - compressor gas consumed). The TLGA may be input either directly
using *MAXGAS, or indirectly from the field's gas production rates using *RECYCLE. In
either case the user may specify that a compressor consumes some of the available gas. This
gas consumption is directly proportional to the total lift gas injected (TLGI) which may be
different from TLGA.
An iterative procedure is now started to determine the optimum lift gas allocations to each
well. The TLGA is divided into equal "packets" of gas to be allocated iteratively.
The following equation is used to optimize oil production for each "packet" of lift gas
allocated, and is summed over the entire range of TLGA:
d (oil rate )
max
SUM _ over _ packets
d (gas inj rate )
The preceding summation requires a derivative of the oil rate with respect to gas injected,
which is calculated from the tubing performance tables (*PTUBE1). During this derivative
calculation, wells which require a gas "kick off" rate to start oil production (derivative of
zero) are assumed initially to start oil production immediately with lift gas injection. Then, if
for the particular "packet" of lift gas allocation a well requiring a "kick off" gas rate turns out
to possess the maximum derivative, then enough lift gas will be assigned to that well to "kick
off" it's production.
If the derivative in the above equation is less than the minimum gas cost factor (gc_fact) then
the gas "packet" will not be assigned to that well. The gas "packet" allocation process will
end when either the TLGA is reached, or when each well has reached its minimum derivative
specified by gc_fact.
Example:
*GLOPT 0 0.19
-or*GLOPT .00002 0.05
*RECYCLE 'group1' 0.97 1.0E+6 1.0E+8
-or*GLOPT 0 0 *MAXGAS 1.0E+7
User's Guide IMEX
Well Element Geometry (Conditional)
*GEOMETRY
PURPOSE:
*GEOMETRY specifies the well geometric characteristics to be used by the simulator to

calculate the well index internally.
FORMAT:
*GEOMETRY
(*I) (rad geofac wfrac skin)

(*J)
(*K)
DEFINITIONS:
*I
This subkeyword specifies that the wellbore is parallel to the I axis.
*J
This subkeyword specifies that the wellbore is parallel to the J axis.
*K
This subkeyword specifies that the wellbore is parallel to the K axis.
rad
A real number specifying the well radius (m | ft | cm).
geofac
A real number specifying the geometric factor for the well element. This
value is not used when *GEOA or *KHA is used to define the well index
calculation on the *PERF (or *PERFV, *PERFVHY) card. The Peaceman
well index calculation used when *GEOA or *KHA is specified does not
require an external geometry factor (see below).
wfrac
A real number between 0. and 1. specifying the fraction of a circle that the
well models. (Usually one; set to one half or one quarter if the well is at the
edge or corner of the grid block on the grid boundary.)
skin
A real number specifying the well skin factor. Where:
q
re
rad
areap
=
=
=
2pi * kh * wfrac / (ln(re/rad)+skin) *

(Phase mobility) * (Pblock - Pwell)
geofac*sqrt((areap)/(3.14159*wfrac))
wellbore radius
area perpendicular to the wellbore
(DX*DY for a vertical well)
User's Guide IMEX
In the Cases where *GEOA or *KHA are used, the effective radius re is
calculated from the following expressions (Peaceman D.W.: Interpretation of
Well-Block Pressures in Numerical Reservoir Simulation with Non-Square
Grid Blocks and Anisotropic Permeability, SPEJ (June 1983) pp. 531-543.):
For a vertical well (direction = K):
1/ 2
1/ 2

DX 2 ky + DY 2 kx
ky
kx

re = 0.28
1/ 4
1/ 4
kx
ky
+
kx
ky
1/ 2
For a horizontal well parallel to the J direction (direction = J):
re = 0.28
1/ 2
1/ 2
kz
kx
DX 2
+ DZ 2
kx
kz
kz

kx
1/ 4
kx
+
kz
1/ 2
1/ 4
For a horizontal well parallel to the I direction (direction = I):
re = 0.28
1/ 2
1/ 2

2 ky
DY 2 kz
+ DZ
ky
kz
kz

ky
1/ 4
ky
+
kz
1/ 2
1/ 4
DEFAULTS:

CONDITIONS:
*GEOMETRY may be used with *INJECTOR
*MOBWEIGHT or with *PRODUCER.
It comes before any of the well completion keywords that will use GEO (*GEOA) or *KH
(*KHA). All subsequent well completion keywords will use the last *GEOMETRY card to
calculate the well index. Required if *PERF *GEO (*GEOA), *PERFV *GEO (*GEOA),
*PERFVHY *GEO (*GEOA), *PERF *KH (*KHA), *PERFV *KH (*KHA), or
*PERFVHY *KH (*KHA) are used.
User's Guide IMEX
EXPLANATION:
For example, to define two wells with identical well geometry, and a third with a larger
radius, the input data would appear as:
** 3 wells are defined.
*WELL 1 'Producer 1' *VERT 12 14
...
** The well geometries are input for wells 1,2
** rad geofac wfrac skin
*GEOMETRY *K 0.375 0.2488 1.0 0.0
** The well completion is defined for wells 1,2
*PERFV *GEO 1:2
** kf
wi
setn
2:4
1.
1
** Well geometries are input for well 3
** rad geofac wfrac skin
*GEOMETRY *K 0.5 0.2488 1.0 0.0
** The well completion is defined for well 3
*PERFV *GEO 3
** kf
wi
setn
2:4
1.
1
The variable rad is required if it has not been entered elsewhere in the data set.
If *GEOMETRY is used, then rad must be equal to the well radius entered with *REFINE
*HYBRID. Since the well is assumed to be in the center of the grid block geofac is assumed
to be 0.50. Hybrid grid is used to model, especially in producers, more accurately, near-well
effects, such as breakthrough.
User's Guide IMEX
Location of Well Completions (Conditional)
*PERF
PURPOSE:
*PERF specifies the location of the well completion grid blocks.

FORMAT:
*PERF *WI wn
{locat.}
:
{well index}
:
({rel. perm.})
:
(*KRPERF
{locat.}
:
{special rel perm dat}

:)
({status})
:
({connection})
:
-or*PERF *GEO (turbulent_skin) (*QUAD|*PSEUDOP) wn

{locat.}
:
{well index}
:
({rel. perm.})
:
(*KRPERF
{locat.}
:

:)
({status})
:
({connection})
:
({status})
:
({connection})
:
-or*PERF *KH (turbulent_skin) (*QUAD|*PSEUDOP) wn

{locat.}
:
{well index}
:
({rel. perm.})
:
(*KRPERF
{locat.}
:

:)
-or*PERF *GEOA (turbulent_skin) (*QUAD|*PSEUDOP) wn

{locat.}
:
{well index}
:
(*KRPERF
{locat.}
:

:)
User's Guide IMEX
({rel. perm.})
:
({status})
:
({connection})
:
-or*PERF *KHA (turbulent_skin) (*QUAD|*PSEUDOP) wn

{locat.}
:
{well index}
:
({rel. perm.})
:
(*KRPERF
{locat.}
:

:)
({status})
:
({connection})
:
DEFINITIONS:
(turbulent_skin)
(*TURB) | (*TURB_GRM) | (*TURB_FG1) | (*TURB_FG2)
{location}
{well index}
wi | ff (di) | kh (di)
{relative permeability}
(setn) (swcon) (sgcrit) (sorw) (sorg)
{special relative permeability data}
Special relative permeability table set number and end point data defined in a
separate manual page for the subkeyword *KRPERF.
{status}
(*OPEN) | (*AUTO) | (*CLOSED)
{connection}
(*FLOW-TO ily) | (*FLOW-FROM ily) (*REFLAYER)
*WI
Keyword indicating that well indices of layers are input directly for each well
layer. If none of *WI, *GEO, *GEOA, *KH or *KHA are present then this
is the default.
*GEO
Keyword indicating that well indices of layers are to be calculated from the
geometric information on the last *GEOMETRY keyword and from the
dimensions and permeability of the grid block in which the completion occurs.
User's Guide IMEX
*KH
geometric information on the last *GEOMETRY keyword and from the KH
specified for each well layer.
*GEOA
*GEOA uses the anisotropic version of Peacemans well model to calculate
effective well radius. Geofac from the *GEOMETRY keyword is ignored as
geofac for Peacemans model is 0.28 and independent of aspect ratio and
permeability anisotropy (see *GEOMETRY keyword for reference).
*KHA
specified for each well layer. *KHA uses the anisotropic version of
Peacemans well model to calculate effective well radius. Geofac from the
*GEOMETRY keyword is ignored as geofac for Peacemans model is 0.28
and independent of aspect ratio and permeability anisotropy (see
*GEOMETRY keyword for reference).
*TURB | *TURB_GRM | *TURB_FG1 | *TURB_FG2
Keywords indicating that turbulent skin effects are to be considered due to
high gas velocity near the wellbore which results in a deviation from Darcy's
law. The mechanical skin factor will be modified based on D factors.
Keyword *TURB indicates the D factor is given by user input (di of well
index). If *TURB_GRM, *TURB_FG1 or *TURB_FG2 is used, the
simulator calculates D factors based on the correlation of Geertsma (1974) or
on the two correlations of Fredrick and Graves (1994) respectively. See
keyword *NONDARCY for a description of these correlations.
If none of the *TURB, *TURB_GRM, *TURB_FG1 or *TURB_FG2
keywords are present, turbulent effects will be ignored. These subkeywords
can only be used with the *GEO, *GEOA, *KH and *KHA subkeywords.
*TURB
Keyword indicating that turbulent skin effects are to be considered due to
law. The mechanical skin factor will be modified based on input D factors.
If *TURB is not present, turbulent effects will be ignored. This subkeyword
*QUAD
Keyword indicating that a quadratic inflow relationship is to be used for
calculating the gas inflow/outflow from a well. The density of the gas near
the wellbore is assumed to be proportional to its pressure yielding:
User's Guide IMEX
q = wi * (Phase mobility) * (Pblock**2 - Pwell**2)/(2*Pblock)

as the inflow relationship. If *QUAD is not present, the standard
formulation is assumed. This subkeyword can only be used with the *GEO
(*GEOA) and *KH (*KHA) subkeywords.
*PSEUDOP
Subkeyword indicating that the variation in both gas density and gas
viscosity with pressure is to be taken into account in computing the gas flow
rate into or out of the wellbore. The expression used for the gas flow rate
when *PSEUDOP is in force is
Eg
q = wi * K rg * gstc *

* (Pblock Pwell) ,
ave
with the average value for the quotient of gas expansion factor and gas
viscosity computed as
Eg
= 0.5 * g

ave
Eg
block
well
with the gas expansion factor and gas viscosity evaluated directly from input
PVT table at both pressures. This subkeyword can be used only with the
*GEO, *GEOA, *KH, and *KHA perforation options.
*KRPERF
Optional subkeyword used to enter special relative permeability table set
number and end point data. Please see separate *KRPERF manual page for
details of this subkeyword.
wn
Well sequence number.
if
An integer specifying the I direction grid block index of the fundamental grid,
or a range of I direction grid blocks that the well is completed through. (See
explanation).
jf
An integer specifying the J direction grid block index of the fundamental
grid, or a range of J direction grid blocks that the well is completed through.
(See explanation).
kf
An integer specifying the K direction grid block index of the fundamental
grid, or a range of K direction grid blocks that the well is completed through
(See explanation).
User's Guide IMEX
ir1 jr1 kr1

Integers specifying the I, J, K direction grid block index of the first level of
refined grid that the well is completed through.
irn jrn krn
Integers specifying the I, J, K direction grid block index of the n'th level of
refined grid that the well is completed through. When using a hybrid grid the
last level refers to the R, Theta, and Z location of the hybrid radial grid.
wi
Well index for the layer.
Production for the layer for each phase, and injection for MOBILITY
WEIGHTED injectors is found from:
q = wi * (Phase mobility) * (Pblock - Pwell)
where the units for wi are (md*m | md*ft | md*cm).
If the well is a MOBILITY UNWEIGHTED INJECTOR, then the equation is:
q = wi * (Pblock - Pwell)
with units for wi of (m3/kPa-day | STB/psi-day or SCF/psi-day | cm3/kPamin | m3/kg/cm2-day).
ff
The well index calculated from geometric factors on the *GEOMETRY
keyword and the grid block dimensions and permeability is multiplied by this
factor. It can be used to account for partial completions through a grid block.
kh
KH for the layer.
q = 2pi * kh * wfrac / (ln(re/rad)+skin) * (Phase mobility) * (Pblock - Pwell)
re = geofac*sqrt((areap)/(3.14159*wfrac)) when *GEO or *KH used, see
*GEOMETRY keyword for formula when *GEOA or *KHA used
rad = wellbore radius
areap = area perpendicular to the wellbore (DX*DY for a vertical well)
where the units for kh are (md*m | md*ft | md*cm).
di
D factor for the layer, determined from a 2 cycle build-up test. The specified or
calculated well productivity will be adjusted by a turbulent skin factor which is
given by Di*|qi|. The D factor is used to modify the well index as follows:
q = wi' * (Phase mobility) * (Pblock - Pwell)
User's Guide IMEX
wi' = 2pi * kh * wfrac / (ln(re/rad) + skin + D*|qi|) * (Phase mobility)

The units for the D factor are (days/ft3 | days/m3 | mins/cm3)
setn
Relative permeability set to use when calculating the phase mobility in the
well equation. If not present, the relative permeability for the grid block will
be used. See *KRPERF keyword
This option can be used for partial completions where relative permeability
endpoints may be different. For example, if a producer is completed only in
the top quarter of the grid block, then the water will have to cone up to the
well (let's say Sw = 0.67) before water becomes mobile in the well.
swcon
Connate water saturation of the well layer to scale the relative permeability
table given by 'setn' swcon corresponds to the first table entry in the *SWT
table. If swcon is not present then the swcon of the relative permeability
table number 'setn' will be used. See *KRPERF keyword.
sgcrit
Critical gas saturation of the well layer to scale the relative permeability table
given by 'setn' corresponds to the table entry in the *SGT table where Krg is
zero. If sgcrit is not present then the sgcrit of the relative permeability table
number 'setn' will be used. See *KRPERF keyword.
sorw
Residual oil saturation of the well layer to scale the relative permeability
table given by 'setn' for the water-oil relative permeability sorw corresponds
to the table entry in the *SWT table where Krow is zero. If sorw is not
present then the sorw of the relative permeability table number 'setn' will be
used. See *KRPERF keyword.
sorg
table given by 'setn' for the gas-oil relative permeability sorg corresponds to
the table entry in the *SGT or *SLT table where Krog is zero. If sorg is not
present then the sorg of the relative permeability table number 'setn' will be
*OPEN
This subkeyword specifies that the well layer is open (perforated). This is
the default.
User's Guide IMEX
*AUTO
This subkeyword specifies that the well layer is currently plugged but is a
candidate for automatic recompletion. This automatic recompletion is
specified by using the *RECOMPLETE action of the *MONITOR keyword.
*CLOSED
This subkeyword specifies that the well layer is geometrical node point for
the purpose of defining the well trajectory. The layer will remain plugged
unless overridden by a new *PERF card selecting the layer status at that
location as *OPEN or *AUTO.
*FLOW-TO
This keyword is used to establish a "child to parent" relationship between the
current layer and a layer ALREADY specified under the current *PERF card
for a producer. For the first layer specified under the *PERF the parent is the
surface. This default is applied if this keyword is missing for the first layer.
For subsequent layers the default is the previous layer specified in the
sequence. The index used to reference a given layer for the "child to parent"
relationship is simply the count of the number of well layers entered to that
point that the parent layer is specified under the current *PERF card. The
range specification operator : can still appear, but only once per layer card.
These points are best illustrated with an example of a dual lateral well with
legs or "branches" parallel to the i and j axis:
*PERF
** i
1
1
1
1
2
3
*GEO 1
j
k
1 1:3
1
4
2
4
3
4
1
4
1
4
ff
status
1.0
1.0 *CLOSED
1.0
1.0
1.0
1.0
connection
FLOW-TO 'SURFACE'
FLOW-TO
3
FLOW-TO
4
FLOW-TO
5
FLOW-TO
4
FLOW-TO
7
**
**
**
**
**
**
1-3
4
5
6
7
8
This corresponds to the following geometry, assuming *KDIR *DOWN

1
j
6
2
k
3
4
5
7
The first line under the *PERF specifies layers 1-3. If the *FLOW-TO
keyword was not specified for this line then *FLOW-TO 'SURFACE' would
have been defaulted. The range in k direction is interpreted as well layer 1
connected to 2. Well layer 2 connected to 3. Layer flow is connected to
layer 3. In general if the *FLOW-TO keyword is missing for layer N+1, then
User's Guide IMEX
*FLOW-TO N is assumed. Therefore the *FLOW-TO keyword for layer 4 is

not strictly required. Note also that layer 4 is assigned the status of
*CLOSED. This node is defined simply for geometrical convenience, there
can be no flow from it. In this case by default the reference layer for BHP
calculation is the first layer. However any of the layers 1-8 could have been
designated as reference layer with the keyword *REFLAYER.
ily
The index of the parent layer used to establish a "child to parent"
relationship. The parent layer must already have been defined under the
current *PERF card to be referenced.
*FLOW-FROM
This keyword is used to establish a "child to parent" relationship between the
current layer and a layer ALREADY specified under the current *PERF card for
an injector. Please see documentation under the *FLOW-TO keyword above.
*REFLAYER
By default the first layer specified under the current *PERF card is used to
measure the wellbore flowing pressure (pwf). The user may select a different
layer for this purpose by the use of this keyword. This keyword can only
appear once per *PERF card.
DEFAULTS:

CONDITIONS:
follow *GEOMETRY.
If the *GEO or *GEOA keyword is present, then *PERF must come after at least one
*GEOMETRY. One of *PERF, *PERFV, or *PERFVHY is required.
If *QUAD or *PSEUDOP are not specified, then the drawdown pressure is calculated by the
standard formulation. If *TURB is used without the D factor being specified, then turbulent
skin effects are neglected. The keywords *TURB, *TURB_GRM, *TURB_FG1,
*TURB_FG2, *QUAD and *PSEUDOP cannot be used with *WI. These keywords must
immediately follow the *GEO, *GEOA, *KH or *KHA keywords and precede the well list.
If using *REFINE *HYBRID use *PERF or *PERFVHY.
EXPLANATION:
This keyword specifies the grid blocks in which a well is completed.

*PERF is used for horizontal or deviated wells, where the completions are not in a single
vertical column of grid blocks. (For vertical wells use the *PERFV keyword).
If the perforation location integers identify an inactive grid block (for example, a null or
pinched-out block), the user can choose one of two possible actions:
User's Guide IMEX
1. The perforation can be completed with a status of CLOSED, in which case the
perforation will be included for the purposes of head calculation but no fluid will
flow in the completion; or
2. The completion can be rejected by the simulator with a message identifying the
suspect perforation so that data may be modified.
Action (1) is the default. Action (2) can be specified with the *NULL-PERF keyword.
...
**
rad
geofac
*GEOMETRY *K 0.375 0.2488
wfrac
1.0
skin
0.0
*PERF *GEO 1
** if jf
kf ff (setn)
12
6 2:4 1.
1
13
6 5
.5
2
*PERF *GEO *TURB *QUAD 2
** if jf
kf
ff
di
12
6
2:4 1.
0.00002
13
6
5
.5
0.00004
(setn)
1
2
*PERF *GEO *TURB *PSEUDOP 3

** if jf
kf
ff
di
12
6
2:4 1.
0.00002
13
6
5
.5
0.00004
(setn)
1
2
*PERF *KH 1
** if jf
kf
12
6
2:4
13
6
5
kh
(setn)
100.
1
500.
2
*PERF *KH *TURB *QUAD 2

** if jf
kf
ff
di
12
6
2:4 1.
0.00002
13
6
5
.5 0.00004
*PERF 3
** if jf
16
8
17
8
18:20
8
kf
4
5
6
wi
1.56
2.34
12.4
(setn)
1
2
3
*PERF *KH *TURB *PSEUDOP 4

** if jf
kf
ff
di
12
6
2:4 1.
0.00002
13
6
5
.5 0.00004
User's Guide IMEX
(setn)
1
2
(setn)
1
2
Location of Vertical Well Completions (Conditional)
*PERFV
PURPOSE:
*PERFV specifies the location of the well completion grid blocks for vertical wells. The
*VERT subkeyword must have been used with the *WELL keyword used to define the well.
FORMAT:
*PERFV *WI wnums

kf
:
wi
:
(*KRPERF
kf
:
(setn)
:
(swcon)
:
(sgcrit)
:
(sorw)
:
(sorg)
:
(*OPEN)
(*AUTO)

:)
-or*PERFV *GEO (turbulent_skin) (*QUAD|*PSEUDOP) wnums

kf
:
ff
:
(*KRPERF
kf
:
(di)
:
(setn)
:
(swcon)
:
(sgcrit)
:
(sorw)
:
(sorg)
(*OPEN)
(*AUTO)

:)
-or*PERFV *KH (turbulent_skin) (*QUAD|*PSEUDOP) wnums

kf
:
ff
:
(di)
:
(setn)
:
(swcon)
:
(sgcrit)
:
(sorw)
:
(sorg)
(*OPEN)
(*AUTO)
-or*PERFV *GEOA (turbulent_skin) (*QUAD|*PSEUDOP) wnums

kf
:
ff
:
(*KRPERF
kf
:
(di)
:
(setn)
:
(swcon)
:
(sgcrit)
:
(sorw)
:
(sorg)
(*OPEN)
(*AUTO)

:)
User's Guide IMEX
-or*PERFV *KHA (turbulent_skin) (*QUAD|*PSEUDOP) wnums

kf
:
ff
:
(di)
:
(*KRPERF
kf
:
(setn)
:
(swcon)
:
(sgcrit)
:
(sorw)
:
(sorg)
(*OPEN)
(*AUTO)

:)
DEFINITIONS:
(turbulent_skin)
*WI
layer. If none of *WI, *GEO *GEOA, *KH or *KHA are present then this is
the default.
*GEO
*KH
*GEOA
*KHA
User's Guide IMEX

*TURB
can only be used with the *GEO *GEOA *KH and *KHA subkeywords.
*QUAD
q = wi * (Phase mobility) * (Pblock**2 - Pwell**2)/(2*Pblock)
as the inflow relationship. If *QUAD is not present, the standard formulation
is assumed. This subkeyword can only be used with the *GEO, *GEOA, *KH
and *KHA subkeywords.
*PSEUDOP
Eg

* (Pblock Pwell) ,
ave
Eg
= 0.5 * g

ave
Eg
block
well
User's Guide IMEX
*KRPERF
wnums
Well sequence number, or a number and/or range of well sequence numbers.
kf
An integer specifying the K direction grid block index of the fundamental
grid, or a range of K direction grid blocks that the well(s) is(are) completed
through.
wi
If the well is a MOBILITY UNWEIGHTED INJECTOR, then the equation
is:
with units for wi of (m3/kPa-day | STB/psi-day or SCF/psi-day | cm3/kPaday | m3/kg/cm2-day).
ff
kh
KH for the layer.
User's Guide IMEX

di
D factor for the layer, determined from a 2 cycle build-up test. The specified
or calculated well productivity will be adjusted by a turbulent skin factor which
is given by Di*|qi|. The D factor is used to modify the well index as follows:
where:
setn
be used. See *KRPERF keyword.
swcon
table number 'setn' will be used. See *KRPERF keyword.
sgcrit
User's Guide IMEX
sorw
sorg
*OPEN
the default.
*AUTO
DEFAULTS:

CONDITIONS:
follow *WELL.
If the *GEO or *GEOA keyword is present, then *PERFV must come after at least one
skin effects are neglected. Keywords of *TURB, *TURB_GRM, *TURB_FG1,
immediately follow the *GEO *GEOA, *KH or *KHA keywords and precede the well list.
For a given well, the first layer defined on the most recent *PERFV or *PERFVHY line is the
layer at which the well's bottom hole pressure is calculated, and the layer from which the pressure
drop to the surface is calculated when well head pressures are evaluated. The same holds for the
most recent *PERF line if the *REFLAYER subkeyword of *PERF was not used.
User's Guide IMEX
EXPLANATION:
This keyword specifies the grid blocks in which a vertical well is completed. If several vertical
wells are all completed in the same layers, they may all be defined with one *PERFV keyword.
...
**
rad
geofac wfrac
*GEOMETRY *K 0.375 0.2488 1.0
*PERFV *GEO 2:3
** kf
ff (setn)
2:4 1.
1
5
.5
2
*PERFV *KH *TURB *QUAD 1
** kf
kh
di
(setn)
2:4 100. 0.0002
1
5
500. 0.0004
2
*PERFV 1 4
** kf
ff
(setn)
2:4 1.56
1
5
1.1
2
*PERFV *KH *TURB *PSEUDOP 6
** kf
kh
di
(setn)
2:4 100. 0.0002
1
5
500. 0.0004
2
skin
0.0
User's Guide IMEX
Location of Vertical Well Completions for Hybrid Refined

Grids (Conditional)
*PERFVHY
PURPOSE:
*PERFVHY specifies the location of the well completion grid blocks for vertical wells
completed in hybrid refined grid blocks. The *VERT subkeyword must have been used with
the *WELL keyword used to define the well.
FORMAT:
*PERFVHY *WI wnums

kf
:
ff
:
(wi)
:
(*KRPERF
kf
:
(setn) (swcon)
:
:
(sgcrit)
:
(sorw)
:
(sorg)
:
(*OPEN)
(*AUTO)

:)
-or*PERFVHY *GEO (turbulent_skin) (*QUAD|*PSEUDOP) wnums

kf
:
kr
:
(*KRPERF
kf
:
ff
:
(di)
:
(setn) (swcon) (sgcrit) (sorw) (sorg) (*OPEN)

:
:
:
:
:
(*AUTO)

:)
-or*PERFVHY *GEOA (turbulent_skin) (*QUAD|*PSEUDOP) wnums

kf
:
kr
:
(*KRPERF
kf
:
ff
:
(di)
:

:
:
:
:
:
(*AUTO)

:)
-or*PERFVHY *KHA (turbulent_skin) (*QUAD|*PSEUDOP) wnums
User's Guide IMEX
kf
:
kr
:
(*KRPERF
kf
:
ff
:
(di)
:

:
:
:
:
:
(*AUTO)

:)
DEFINITIONS:
(turbulent_skin)
*WI
layer. If none of *WI, *GEO, *GEOA, *KH or *KHA are present then this
is the default.
*GEO
*KH
*GEOA
*KHA
User's Guide IMEX

*TURB
*QUAD
q = wi * (Phase mobility) *
(Pblock**2 - Pwell**2)/(2*Pblock)
as the inflow relationship. If *QUAD is not present, the standard

formulation is assumed. This subkeyword can only be used with the *GEO,
*GEOA, *KH and *KHA subkeywords.
*PSEUDOP
Eg

* (Pblock Pwell) ,
ave
Eg
User's Guide IMEX
= 0.5 * g

ave
Eg
block
well
*KRPERF
wnums
Well sequence number, or a number and/or range of well sequence numbers.
kf
An integer specifying the K direction grid block index of the fundamental grid,
or a range of K direction grid blocks that the well(s) is(are) completed through.
kr
An integer specifying the K direction grid block index of the hybrid refined
grid, or a range of K direction refined grid blocks that the wells are completed
through.
wi
If the well is a MOBILITY UNWEIGHTED INJECTOR, then the equation is:
with units for wi of (m3/kPa-day | STB/psi-day or SCF/psi-day | cm3/kPaday | m3/kg/cm2-day).
ff
kh
KH for the layer.
User's Guide IMEX
re = geofac*sqrt((areap)/(3.14159*wfrac))
when *GEO or *KH used, see *GEOMETRY keyword for formula when
*GEOA or *KHA used
di
D factor for the layer, determined from a 2 cycle build-up test. The specified
or calculated well productivity will be adjusted by a turbulent skin factor
which is given by Di*|qi|. The D factor is used to modify the well index as
follows:
where:
setn
be used. See *KRPERF keyword.
swcon
table number 'setn' will be used. See *KRPERF keyword
User's Guide IMEX
sgcrit
sorw
sorg
*OPEN
the default.
*AUTO
DEFAULTS:

CONDITIONS:
follow *WELL.
If the *GEO or *GEOA keyword is present, then *PERFVHY must come after at least one
skin effects are neglected. Keywords of *TURB, *TURB_GRM, *TURB_FG1,
immediately follow the *GEO, *GEOA, *KH or *KHA keywords and precede the well list.
User's Guide IMEX
For a given well, the first layer defined on the most recent *PERFV or *PERFVHY line is the
layer at which the well's bottom hole pressure is calculated, and the layer from which the
pressure drop to the surface is calculated when well head pressures are evaluated. The same
holds for the most recent *PERF line if the *REFLAYER subkeyword of *PERF was not used.
EXPLANATION:
This keyword specifies the hybrid refined grid blocks in which a vertical well is completed. If
several vertical wells are all completed in the same layers, they may all be defined with one
*PERFVHY keyword.
*WELL 1 'Producer 1' *VERT
...
**
rad
geofac
*GEOMETRY *K 0.375 0.2488
12 16
10 5
21 3
17 12
wfrac
1.0
skin
0.0
*PERFVHY *GEO 2:3

** kf kr
ff (setn)
2:4 1:3 1.
1
5
2
0.5
2
*PERFVHY *KH
** kf
kr
2:4 1:3
5
2
*TURB *QUAD 1
kh
di
(setn)
100. 0.0002
1
500. 0.0004
2
*PERFVHY 1 4
** kf
kr
ff
(setn)
2:4 1:3 1.56
1
5
2
1.1
2
*PERFVHY *KH
** kf
kr
2:4 1:3
5
2
User's Guide IMEX
*TURB *PSEUDOP 1
kh
di
(setn)
100. 0.0002
1
500. 0.0004
2
If *GEOMETRY is used, then rad must be equal to the well radius entered with *REFINE
*HYBRID. Since the well is assumed to be in the center of the grid block geofac is assumed
to be 0.50. Hybrid grid is used to model, especially in producers, more accurately, near-well
effects, such as breakthrough.
User's Guide IMEX
Geometric Data for Deviated Well Completions (Conditional)

*LAYERXYZ
PURPOSE:
*LAYERXYZ allows the user to supply geometric information specifying deviated perforations
perforations in which the wellbore direction is not parallel to one of the local coordinate axes.
FORMAT:
*LAYERXYZ wname
{locat.}
:
{deviated layer information}

:
DEFINITIONS:
wname
Single well name within single quotes specifying the well to which the
following deviated layer specifications apply. No wild-carding is supported
in this situation.
{location}
These triples of integers specify the layers of well wn to be treated as
deviated layers. It is valid to name some of well wns layers and not others
in the *LAYERXYZ lines, but layers not mentioned do not acquire the
deviated status and will be treated as *I, *J, or *K perforations. Any layer
named under *LAYERXYZ must already have been defined for the well
using a *PERF or *PERFV statement.
{deviated layer information}
(x1 y1 z1 x2 y2 z2 plength | *UNDEVIATED)
if
jf
kf
ir1 jr1 kr1
User's Guide IMEX
irn jrn krn

x1 y1 z1
Coordinates of the entry point of the wellbore into the perforated grid
block. (m | ft). See explanation.
x2 y2 z2
Coordinates of the exit point of the wellbore from the perforated gridblock.
(m | ft). See explanation.
plength
Perforated length of the deviated wellbore within the named grid block (m | ft).
See explanation.
*UNDEVIATED
Subkeyword indicating that the layer is henceforth to be treated as an
undeviated layer with the direction (*I,*J, or *K) which the layer had when
initially perforated.
DEFAULTS:
Layers named under *LAYERXYZ with geometric information (as opposed to the
*UNDEVIATED subkeyword) are flagged as deviated within the simulator; undeviated is the
default status assigned at the beginning of the run. The undeviated status can be reimposed
through the *UNDEVIATED subkeyword.
CONDITIONS:
The named layers must all have been previously created for well wn with *PERF or *PERFV
lines using the *GEO (or *GEOA) option for computation of well indices. If another well
index calculation was specified, the layer is rejected as invalid for deviated perforation. If
geometric information is supplied, all seven of the real numbers x1, y1, z1, x2, y2, z2,
plength must be specified or an error is generated. Not all of a wells layers need to be
named under *LAYERXYZ; those omitted are treated as undeviated according to the original
*PERF specification.
EXPLANATION:
This keyword specifies geometric information which allow the well indices to be calculated for
perforations in which the wellbore direction does not parallel one of the gridblock coordinate
axes. The point coordinates x1,y1,z1 and x2,y2,z2 must be expressed in the Cartesian
coordinate system underlying the simulation grid. The entry point and exit point specify the
direction of the wellbore; the perforation length gives the actual perforation length through the
named gridblock. It is valid for plength to exceed the distance between (x1,y1,z1) and
(x2,y2,z2) in order to allow more freedom in matching well indices. However the distance
between (x1,y1,z1) and (x2,y2,z2) must be positive; if (x1,y1,z1) and (x2,y2,z2) are the same
point an error is generated. The identities of entry and exit points can be exchanged with no
difference in the well index calculation. The ff factor entered under the *PERF or *PERFV line
User's Guide IMEX
when the layer was created still apply to the deviated layer. If the user intends that plength
alone control the well index calculation, ff should be entered as 1.0 on the *PERF line for the
layer.
The deviated well index is calculated as
WI = 2*wfrac*K*plength*ff/(ln(re/rw) + ss)
The well angular fraction wfrac, the completion factor ff, the well radius rw, and the skin
term ss are entered on the relevant *GEOMETRY or *PERF.. lines. plength is read directly
in the *LAYERXYZ data.
The drainage radius re is computed from the information entered under *LAYERXYZ as
follows. When the wellbore is parallel to the D axis (D is I, J, or K), re(D) is calculated as
Re(D) = geofac*sqrt(V/(*xh(D)*wfrac))
(unless the *GEOA option is used, see *GEOMETRY for details of how effective radius is
calculated for direction D (D is I, J, or K).)
where V is the bulk volume of the perforated grid block, xh(D) is the grid block thickness in
the direction D, and geofac is the geometric factor entered with the *GEOMETRY keyword.
Once re(D) has been calculated for the three directions I,J,K, an interpolation to the deviated
wellbore direction is done as follows. Let u be a unit vector in the wellbore direction (x2x1,y2-y1,z2-z1). It makes no difference whether u points with or against the fluid flow in the
wellbore. Let i, j, and k be unit vectors pointing in the local I, J, and K directions for the
block in which the layer is perforated. These are determined from data entered for the block
in the Reservoir Description section of the data set. Note that for corner point grids (see the
Reservoir Description section of this manual) these directions are not necessarily aligned with
the underlying Cartesian axes of the grid (those in which the coordinates x1, y1, etc. are
defined) nor are the vectors i, j, and k necessarily mutually orthogonal.
Define cos(thetaI) as the dot product ui, and let cos(thetaJ) and cos(thetaK) be similarly
defined. Define sin**2(thetaI) as 1 cos**2(thetaI) and similarly for J and K. Then the
interpolated value of re to apply in the direction of the wellbore is
re(u)
(re(I)*cos**2(thetaI)*sin**2(thetaJ)*sin**2(thetaK) +
re(J)*cos**2(thetaJ)*sin**2(thetaK)*sin**2(thetaI) +
re(K)*cos**2(thetaK)*sin**2(thetaI)*sin**2(thetaJ))/S,
where S is the sum of the three trigonometric weighting factors in the numerator.
The completion planar averaged permeability K is computed similarly except that re(I), re(J),
and re(K) are replaced with
K(I)
K(J)
K(K)
=
=
=
User's Guide IMEX
sqrt(Ky*Kz)
sqrt(Kz*Kx)
sqrt(Kx*Ky)
EXAMPLES:
Example 1:
*LAYERXYZ WELL-NNE17
65 23 5
** x1
y1
z1
2287.49 1457.64 3949.09
x2
y2
z2
plength
2284.34 1460.23 3944.28 2.67
A completion for well WELL-NNE17 in block 65 23 5 must already have been created with
a *PERF.. line for the above to be valid.
Example 2:
*LAYERXYZ WELL-MULTI-REF
16 48 11 / 1 1 2 / 2 2 1
** x1
y1
z1
x2
102.11 493.74 2285.53
102.48
y2
z2
plength
494.87 2284.13
2.67
In this example, the already defined completion of well WELL-MULTI-REF in the

multiply-refined grid block 16 48 11 / 1 1 2 / 2 2 1 is flagged as deviated and the seven values
above are assigned to x1, y1, z1, x2, y2, z2, and plength.
User's Guide IMEX
Simplified Geometric Data for Deviated Well Completions

(Conditional)
*LAYERIJK
PURPOSE:
*LAYERIJK allows the user to supply a layer direction for each layer. Perforations are parallel
to one of the local coordinate axes but can vary from layer to layer. The directions defined on
the *LAYERIJK keyword override the well direction specified on the *GEOMETRY keyword.
FORMAT:
*LAYERIJK wname
{locat.}
:
{layer direction}
:
DEFINITIONS:
wname
in this situation.
{location}
These triples of integers specify the layers of well wn to be treated as
deviated layers. It is valid to name some of well wns layers and not others
in the *LAYERIJK lines. Layers not mentioned do not acquire the deviated
status; layer direction is then defined by the currently active GEOMETRY
keyword. Any layer named under *LAYERIJK must already have been
defined for the well using a *PERF or *PERFV statement.
{layer direction}
(*I | *J | *K | *UNDEVIATED)
if
jf
kf
User's Guide IMEX
ir1 jr1 kr1

irn jrn krn
*I
Signifies the layer is perforated in the local I direction.
*J
Signifies the layer is perforated in the local J direction.
*K
Signifies the layer is perforated in the local K direction.
*UNDEVIATED
undeviated layer with the direction (*I,*J, or *K) which the layer had when
initially perforated (defined by active *GEOMETRY keyword).
DEFAULTS:
Layers named under *LAYERIJK with geometric information *I, *J or *K (as opposed to the
*UNDEVIATED subkeyword) are flagged as deviated within the simulator; undeviated is the
default status assigned at the beginning of the run. The undeviated status can be reimposed
through the *UNDEVIATED subkeyword.
CONDITIONS:
The named layers must all have been previously created for well wn with *PERF or *PERFV
lines using the *GEO (or *GEOA) option for computation of well indices. If another well
index calculation was specified (example *KH), the layer is rejected as invalid for deviated
perforation. Not all of a wells layers need to be named under *LAYERIJK; those omitted
are treated as undeviated according to the original *PERF specification.
EXPLANATION:
This keyword specifies geometric information that allows the well indices to be calculated for
perforations in which the wellbore direction is parallel to one of the gridblock coordinate axes
but varies by layer. The ff factor entered under the *PERF or *PERFV line when the layer
was created still apply to the deviated layer.
The deviated well index is calculated in a manner identical to the LAYERXYZ keyword with
the following additional assumptions made. It is assumed a perforation enters a block at the
center (barycenter) of one face perpendicular to the layer direction (*I, *J, or *K) and exits at
the center (barycenter) of the other face. The length of the perforation is the length of the line
connecting one face center to the other.
User's Guide IMEX
For example, a *LAYERIJK *I perforation would follow the path defined by connecting the
center of the lower I face (the face connecting block I with block I-1) to the center of the upper I
face (the face connecting the block I with the block I+1). The length of the connection is the
distance between the centers of the lower I face and upper I face. Using the notation defined in
the *LAYERXYZ keyword explanation, we also assume that re(u) equals re(I) and K(u) = K(I).
A *LAYERIJK *J or *LAYERIJK *K perforation would have its perforation length,
effective radius, and well permeability calculated in an analogous fashion (references to I
would be replaced by J or K in the above paragraph).
EXAMPLES:
Example 1:
*LAYERIJK WELL-NNE17
65 23 5 *I
A completion for well WELL-NNE17 in block 65 23 5 must already have been created with
a *PERF.. line for the above to be valid, the well perforation is parallel to the I direction
Example 2:
*LAYERIJK WELL-MULTI-REF
16 48 11 / 1 1 2 / 2 2 1 *J
In this example, the already defined completion of well WELL-MULTI-REF in the

multiply-refined grid block 16 48 11 / 1 1 2 / 2 2 1 is flagged as deviated, the well perforation
is parallel to the J direction.
User's Guide IMEX
Data for Hydraulic Fracture Proxy (HFP) Model (Conditional)

*LAYERHFP
PURPOSE:
productivity indices which act as proxies for vertical fractures (assuming pseudo radial flow) on
a layer by layer basis.
FORMAT:
*LAYERHFP wname
{locat.}
:
{HFP Layer Information}

:
DEFINITIONS:
wname
in this situation.
{location}
These triples of integers specify the layers of well wname to be treated as
fractured layers. It is valid to name some of wells layers and not others in
the *LAYERHFP lines, but layers not mentioned are not treated as proxies
for vertical fractures. Any layer named under *LAYERHFP must already
have been defined for the well using a *PERF or *PERFV statement.
{HFP Layer Information}
(hf_hlen hf_wid hf_perm hf_bet_a hf_bet_b hf_ccf | *DEFAULT)
if
jf
kf
ir1 jr1 kr1
User's Guide IMEX
irn jrn krn

hf_hlen
Fracture half length. (m | ft).
hf_wid
Fracture width (m | ft).
hf_perm
Fracture Permeability (mD).
hf_bet_a
Constant term in the non Darcy (Beta) factor correlation. See below.
hf_bet_b
Permeability term in the non Darcy (Beta) factor correlation. See below.
Hf_bet_a and hf_bet_b are used to determine the non Darcy term for the
fracture.
frac (p) = hf_bet_a x (Krp x hf_perm) hf_bet_b p=oil, water, gas, or solvent,
Hf_perm is the fracture permeability in mD and hf_bet_a must be entered so as
to yield frac (p) in units 1/ft. Krp is the relative permeability of the phase whose
frac (p) is being calculated. Regardless of the unit system employed in the
simulator, frac has units of 1/ft.
hf_ccf
Fracture conductivity correction factor for non Darcy flow. See explanation,
normally set to 0.50.
*DEFAULT
unfractured layer.
DEFAULTS:
Layers named under *LAYERHFP with geometric information and non Darcy information
(as opposed to the *DEFAULT subkeyword) are flagged as hydraulically fractured within the
simulator; *DEFAULT is the default (unfractured) status assigned at the beginning of the run.
Unfractured status can be reimposed through the *DEFAULT subkeyword. All 6 HFP
entries must be defined if the proxy model is used (hf_hlen, hf_wid, hf_perm, hf_bet_a,
hf_bet_b, hf_ccf).
User's Guide IMEX
CONDITIONS:
The named layers must all have been previously created for well wname with *PERF or
*PERFV lines using the *GEO *KH *GEOA or *KHA option for computation of well indices
(cannot be used with the *WI option). If geometric and non Darcy information is supplied, all
of the input (hf_hlen, hf_wid, hf_perm, hf_bet_a, hf_bet_b, and hf_ccf) must be specified or an
error is generated. Not all of a wells layers need to be named under *LAYERHFP; those
omitted are treated as unfractured according to the original *PERF specification.
EXPLANATION:
This keyword specifies input of geometric and non Darcy parameters which allows the well
productivity index (PI) to be used as a proxy for a vertical hydraulic fracture, assuming the flow
pattern is pseudo radial. The work is based on the work of Pratt, Cinco-lay and Samaniego.
Later Guppy and Gidley developed a similar non Darcy extension. This option applies these
concepts to dynamically calculate well productivities within a reservoir simulator.
Early work in well test analysis of fractured wells has shown that for the pseudo radial flow
period an infinite conductivity fractured well behaves like an unfractured well with an effective
wellbore radius of the fracture half length.
Low and finite conductivity wells can be represented through the use of an effective wellbore
radius as well. In this case, the effective radius is a function of dimensionlesss fracture
conductivity (KfBf)D and fracture length. This analysis applies to Darcy flow in fractures
only.
The extension which allows the application of these ideas to non Darcy flow is to calculate a
resistance factor due to non Darcy flow in the fracture which lowers the fracture conductivity.
The resistance factor, which is defined as 1/(1+Forchfr), is a function of dynamically changing
reservoir properties and flow in the fracture, all of which vary from timestep to timestep and so
must be reevaluated at every timestep.
For each phase (p), at each timestep, for each layer, we calculate the Fracture Forchheimer
Number (Forchfr) as:
Forch fr =
hf _ ccf x frac (p ) x Density(p ) x K rp x hf _ permx Q(p )

Vis cos ity(p ) x hf _ wid x (layer vertical height )
Where frac (p) = hf_bet_a x (Krp x hf_perm) hf_bet_b

If hf_ccf is 0.5, this assumes the non Darcy effects in each half of the fracture are based on of
the layers total flow. Q(p) is the production rate of phase p. Krp is the relative permeability of
phase p.
The Darcy dimensionless fracture conductivity is calculated as:
(K f B f )D
hf _ perm x hf _ wid
(Formation perm ) x (fracture 1 / 2 length )
The dimensionless conductivity is reduced by 1/(1+Forchfr) and this reduced dimensionless

conductivity is used to calculate (using the function graphed below) an effective well radius
from which a fracture proxy well PI is calculated.
hf_ccf is the fracture conductivity correction factor which is used to relate the flow of the layer
to the flow in each wing of the fracture. As the Forchheimer number in each wing of the well is
User's Guide IMEX
normally based on no more than of the well layer rate, we assume the hf_ccf should normally
be no greater than 0.5. This implies that the maximum Forchheimer number is based on no
more than the well layer rate.
Normally hf_ccf should be set to 0.5, hf_ccf values smaller or larger than 0.5 can be used in
order to match observed well pressures. For example; if a significant amount of production
enters the fracture right at the well, the Forchheimer number in the fracture would be less than
if hf_ccf was assumed to be 0.5. In this case lower values of hf_ccf might be used to account
for the production which effectively bypasses the fracture.
hf_ccf = 0.0 can be used to model Darcy flow in the fracture (completely ignoring non Darcy
effects), by making the Forchheimer number equal to 0.0. Darcy flow tends to estimate very
large effective well radii; the model will limit this to be no more than 90% of the well blocks
effective radius (Re). Using this model to model Darcy flow is only useful in illustrating the
large influence non Darcy effects have on fracture productivity.
Effective Wellbore Radius vs. Dimensionless Fracture Conductivity for a Vertical

Fracture
EXAMPLES:
Example 1:
*LAYERHFP WELL-NNE17
**
frac
frac
**
half
width
**
length
65 23 5 150.0
0.00833
65 23 6 130.0
0.00800
frac
perm
Beta
Beta
A-term B-term
conductivity
correction
factor
50000 27.3e9 -1.1045 0.5
45000 27.3e9 -1.1045 0.5
Completions for well WELL-NNE17 in blocks 65 23 5 and 65 23 6 must already have been
created with a *PERF line for the above to be valid.
User's Guide IMEX
Example 2:
*LAYERHFP WELL-SINGLE-REF
**
frac
frac
frac Beta
Beta
conductivity
**
half
width
perm A-term B-term correction
**
length
factor
16 48 11 / 2 2 1 80.0
0.00833 100000 27.3e9 -1.1045 0.5
In this example, the already defined completion of well WELL-SINGLE-REF in the refined
grid block 16 48 11 / 2 2 1 is flagged as being a proxy for a hydraulic fracture.
Example 3:
*LAYERHFP WELL-SINGLE-REF
** The layer is unfractured (may have previously
** been a proxy for a fracture)
16 48 11 / 2 2 1 *DEFAULT
In this example, the already defined completion of well WELL-SINGLE-REF in the refined
grid block 16 48 11 / 2 2 1 is flagged as not being a proxy for a hydraulic fracture.
In the examples above the Beta A-term (hf_bet_a) = 27.3e9 and the B-term (hf_bet_b) = 1.1045 were taken from an example in the textbook by Dake (Fundamentals of Reservoir
Engineering).
The correlation for developed by Cooke, C.E. Conductivity of Fracture Proppants in
Multiple Layers J.Pet.Tech, (Sept 1973) pp1101-1107 is relevant for fractured systems and
is presented below (frac in 1/ft, Kf in mD).
Proppant Sand Mesh Size
hf_bet_a
hf_bet_b
8-12
538.108E9
-1.24
10-20
850.525E9
-1.34
20-40
3411.752E9
-1.54
40-60
2143.503E9
-1.60
Useful conversion factors are:

1 atm-sec2/gm = 1.01325E6 cm-1, 1 ft = 30.48 cm,
1 mD = 0.9869233E-11 cm2
User's Guide IMEX
Special Relative Permeability Data for Well Completions

(Conditional)
*KRPERF
PURPOSE:
*KRPERF is an optional subkeyword of the *PERF, *PERFV, and *PERFVHY keywords. It

specifies the relative permeability table set numbers (set numbers) and end points for well
completions (layers). *KRPERF, if used, must immediately follow the sequence of layer
lines introduced by the last *PERF.. keyword and must precede the next keyword. Only
layers named in the immediately preceding layer list can be referred to using *KRPERF.
Some relative permeability information can be entered directly in the *PERF.. lines without
using *KRPERF; please see the manual entries for the *PERF.. keywords for details.
However there are other special relative permeability data which can be entered only through
the *KRPERF subkeyword. We recommend that users with new data sets enter all
completion end point and relative permeability set numbers on the *KRPERF card.
FORMAT:
*KRPERF
{locat.}
:

:
DEFINITIONS:
{location}
The same layer identification used in the preceding *PERF, *PERFV, or
*PERFVHY keyword. The format for the layer identification differs among
those keywords. Please see the appropriate manual page.
A set of one or more of the subkeywords *SETN, *SWCON, *SWCRIT,
*SORW, *SOIRW, *SGCON, *SGCRIT, *SORG and *SLCON, each
followed by a number (integer or real) giving the rel. perm. table set number
or end point value.
*SETN
Introduces the integer relative permeability table set number (set number) used
to calculate the mobility entering the well index expression. If not specified,
the relative permeability curve set number for the perforated grid block is used.
*SWCON
Introduces the real number specifying the connate water saturation for the
completion, which scales the *SWT relative permeability table. *SWCON is the
smallest water saturation possible the in *SWT table. One minus *SWCON is
the maximum oil saturation possible in the *SWT table. If the keyword is not
present see DEFAULTS below for a description of how the perforation value
of this endpoint is obtained.
User's Guide IMEX
*SWCRIT
Introduces the real number specifying the critical water saturation for the
completion, which scales the *SWT relative permeability curve. *SWCRIT
is the water saturation at which water first becomes mobile in the two phase
oil-water table. *SWCRIT should be equal to or greater than *SWCON. If
the keyword is not present see DEFAULTS below for a description of how
the perforation value of this endpoint is obtained.
*SOIRW
Introduces the real number specifying the irreducible oil saturation for the
completion, which scales the *SWT relative permeability curve. *SOIRW is
the smallest oil saturation in the two phase oil-water table, one minus
*SOIRW is the maximum water saturation in the table. If the keyword is not
*SORW
Introduces the real number specifying the residual oil saturation for the
completion, which scales the *SWT relative permeability curve. *SORW is the
oil saturation in the two phase oil-water table at which oil first becomes mobile.
*SORW should be equal to or greater than *SOIRW. If the keyword is not
*SORMAX
Introduces the real number specifying the maximum residual oil saturation
for the completion, which scales the Sormax value of the rock type of the
completion. It is only effective when the Krow hysteresis or trapped oil
hysteresis is modeled for the rock type of the completion. Sormax is the oil
saturation in the two phase oil-water system at which oil first becomes
immobile during the boundary/primary imbibition process (the imbibition
starting from oil saturation equals to 1-Swcon). If the keyword is not present
see DEFAULTS below for a description of how the perforation value of this
endpoint is obtained.
*SGCON
Introduces the real number specifying the connate gas saturation for the
completion, which scales the *SGT or *SLT relative permeability curve.
*SGCON is the smallest gas saturation in the two phase gas-liquid relative
permeability table. If the keyword is not present see DEFAULTS below for a
description of how the perforation value of this endpoint is obtained.
User's Guide IMEX
*SGCRIT
Introduces the real number specifying the critical gas saturation for the
*SGCRIT is the gas saturation in the two phase gas-liquid relative
permeability table at which gas first becomes mobile. *SGCRIT should be
equal to or greater than *SGCON. If the keyword is not present see
DEFAULTS below for a description of how the perforation value of this
endpoint is obtained.
*SLCON
Introduces the real number specifying the connate liquid saturation for the
*SLCON is the smallest liquid saturation entry in the two phase gas-liquid
relative permeability table. If the keyword is not present see DEFAULTS
below for a description of how the perforation value of this endpoint is
obtained.
*SORG
Introduces the real number specifying the residual oil saturation for the
*SORG is the oil saturation in the two phase gas-liquid relative permeability
table at which oil becomes mobile. The definition of *SORG changes
depending on whether the *NOSWC option is active. If *NOSWC is not
active, the residual liquid saturation equals *SORG plus *SWCON, if
*NOSWC is active, the residual liquid saturation equals *SORG. If the
keyword is not present see DEFAULTS below for a description of how the
perforation value of this endpoint is obtained.
DEFAULTS:
For the saturation endpoints which are not explicitly defined on the current *KRPERF
keyword, the following algorithm is used to obtain their values.
If SETN is read in on the *KRPERF keyword (or defined on the *PERF line) for the layer in
question, unassigned saturation endpoints are obtained from the original table value of the
rock type assigned in *SETN. Current grid block saturation endpoints are ignored.
If SETN has not been read in on the current KRPERF keyword (or defined on the *PERF
line) for the layer in question, unassigned saturation endpoints are obtained from the current
grid block saturation endpoints.
*SETN, *SWCON, *SGCRIT, *SORW, and *SORG can be set either directly in the layer
line after *PERF, *PERFV, or *PERFVHY, or after *KRPERF. If no relative permeability
data are entered in the *PERF.. layer line, then defaults are set (or reset in the case of
reperforation), and non-default values, if wanted, must be established or re-established with
*KRPERF. The *SWCRIT , *SOIRW, *SORMAX, *SGCON and *SLCON end points can
only be set with the *KRPERF subkeyword.
User's Guide IMEX
When values of *SETN, *SWCON, *SGCRIT, *SORG, and *SORG are read in on the
various *PERF cards, their values will be overridden by the values of these variables
encountered on the *KRPERF card.
Please see the descriptions above of *SETN, *SWCRIT, *SWCON, *SGCRIT, *SGCON,
*SORW, *SOIRW, *SORG, and *SLCON, for more detailed defaulting information, as well
as the manual pages for the *PERF.. keywords.
Please refer to the detailed discussion of relative permeability scaling given in the manuals
Rock-Fluid Data section under the array keywords *SWCON, *SWCRIT, *SORW,
*SOIRW, *SGCON, *SGCRIT, *SLCON, or *SORG.
When the backward compatibility option *SCALING-OLD is used for a relative permeability
set (see *RPT keyword). Only the variables *SETN, *SWCON, *SGCRIT, *SORG, and
*SORG may be entered on the *KRPERF card. The values of the new variables are ignored.
We recommend that users employing *SCALING-OLD do not use the *KRPERF option.
The following rules are used to resolve inconsistencies in the end point data.
1. Critical water saturation is less than connate water saturation. Critical water
saturation is reset to connate water saturation.
2. Irreducible oil saturation in the water-oil table is greater than residual oil. Residual
oil saturation is reset to irreducible oil saturation.
3. Irreducible oil saturation in the gas-liquid table (Soirg) is greater than residual oil.
Irreducible oil saturation is reset to residual oil saturation. Irreducible oil is
calculated from *SLCON and *SWCON. If *NOSWC is not active, Soirg equals
*SLCON minus *SWCON. If *NOSWC is active, Soirg equals *SLCON.
4. Critical gas saturation is less than connate gas saturation. Critical gas saturation is
reset to connate gas saturation.
5. If when *NOSWC is not active, connate liquid saturation in the gas-liquid table is
less than connate water saturation, connate liquid saturation is reset to connate
water saturation.
CONDITIONS:
This subkeyword must be located in the Well and Recurrent Data keyword group, after
*PERF, *PERFV, or *PERFVHY and before the next keyword.
All layers named in {location} after *KRPERF must have appeared in the immediately
preceding set of layer perforation lines.
The *KRPERF subkeyword is optional, and when it appears there is no requirement that all
perforated layers of the well must be named under *KRPERF; however, layers not named
under *KRPERF receive the table set numbers and end point values set by default or set
directly in the *PERF.. lines.
EXPLANATION:
This subkeyword provides an alternative to entering completion relative permeability table

set numbers and end points on the *PERF, *PERFV, or *PERFVHY lines. In addition this
subkeyword allows the user access to more end points.
User's Guide IMEX
Values for *SETN, *SWCON, *SORW, *SORG, and *SGCRIT may be entered either in the
*PERF.. line or using the *KRPERF keyword; *SWCRIT, *SOIRW, *SGCON and *SLCON
can be entered only by using the *KRPERF keyword.
This option can be used to represent effects such as partial completions where relative
permeability end points may be altered. For example, if a producer is completed only in the
top quarter of a grid block, then the water must cone up to the perforation (e.g. Sw = 0.67)
before water can flow into the wellbore. So a water relative permeability set with a critical
water saturation of 0.67 can be assigned to the partial completion using *SETN without
affecting reservoir flow.
Examples of Usage
Example 1:
...
**
rad
geofac
*GEOMETRY *K 0.375 0.2488
wfrac
1.0
skin
0.0
*PERF *GEO Producer 1

** if
jf kf
ff
setn swcon sgcrit sorw
12
6
2:4 0.873 7
0.20
0.15
0.17
*KRPERF
12
6 2:3 *SWCRIT 0.22 *SOIRW 0.14
sorg
0.15
In this example, the completions of well 1 in layers 2 and 3 have the completion critical water
saturation set to 0.22 and the completion irreducible oil saturation set to 0.14. Note that
*SWCRIT exceeds the connate water saturation of 0.20 set in the layer perforation line, and
that the value of the connate oil saturation (0.14) is less than the residual oil saturation on the
water-oil curve (0.17).
Example 2:
*PERFV *KH WELL*
** kf
ff
1:4
1.0
*KRPERF
1:3 *SETN 2
*SORW 0.12
*SOIRW 0.10
4 *SETN 3
*SWCRIT
*SORG
*SLCON
*SWCRIT
0.15
0.10
0.17
0.25
*SGCRIT 0.10
*SWCON 0.12
*SGCON 0.00
In this example, all wells with WELL as the first four characters of their names are
completed in layers 1 through 4 at the I and J blocks specified after *VERT in the defining
*WELL line. If any well in this set does not have I and J indices specified, the *PERFV line
is not valid. In the line following *PERFV, the defaults are set for all of the layer relative
permeability variables (since they are not referred to). The first line following *KRPERF
resets all nine special relative permeability variable values to the entered values for layers 1:3
of the listed wells; for layer 4 a table set number (3) is provided and all other quantities
except *SWCRIT default. In layer 4, *SWCRIT is being set high to reduce water production.
User's Guide IMEX
Group Production Constraints (Optional)
*GCONP
PURPOSE:
*GCONP is used to specify group production controls.

FORMAT:
*GCONP 'group_name_1' 'group_name_2' ... 'group_name_n'

(*MAX)
(*STO)
(*STG)
(*STW)
(*STS)
(*STL)
(*BHF)
value
(*MIN)
(*MNP)
(*CPP)
(*STO)
(*STG)
(*STW)
(*STS)
(*STL)
(*BHF)
(*MNP)
(*CPP)
value
(*TARGET)
(*VREP)
(*RECYCLE)
(*PMAINT)
(*GAS)
(*WATER)
(*SOLVENT)
(*PMSECT)
(*PMTARG)
(*PMCOEF)
value
(*STOP)
(*CONT)
(*SHUTMOW)
(*SHUTMOL)
(*SHUTMOLDOWN)
(*SHUTMOLUP)
(*SHUTALL)
(*STOP)
(*CONT)
(*SHUTMOWS)
vrep_frac
recyc_frac
sector_name
p_targ
c1 c2 c3
(*IPP)
DEFINITIONS:
'group_name_1', group_name_2, , group_name_n

Are the groups to which the following constraints apply. The wells that are
connected to each group must already have been specified using the *WELL
keyword. Production targets are apportioned according to either instantaneous
production potentials or to guide rates specified using the keyword *GUIDEP.
*MAX
Specifies that the constraint is a maximum constraint. This value becomes a
target for the group only as the result of a violation with the *CONT action.
User's Guide IMEX
*MIN
Specifies that that the constraint is a minimum constraint. This value
becomes a target for the group only as the result of a violation with the *CONT
action. *STOP and *CONT are the only actions available for MIN constraints,
and MNP (manifold pressure) and CPP (compressor or surface pressure) are
the only group production constraints which accept the MIN designation.
*TARGET
This subkeyword specifies a target production rate for the group. The
specified stream rate is set as a target to be met by the group. If another
constraint with action *CONT is violated, the rate target shifts to that
constraint and the target set by the *TARGET keyword ceases to have an
effect. There is no action associated with *TARGET since a target is not
checked for violation.
*VREP
This subkeyword specifies a voidage replacement production target. This
indicates that the production wells connected to this group produce an
amount of the bottom-hole fluid in proportion to the total bottom-hole fluid
injected into the reservoir by the injection wells connected to this group.
*RECYCLE
This subkeyword specifies a recycling production target. This indicates that
the production wells connected to this group produce such that the phase
injected by the injection wells connected to this group as specified by *GAS,
*WATER or *SOLVENT is reproduced (recycled) out of the reservoir.
*PMAINT
This subkeyword specifies that the group production rates shall be adjusted
so as to maintain the hydrocarbon volume weighted average pressure in a
particular region/sector (*PMSECT) at a desired level (*PMTARG).
*STO
constraint. Zero rates are allowed and will have the same effect as shutting
in all the wells connected to that group.
*STG
constraint. Zero rates are allowed and will have the same effect as shutting
in all the wells connected to that group.
*STW
shutting in all the wells connected to that group.
User's Guide IMEX
*STS
shutting in all the wells connected to that group.
*STL
STB/day | cm3/min) constraint. Zero rates are allowed and will have the
same effect as shutting in all the wells connected to that group.
*MNP
This subkeyword introduces a manifold pressure (kPa | psi | kPa | kg/cm2)
constraint. This may only be applied if the listed groups have all had
production specified as going through a manifold with the *MANIFOLD
keyword.
*CPP
This subkeyword introduces a compressor (surface) pressure (kPa | psi | kPa |
kg/cm2) constraint. This may only be applied if the listed groups have all
had production specified as going through a manifold with the *MANIFOLD
keyword.
*PMSECT
Introduces a single sector identified by the name sector_name whose
average hydrocarbon pore-volume pressure is to be maintained.
sector_name must be 16 characters maximum and must have already been
defined in the RESERVOIR DESCRIPTION section in the input data.
Defaulted to Entire Field.
*PMTARG
Introduces the targeted average hydrocarbon pore-volume pressure (kPa | psi
| kPa | kg/cm2) for the sector. The value p_targ must be a real number larger
than one atmosphere. Defaulted to the current sector pressure.
*PMCOEF
Introduces the control coefficients used in the pressure control strategy. The
values (c1, c2, c3) must be non-negative real numbers. Defaulted to the
internally estimated values. Currently the internal calculation of c1.c2 and c3
does not take into account aquifer influx. If aquifer influx is strong it may be
necessary for the user to optimize these coefficients
*IPP
This subkeyword specifies the use of instantaneous production potential
(IPP) to distribute the production between wells or groups to meet the
specified group production target.
User's Guide IMEX
*BHF
This subkeyword identifies a bottom-hole fluid rate (rm3/day | rbl/day |
rcm3/min) constraint. Zero rates are allowed and have the same effect as
shutting all the wells connected to the group.
vrep_frac
When the voidage replacement subkeyword is used (*VREP) the vrep_frac is
the voidage replacement ratio. A ratio of 1.0 indicates that the bottom-hole
fluid injected is completely produced.
recyc_frac
When the recycle subkeyword is used (*RECYCLE) the recyc_frac is the
recycled fraction of the indicated surface stream, which is imposed as a
group production target. A fraction of 1.0 indicates complete
reproduction/recycling of the injected surface stream.
value
Constraint value.
*STOP
Action subkeyword indicating that if the constraint cannot be met then the
simulation should be stopped.
*CONT
Action subkeyword indicating that the simulation continues with the violated
constraint switched to target constraint. This is the default action if no action
subkeyword is found.
*SHUTMOW
Action subkeyword indicating that if a maximum stock tank rate is exceeded
for a group, then the most offending well (MOW the well with the highest
rate of the named surface stream) should be shut. Not available for MNP and
CPP constraints.
*SHUTMOL
for a group, then the most offending layer (MOL) in the most offending well
(the well with the highest rate of the named surface stream) should be shut.
Not available for MNP and CPP constraints.
*SHUTMOLDOWN
for a group, then the most offending layer (MOL) and the layers below it in
the most offending well (the well with the highest rate of the named surface
stream) should be shut. Not available for MNP and CPP constraints.
User's Guide IMEX
*SHUTMOLUP
for a group, then the most offending layer (MOL) and the layers above it in
the most offending well (the one with the highest rate of the named surface
stream) should be shut. Not available for MNP and CPP constraints.
*SHUTMOWS
for a group, then a list of prioritized most offending wells (MOWS the
wells with the higher rates of the named surface stream) should be shut. Not
available for MNP and CPP constraints.
*SHUTALL
for a group, then all currently open wells in the group should be shut. Not
available for MNP and CPP constraints.
DEFAULTS:
Optional keyword. Default is no production constraints on groups. Default action is *CONT.

CONDITIONS:
This keyword must be located in the Well and Recurrent Data keyword group. A group must
be defined before it can be given any constraint values. Constraint types *TARGET,
*RECYCLE, *VREP and *PMAINT are exclusive in one timecard, and only the latest
counts. If a group is assigned by *GCONP with the following dependent constraints:
*VREP, *RECYCLE, or *PMAINT, such a group cannot be assigned by *GCONI with these
constraints for any of its injection streams. Error message will be issued if the consistency of
the dependent constraints is violated.
EXPLANATION:
*GCONP is used to specify constraints on how much fluid is produced in the group.
*GCONP can also be used to specify voidage replacement, recycling or pressure maintenance
targets (see explanations for keyword *GCONI). This allows group controls to adjust
production rates in response to injection.
*IPP or *GUIDEP (which is described on a separate manual page) are used to distribute the
injection and production rates between wells and groups. This can be done in two ways:
1. The instantaneous production potential (IPP), which is computed internally by the
simulator, can be used to determine the apportionment of production targets among
wells. This is the default.
2. The user may specify guide rates, which are then used to determine the
apportionment.
User's Guide IMEX
When IPP is used to distribute the target, each contributing group or well is assigned a rate in
proportion to its maximum production rate for the surface stream for which the target is
defined. This has the effect that either the target can be met, and all of the contributors are
assigned rates which represent the same fraction of their maximum rate, or the target cannot
be met and all contributors are assigned exactly their maximum rates.
*GUIDEP overrides *IPP if *GUIDEP is specified.
Example:
*GCONP Field
*MAX *STW 10000. *SHUTMOLDOWN
This specifies that group Field has a maximum stock tank water production constraint, with
the action upon violation being to shut the most offending layer of the most offending well
(layer and well with highest water rate) and any lower perforations.
Example:
*GCONP Group-1
*TARGET *MNP 2800.
This sets a manifold pressure target of 2800. units for the group Group-1. This group must
have had its production assigned manifold treatment with the *MANIFOLD keyword.
Group-1 must have wells directly attached to it.
Example: Use IPP to distribute production
*GCONP 'Group1'
*MAX
*STG 100000.0
*TARGET *STO
1000.0
*IPP
Example: Use oil phase guide rates to distribute production.

*GCONP 'Group1'
*MAX
*STG 100000.0
*TARGET *STO
1000.0
*GUIDEP *STO 'well-1' 'well-2' 'well-3'
200.0
150.0
650.0
User's Guide IMEX
Group Injection Constraints (Optional)
*GCONI
PURPOSE:
*GCONI is used to specify group injection controls.

FORMAT:
*GCONI 'group_name_1' 'group_name_2' ... 'group_name_n'

(*MAX)
(*TARGET)
(*VREP)
(*RECYCLE)
(*PMAINT)
(*STG)
(*STW)
(*STS)
(*BHG)
(*BHW)
(*GMP)
(*WMP)
(*SMP)
(*GCP)
(*WCP)
(*SCP)
(*STG)
(*STW)
(*STS)
(*BHG)
(*BHW)
(*GMP)
(*WMP)
(*SMP)
(*GCP)
(*WCP)
(*SCP)
(*GAS)
(*WATER)
(*SOLVENT)
(*GMKUP)
(*WMKUP)
(*SMKUP)
(*GAS)
(*WATER)
(*SOLVENT)
(*GAS)
(*WATER)
(*SOLVENT)
value
(*STOP)
(*CONT)
value
vrep_frac
recyc_frac
(make_up volume)
(*PMSECT)
(*PMTARG)
(*PMCOEF)
(*PMMAXR)
sector_name
p_targ
c1 c2 c3
d1 d2
*IIP
User's Guide IMEX
DEFINITIONS:

connected to each group must already have been specified using the *WELL
keyword. The injection targets are apportioned according to either the
instantaneous injection potentials or to the guide rates specified using the
keyword *GUIDEI.
*MAX
Specifies that the constraint is a maximum constraint which is checked for
violations. This value becomes an injection rate target for the group only as
the result of a violation with the *CONT action.
*TARGET
This subkeyword specifies a target injection rate for the group. The specified
stream rate is set as a target to be met by the group. If another constraint with
action *CONT is violated, the rate target shifts to that constraint and the target
set by the *TARGET keyword ceases to have an effect. There is no action
associated with *TARGET since a target is not checked for violation.
*STG
maximum or target. Zero rates are allowed and have the same effect as
shutting in all the gas injection wells connected to that group.
*STW
cm3/min) maximum or target. Zero rates are allowed and have the same
effect as shutting in all the water injection wells connected to that group.
*STS
effect as shutting in all the solvent injection wells connected to that group.
*BHG
This subkeyword identifies a reservoir gas rate (m3/day | scf/day | cm3/min)
maximum or target. Zero rates are allowed and have the same effect as
shutting in all the gas injection wells connected to that group.
*BHW
This subkeyword identifies a reservoir water rate (m3/day | STB/day |
effect as shutting in all the water injection wells connected to that group.
User's Guide IMEX
*VREP
This subkeyword introduces a voidage fraction injection target. This indicates
that the injection wells connected to this group inject such that the voidage
created by the producers connected to this group is replaced. In this case
*GAS, *WATER and *SOLVENT specify which phase is to be injected to
replace the voidage. If more than one phase is being injected to replace the
voidage then there must be one *VREP keyword for each phase. These primary
voidage replacement streams are handled independently. One make-up stream
can be supplemented with *GMKUP, *WMKUP or *SMKUP to meet a total
voidage replacement fraction. One of *GAS, *WATER, *SOLVENT,
*GMKUP, *WMKUP or *SMKUP must be present for each *VREP keyword.
*RECYCLE
This subkeyword introduces a recycling injection target. This indicates that the
injection wells connected to this group inject such that the phase produced by
the production wells connected to this group as specified by *GAS, *WATER
or *SOLVENT is recycled (reinjected) into the reservoir at the same surface
conditions specified for the injectors.
*PMAINT
This subkeyword specifies that the group injection rates shall be adjusted so
as to maintain the hydrocarbon volume weighted average pressure in a
particular region/sector (*PMSECT) at a desired level (*PMTARG).
*GMP
This subkeyword introduces a gas manifold pressure (kPa | psi | kPa |
kg/cm2) injection constraint. This subkeyword can only be entered if all of
the listed groups have had gas injection identified as going through a
manifold with the *MANIFOLD keyword.
*WMP
This subkeyword introduces a water manifold pressure (kPa | psi | kPa |
the listed groups have had water injection identified as going through a
*SMP
This subkeyword introduces a solvent manifold pressure (kPa | psi | kPa |
the listed groups have had solvent injection identified as going through a
*GCP
This subkeyword introduces a gas compressor (surface) pressure (kPa | psi |
kPa | kg/cm2) injection constraint. This subkeyword can only be entered if
all of the listed groups have had gas injection identified as going through a
manifold with the *MANIFOLD keyword. Also, a hydraulics table for
User's Guide IMEX
calculation of the gas manifold-to-surface pressure drop must have been

identified for all of the listed groups with the *GPTABLE keyword.
*WCP
This subkeyword introduces a water surface pressure (kPa | psi | kPa |
the listed groups have had water injection identified as going through a
calculation of the water manifold-to-surface pressure drop must have been
*SCP
This subkeyword introduces a solvent compressor (surface) pressure (kPa | psi |
kPa | kg/cm2) injection constraint. This subkeyword can only be entered if all
of the listed groups have had solvent injection identified as going through a
calculation of the solvent manifold-to-surface pressure drop must have been
*PMSECT
Introduces a single sector identified by the name sector_name whose
average hydrocarbon pore-volume pressure is to be maintained.
sector_name must be 16 characters maximum and must have already been
defined in the RESERVOIR DESCRIPTION section in the input data.
Defaulted to Entire Field.
*PMTARG
Introduces the targeted average hydrocarbon pore-volume pressure (kPa | psi
| kPa | kg/cm2) for the sector. The value p_targ must be a real number larger
than one atmosphere. Defaulted to the current sector pressure.
*PMCOEF
Introduces the control coefficients used in the pressure control strategy. The
values (c1, c2, c3) must be non-negative real numbers. Defaulted to the
internally estimated values.
*PMMAXR
Introduces the maximum surface rate of the specified injection stream that is
available for the pressure maintenance. It is the summation of two sources: a
fraction (d1) of the total produced (from the same group) and a make-up rate
(d2). The values must be non-negative real numbers. The fraction d1 is
nondimensional and defaulted to 0. The make-up rate d2 is (m3/day | bbl/day
| cm3/min) for water and (m3/day | scf/day | cm3/min) for gas or solvent, and
is defaulted to 1.0e+20.
User's Guide IMEX
*IIP
This subkeyword specifies the use of instantaneous injection potential (IIP)
to distribute the injection between wells or groups to meet the specified
group injection target.
*GAS
Specifies that the gas phase to that is to be injected for voidage replacement,
recycle pressure maintenance.
*WATER
Specifies that the water phase to that is to be injected for voidage
replacement, recycle or pressure maintenance.
*SOLVENT
Specifies that the solvent phase to that is to be injected for voidage
replacement, recycle or pressure maintenance.
*GMKUP
Specifies that gas phase is the make-up stream supplemented to meet the
total voidage replacement fraction.
*WMKUP
Specifies that water phase is the make-up stream supplemented to meet the
*SMKUP
Specifies that solvent phase is the make-up stream supplemented to meet the
value
Constraint value.
vrep_frac
When the voidage replacement subkeyword is used (*VREP) the vrep_frac is
the voidage replacement ratio. A ratio of 1.0 indicates complete voidage
replacement by the specified phase. When several values are entered (one for
each of several phases) then the vrep_frac entered for a non make-up phase is
applied as a target for that phase independently of the other specified values.
recyc_frac
When the recycle subkeyword is used (*RECYCLE) the recyc_frac is the
recycled fraction of the indicated surface stream, which is imposed as a
group injection target. A fraction of 1.0 indicates complete recycling of the
surface stream.
User's Guide IMEX
make_up volume
The amount of make up gas or solvent (m3/day | scf/day | cm3/min) to be
injected with the recycled gas or solvent. For recycled water the make up
rates are (m3/day | bbl/day | cm3/min).
*STOP
Action subkeyword indicating that if the constraint cannot be met then the
simulation should be stopped.
*CONT
Action subkeyword indicating that the simulation continues with the violated
constraint switched to target constraint.
DEFAULTS:
Optional keyword. Default is no constraints on groups, make_up volume defaults to 0.0.

CONDITIONS:
be defined, by appearing in the list directly following *GROUP in a *GROUP line or after
the *ATTACHTO keyword on a *GROUP line, before it can be given any constraint values.
Constraint types *TARGET, *RECYCLE, *VREP and *PMAINT are exclusive in one
timecard for the same injector stream (gas, water or solvent), and only the latest entry counts.
If a group is assigned by *GCONI with the following dependent constraints: *VREP,
*RECYCLE, or *PMAINT for any of the injection streams, such a group cannot be assigned
by *GCONP with these constraints for group production. In addition, only one injection
stream can be assigned with voidage makeup (*VREP *GMKUP / *WMKUP / *SMKUP) or
pressure maintenance (*PMAINT) for the same targeted group. Error message will be issued
if the consistency of the dependent constraints is violated.
EXPLANATION:
*GCONI is used to specify maximum or target fluid injection rates for the group. *GCONI
can also be used to specify voidage replacement, recycling and pressure maintenance targets.
This allows group controls to adjust injection rates in response to production.
Examples :
*GCONI Group-1 Group-2
*TARGET *STG 3000000.
This sets stock tank gas injection targets of 3000000. units per day for Group-1 and Group-2.
*GCONI Group-1
*TARGET *STG 1000000.
*MAX *GMP 2394. *CONT
User's Guide IMEX
This sets a stock tank gas target for Group-1 of 1000000. units, and a maximum gas injection
manifold pressure of 2394 units. If maintaining the targeted injection rate causes the
maximum manifold pressure to be exceeded, the group will switch to a gas manifold pressure
target and continue simulation. Group-1 must have had gas injection specified with a
*MANIFOLD keyword and must have a hydraulics table identified for the manifold-tocompressor pressure drop with the *GPTABLE keyword.
*GCONI 'FIELD'
*RECYCLE *GAS 0.8
*VREP *WMKUP 1.0
This sets the model to recycle 80% of the gas produced and make up the voidage replaced to
100% by the injection of water.
Recycling: The following keywords supplement *GCONI and add flexibility to recycling target
specification, making recycling data entry more straightforward: *GPRODGROUP,
*WPRODGROUP, *SPRODGROUP, *GSALESFRAC, *GSALESMAXR, *GFUELFRAC,
*GFUELMAXR, *WMKMAXR, *SMKMAXR, *WMKUPTO, *SMKUPTO,
*GMKMAXR, *GMKUPTO, *WRECYMAXR, *SRECYMAXR, and *GRECYMAXR.
Please see the pages in this manual describing these keywords.
A water recycling target volumetric rate is now calculated as follows in IMEX:

water_injection_target = wrecyc + wmakeup
wmakeup = min(max_water_makeup, total_water_rate - wrecyc)
wrecyc = min(fwrecyc*wprod, max_water_recycling_rate)
In the above, max_water_makeup is the quantity entered after the water recycling fraction
fwrecyc on the *RECYCLE *WATER line or directly using the *WMKMAXR keyword.
total_water_rate is a target injection rate which will be met unless a maximum make-up rate
is exceeded. total_water_rate can be entered only using the *WMKUPTO keyword. wprod
is the water production rate of, by default, the group upon which the water recycling injection
target is imposed. However wprod can be taken as the water production rate of another group
named using the *WPRODGROUP keyword. If no total_water_rate has been entered using
the *WMKUPTO keyword, wmakeup is set equal to the specified maximum make-up rate
and the min operation in the second equation is not performed.
Solvent re-injection is handled similarly to water re-injection:
solvent_injection_target = srecyc + smakeup
smakeup = min(max_solvent_makeup, total_solvent_rate - srecyc)
srecyc = min(fsrecyc*sprod, max_solvent_recycling_rate)
In the above, max_solvent_makeup is the quantity entered after the solvent recycling fraction
fsrecyc on the *RECYCLE *SOLVENT line, or directly using the *SMKMAXR keyword.
total_solvent_rate is a target injection rate which will be met unless a maximum solvent makeup rate is exceeded. total_solvent_rate can only be entered with the *SMKUPTO keyword.
sprod is the solvent production rate of, by default, the group upon which the solvent recycling
injection target is imposed. However sprod can be taken as the solvent production rate of
another group named using the *SPRODGROUP keyword. If no total_solvent_rate has been
entered using the *SMKUPTO keyword, smakeup is set equal to the specified maximum makeup rate and the min operation in the second equation is not performed.
User's Guide IMEX
Gas recycling has been enhanced to allow the specification of fuel and sales rates, and
consequently the expressions for gas recycling rates are more complicated. A volumetric gas
re- injection target is calculated in IMEX as follows:
qinj = qrecyc + qmake_up
qmake_up = min(max_makeup_rate, total_gas_rate - qrecyc)
qrecyc = min(frecyc*(qprod - fuel - sales) , qrecycmax)
fuel = min(frfuel*qprod , max_fuel_rate)
sales = min(frsales*(qprod - fuel) , max_sales_rate)
In the above, qprod is the volumetric rate of gas production from the group upon which the
recycling injection target is imposed or a producing group named using the *GPRODGROUP
keyword. max_makeup_rate is the maximum gas make-up rate entered after frecyc on the
*RECYCLE *GAS line, or separately using the *GMKMAXR keyword. total_gas_rate is a
target gas injection rate which make-up gas will be injected to meet until the maximum makeup rate is reached. total_gas_rate can be entered only using the *GMKUPTO keyword. If
the user does not specify a value for total_gas_rate using the *GMKUPTO keyword, the min
operation in the fourth equation above is not performed and the gas make-up volumetric rate
is set equal to the maximum rate.
frecyc is the gas recycling fraction entered on the *RECYCLE *GAS line. frfuel is a
maximum fraction of the gas production rate which can burned as fuel; it can be entered only
using the *GFUELFRAC keyword for the producing group. max_fuel_rate is an upper limit
to the volumetric fuel rate. It can be entered only with the *GFUELMAXR keyword for the
producing group. If no maximum fuel rate is entered the min operation in the sixth equation
is not performed. max_recycle_rate is a volumetric maximum recycling injection rate for the
targeted group which can be entered only with the *GRECYMAXR keyword; if not entered,
the min operation in the fifth equation is not performed. qprod is the gas production rate in
the group upon which the reinjection target is imposed or the producing group named using
*GPRODGROUP.
frsales is the maximum fraction of the produced gas remaining after the fuel burn which can
be diverted to sales. It can be entered only with the *GSALESFRAC keyword for the
producing group. max_sales_rate is a upper limit to the rate of diversion of produced gas to
sales; it can be entered with the *GSALESMAXR keyword for the producing group.
Pressure Maintenance: *PMAINT specifies one special group control which instructs a
group or groups to adjust the production or injection rates (at reservoir condition) in order to
maintain the average hydrocarbon pore-volume pressure in a particular region (sector) at a
desired level. The group runs essentially on a production (or injection) target rate Qtarg of
bottom-hole fluid in reference to that of the group injection (or production), Qref:
Qtarg = Qref Q
where Q is the bottom-hole correction rate needed to reach or maintain a certain pressure.
The sign designates for production (+) or injection () target. Q has been set to zero for
normal voidage replacement constraints (*VREP under *GCONP / *GCONI). However,
setting Q = 0 may not exactly maintain the pressure due to the involvement of compressible
fluids, and the different reservoir conditions under which Qtarg and Qref are calculated.
User's Guide IMEX
With the pressure control strategy, the correction volumetric rate is calculated as
1
dP
Q = c1 P Pt arg + c 2 + c 3
t
dt
t t
(P Pt arg )dt
where P is the sector pressure at the current time t, and t is the integration interval over the
last several timesteps. The empirical constants c1, c2 and c3 penalize/minimize the object
terms of (1) the difference between current pressure and the target, (2) the pressure
derivative, and (3) the time-averaged pressure difference.
There is little to guide the choice of values of c1, c2, c3. Heuristically, c1 is a damping term
which determines how quickly P will move toward P_targ, and controls the overall accuracy.
Larger values of c1 give a quicker approach to P_targ but may lead to pressure oscillations or
even divergence if it is set too high. As P gets close to P_targ, the derivative term becomes
significant and acts to decrease the magnitude of the time derivative of pressure to avoid
overshooting. The larger c2 is the less likely is overshoot, but the approach to P_targ is
slowed. c3 acts to prevent long-term persistence of pressures either above or below the target.
Larger values of c3 more quickly overcome differences from P_targ, at the possible cost of
some oscillatory behavior.
The Well Management estimates and sets the above constants internally at the beginning of
pressure maintenance if the subkeyword *PMCOEF is absent after *PMAINT, by assuming
that the pressure follows an exponential decline to the set target under pure depletion of an
isothermal point reservoir. It should be noted that there is no guarantee that such a rough
estimate may work most favorably in general situations. Nevertheless, the user is
recommended to run the simulator with the internally-set coefficients. By judging the
pressure convergence behavior in terms of fluctuation and/or shifting, optimal values might
then be achieved by tuning the internally-set values at possibly the same orders and input
through subkeyword *PMCOEF.
Example:
*DATE 1985 1 1
*GCONP Grp-1 *TARGET *BHF 300.
*GCONI Grp-1 *TARGET *STW 150.
*GCONI Grp-1 *TARGET *STG 29000.
** (1)
** (2)
** (3)
*DATE 1987 1 1
*GCONI Grp-1 *PMAINT *GAS
** (4)
Maintain sector Entire Field (default) average pressure at the current value as dated
(default) by adjusting group Grp-1 gas injection rate, using the internally-set control
parameters (default). Line (3) is overridden by line (4). The production and injection rates are
apportioned using the instantaneous potential method (default).
Example:
This is an example using voidage replacement. Here 50% of the voidage is being replaced by
solvent injection and the other 50% by gas injection. Note that voidage replacement is done
at reservoir conditions and consequently surface rates will not agree with the ratio of 50%,
even though the reservoir rates will be in a ratio of 50%.
User's Guide IMEX
*GCONI 'Group1'
*VREP *SOLVENT 0.5
*VREP *GAS
0.5
*IIP
*IIP or *GUIDEI (which is described on a separate manual page) are used to distribute a
group injection target among the contributing wells or groups. This can be done in two ways:
1. The instantaneous injection potential (IIP), which is computed internally by the
simulator, can be used to determine the apportionment of injection targets among
wells or groups. This is the default.
2. The user may specify guide rates, which are then used to determine the apportionment.
When IIP is used to distribute the target, each contributing group or well is assigned a rate in
proportion to its maximum injection rate for the surface stream for which the target is
defined. This has the effect that either the target can be met, and all of the contributors are
assigned rates which represent the same fraction of their maximum rate, or the target cannot
be met and all contributors are assigned exactly their maximum rates.
*GUIDEI overrides *IIP if *GUIDEI is specified.
Example:
Use IIP to distribute injection

*GCONI 'Group1'
*TARGET *STW 1000.0
*IIP
Example:
Use gas phase guide rates to distribute injection.

*GCONI'Group1'
*TARGET *STG
*GUIDEI *STG
User's Guide IMEX
50000.0
'well-1'
2000.0
'well-2'
1500.0
'well-3'
6500.0
Gas Make-up Target for Group Gas Recycling (Optional)

*GMKUPTO
PURPOSE:
*GMKUPTO is used to specify a total group surface gas injection rate which make-up gas, in
general up to a certain maximum rate, will supplement the recycled gas to meet.
In other words the gas rate specified for injection is the rate on the *GMKUPTO keyword. The
rate of make-up gas actually injected may be reduced by the rate of gas being recycled and may
account for the maximum make-up gas rate specified on the *GMKMAXR keyword.
The rate specified on the *GMKUPTO keyword will also be referred to as the make_up_to
rate. The maximum gas make-up rate can be specified either with the *GMKMAXR keyword
or on the *RECYCLE *GAS line following *GCONI. The total injection rate can be entered
only with *GMKUPTO. The rates entered here have an effect only if a gas recycling
injection target is in force. Gas recycling injection targets are specified for groups using the
*GCONI keyword. The total injection rate specified here differs from a group STG target in
that the target entered here will not be met if the make-up gas would have to exceed a
specified maximum rate in order to meet the total rate.
FORMAT:
*GMKUPTO group_name_list total_gas_rates

DEFINITIONS:
group_name_list
A list of one or more valid group names, enclosed in single quotes. The
groups named in this list will acquire the total gas rates specified after the
list. The list need not be contained on a single line.
total_gas_rates
A set of non-negative total injected gas rate values (m3/day | SCF/day |
cm3/min). If only one rate value is entered it is applied to all of the groups
listed. If more than one rate value is entered, then the number of rates entered
must equal the number of groups in the list and the first rate is assigned to the
first group, etc. Zero total rates may be entered, but they are interpreted as
meaning that the groups gas make-up rate is controlled completely by the
maximum make-up rate (make_up_to rate) entered using the *GMKMAXR
keyword. If neither a maximum rate using *GMKMAXR nor a make_up_to
*GMKUPTO rate is specified, then the make-up rate is taken to be zero. The
total gas rate values can be, but need not be, on the same line as the last name
in the group list. The rate values may be spread over more than a single line.
DEFAULTS:
Optional keyword. If not specified, the gas make-up rate is controlled by maximum rates
entered using the *GMKMAXR keyword; if no maximum rate is entered either, then no
make-up gas is injected.
User's Guide IMEX
CONDITIONS:
This keyword must be located in the WELL AND RECURRENT DATA keyword group. A
group must be defined before it can appear in the group list following *GMKUPTO.
Negative values are rejected as being in error.
EXPLANATION:
For a full explanation of how the total gas injection rate enters the calculation of a gas
recycling injection target, please see the pages of this manual describing the *GCONI
keyword. It is important to note that the make up to *GMKUPTO rate is associated with
the group re-injecting the gas, and not with the group producing it. A specified total injected
gas rate has no effect unless a group gas recycling target is currently in force (see the
*GCONI entry in this manual.). The group gas injection rate will be set to the total gas rate
unless this total rate exceeds the injection rate of recycled gas by more than a maximum
make-up rate, in which case the total gas injection rate is set to the recycling rate plus the
maximum make-up rate.
Example : Group1 and Group2 both acquire the single specified total gas injection rate.
*GMKUPTO 'Group1 Group2'
5.0D+7
Here the total gas value is, but need not have been, on the same line as the group names.
User's Guide IMEX
Maximum Make-up Gas Rate for Group Gas Recycling

(Optional)
*GMKMAXR
PURPOSE:
*GMKMAXR is used to specify the maximum surface rate of make-up gas injected as part of
a group gas recycling target to supplement the recycled fluid. This information can also be
specified in a somewhat less straightforward way using the *RECYCLE *RATE
subkeywords of *GCONI when *RECYCLE *GAS is specified. The two methods of
entering the make-up rate can over-write each other. Gas recycling injection targets are
specified for groups using the *GCONI keyword. The numbers entered under *GMKMAXR
have no effect unless a group gas recycling injection target is currently in force.
FORMAT:
*GMKMAXR group_name_list max_rates

DEFINITIONS:
group_name_list
groups named in this list will acquire the maximum make-up rates specified
after the list. The list need not be contained on a single line.
max_rates
A set of non-negative maximum make-up rate values (m3/day | SCF/day |
listed. If more than one rate value is entered, then the number of rates
entered must equal the number of groups in the list and the first rate is
assigned to the first group, etc. Zero maximum rates may be entered, but
they are interpreted as meaning that the groups make-up rate is controlled
completely by the rate entered on the *GMKUPTO keyword. If neither a
maximum rate nor a make_up_to *GMKUPTO rate is specified, then the
make-up rate is taken to be zero. The maximum rate values can be, but need
not be, on the same line as the last name in the group list. The rate values
may be spread over more than a single line.
DEFAULTS:
Optional keyword. If not specified, make-up rate is controlled by the rates entered using the
*GMKUPTO keyword; if no rate is entered using the *GMKUPTO keyword either, then no
make-up gas is injected. Maximum make-up rates are also specified using the *RECYCLE
*RATE subkeywords of *GCONI.
CONDITIONS:
group must be defined before it can appear in the group list following *GMKMAXR.
User's Guide IMEX
EXPLANATION:
For a full explanation of how the maximum make-up rate enters the calculation of a gas
keyword. It is important to note that the maximum make-up rate is associated with the group
re-injecting the gas, and not with the group producing it.
Example : Group1 and Group2 both acquire the single specified maximum make-up gas
injection rate.
*GMKMAXR 'Group1 Group2'
1.0D+7
Here the maximum rate value is, but need not have been, on the same line as the group names.
User's Guide IMEX
Gas Producing Group for Group Recycling (Optional)

*GPRODGROUP
PURPOSE:
*GPRODGROUP is used to specify the group from which gas production rates will be taken
to calculate a gas recycling injection target specified for another group or groups using the
*GCONI keyword.
FORMAT:
*GPRODGROUP group_name_list *FROM 'gas_producing_group_name'

DEFINITIONS:
group_name_list
groups named in this list will have gas recycling injection targets based on
gas production from the named gas producing group. The list need not be
contained on a single line.
*FROM
This subkeyword indicates that all groups named in the list preceding
*FROM are to have their recycling injection targets calculated using the gas
production from the single group with the name
gas_producing_group_name. *FROM may occur on the same line as the
last group in the list but need not. It is valid for *FROM to be on a new line.
gas_producing_group_name
A single valid group name, enclosed in single quotes. This group determines
gas rates and composition for recycling injection targets imposed upon the
listed groups. This group name may be but need not be on the same line as
the *FROM subkeyword.
DEFAULTS:
Optional keyword. If a groups name never appears in a group list following

*GPRODGROUP, then by default that groups own gas production rate and composition are
used to compute gas recycling targets imposed on the group.
CONDITIONS:
group must be defined before it can appear in the group list following *GPRODGROUP.
The fraction of gas recycled to each group in the group_name_list must sum to no more
than 1.0. It is possible to reinject more gas than is produced from the gas producing group
if this is not ensured (The fraction of gas recycled for each group is entered on its *GCONI
keyword after the *RECYCLE *GAS sub-keyword).
User's Guide IMEX
EXPLANATION:
The imposition of a recycling rate target involves using the gas production rates of one group
to determine the gas injection rates in the same or another group. Recycling targets can
involve some additional parameters such as sales gas rate and fuel gas rate which are also
associated with the producing group rather than with the injecting group. Please consult the
pages of this manual describing the *GCONI keyword for a full explanation of the
calculation of gas re-injection targets. Further information is also available from the manual
pages for the other gas recycling keywords, which are *GSALESFRAC, *GSALESMAXR,
*GFUELFRAC, *GFUELMAXR, *GMKMAXR, *GMKUPTO, and *GRECYMAXR.
Example : Group1 and Group2 have their gas recycling injection targets calculated using the
gas produced by Group3.
*GPRODGROUP 'Group1 Group2' *FROM Group3
User's Guide IMEX
Maximum Re-injection Rate for Group Gas Recycling

(Optional)
*GRECYMAXR
PURPOSE:
*GRECYMAXR is used to specify a maximum rate at which gas is re-injected under a group
recycling target. Gas recycling injection targets are specified for groups using the *GCONI
keyword. The numbers entered under *GRECYMAXR have no effect unless a group gas
recycling injection target is currently in force.
FORMAT:
*GRECYMAXR group_name_list max_recyc_rates

DEFINITIONS:
group_name_list
groups named in this list will acquire the maximum gas recycling rates
specified after the list. The list need not be contained on a single line.
max_recyc_rates
A set of non-negative real values (m3/day | scf/day | cm3/min). Zero is accepted
as a value but is interpreted as signifying that no maximum recycling rate is set.
If only a single value is supplied, that value will be applied to all groups in the
list. If more than one value is supplied, the number of maximum recycling rates
must equal the number of group names in the list and the first rate will be applied
to the first group, etc. These maximum rates set, or set an upper limit for, the
volumetric rate at which produced gas is re-injected under a group gas recycling
target. The values may be spread over more than a single line.
DEFAULTS:

*GRECYMAXR, then by default its maximum gas recycling rate is set to zero, which is
interpreted to mean that there is no fixed maximum imposed upon the recycling injection rate.
CONDITIONS:
group must be defined before it can appear in the group list following *GRECYMAXR.
EXPLANATION:
For a full explanation of how the maximum recycling rates enter the calculation of a gas
keyword. It is important to note that the maximum recycling rate values are associated with
the group re-injecting the gas, and not with the group producing it. The recycling volumetric
injection rate is set principally by the recycling fraction entered with the *RECYCLE *GAS
line following *GCONI, but if a maximum has been imposed with the *GRECYMAXR
keyword, the re-injection target is not allowed to exceed this maximum.
User's Guide IMEX
Example : Group1 acquires a maximum recycling injection rate of 3.0D+07 SCF/day and
Group2 of 3.5D+07 SCF/day.
*GRECYMAXR
'Group1
3.0D+07
Group2'
3.5D+07
Here the maximum recycling rates are, but need not have been, on a separate line from the
group names.
User's Guide IMEX
Fuel Consumption Fraction for Group Gas Recycling

(Optional)
*GFUELFRAC
PURPOSE:
*GFUELFRAC is used to specify a maximum fraction of the surface gas produced by a group
which is available for consumption as fuel. Fuel consumption reduces the amount of gas
available for re-injection by the same or another group. Gas recycling injection targets are
specified for groups using the *GCONI keyword. The numbers entered under
*GFUELFRAC have no effect unless a group gas recycling target is currently in force.
FORMAT:
*GFUELFRAC group_name_list fuel_fractions

DEFINITIONS:
group_name_list
groups named in this list will take the fuel fraction values specified after the
fuel_fractions
A set of real values between 0 and 1 inclusive. They may all be zero. If only
a single value is supplied, that value will be applied to all groups in the list.
If more than one value is supplied, the number of fuel fractions must equal
the number of group names in the list and the first fraction will be applied to
the first group, etc. These fractions set, or set an upper limit for, the fraction
of the volumetric rate of gas production which is consumed as fuel, and in
this way affect the amount of gas available for re-injection under a gas
recycling target. The fractions may be spread over more than a single line.
DEFAULTS:
Optional keyword. If a groups name never appears in a group list following *GFUELFRAC,
then by default its fuel fraction value is zero and no gas is consumed as fuel unless a non-zero
fixed fuel consumption rate is set with the *GFUELMAXR keyword.
CONDITIONS:
group must be defined before it can appear in the group list following *GFUELFRAC.
Values outside the range [0,1] are rejected as being in error.
EXPLANATION:
For a full explanation of how the fuel fractions enter the calculation of a gas recycling
injection target, please see the pages of this manual describing the GCONI keyword. It is
important to note that the fuel fraction values are associated with the group producing the reinjected gas, and not with the group injecting it. Fuel consumption will be set at a volumetric
rate of fuel_fraction*gas_production_rate unless a fixed maximum fuel rate has been set
using the *GFUELMAXR keyword.
User's Guide IMEX
Example : Group1 acquires a fuel fraction of 0.10 and Group2 of 0.12.

*GFUELFRAC
'Group1 Group2'
0.10
0.12
Here the fuel fraction values are, but need not have been, on a separate line from the group
names.
User's Guide IMEX
Maximum Fuel Consumption Rate for Group Gas Recycling

(Optional)
*GFUELMAXR
PURPOSE:
*GFUELMAXR is used to specify a maximum rate at which the surface gas produced by a
group is made available for consumption as fuel. Fuel consumption reduces the amount of
gas available for re-injection by the same or another group. Gas recycling injection targets
are specified for groups using the *GCONI keyword. The numbers entered using
*GFUELMAXR have no effect unless a group gas-recycling target is in force.
FORMAT:
*GFUELMAXR group_name_list max_fuel_rates

DEFINITIONS:
group_name_list
groups named in this list will acquire the maximum fuel rates specified after
the list. The list need not be contained on a single line.
max_fuel_rates
A set of non-negative real values (m3/day | SCF/day | cm3/min). Zero is
accepted as a value but is interpreted as signifying that no maximum fuel rate is
set. If only a single value is supplied, that value will be applied to all groups in
the list. If more than one value is supplied, the number of maximum fuel rates
must equal the number of group names in the list and the first rate will be applied
to the first group, etc. These maximum rates set, or set an upper limit for, the
volumetric rate of gas production consumed as fuel and in this way affect the
amount of gas re-injected under a gas-recycling target. The values may be
spread over more than a single line.
DEFAULTS:

*GFUELMAXR, then by default its maximum fuel rate is zero and no gas is consumed as
fuel unless a non-zero fuel fraction is set with the *GFUELFRAC keyword.
CONDITIONS:
group must be defined before it can appear in the group list following *GFUELMAXR.
User's Guide IMEX
EXPLANATION:
For a full explanation of how the maximum fuel rates enter the calculation of a gas recycling
important to note that the maximum fuel rate values are associated with the group producing the
re-injected gas, and not with the group injecting it. Fuel consumption will be set at the entered
maximum fuel rate unless a fuel fraction values has been set for the producing group
*GFUELFRAC keyword. If both the fuel fraction and the maximum fuel rate are set to zero,
then there is no fuel consumption. If only one of the fraction or maximum rate is entered then
the entered quantity is applied. If both are entered then the smaller of the two fuel rates is used.
Example : Group1 acquires a maximum fuel rate of 3.0D+06 SCF/day and Group2 of
3.5D+06 SCF/day.
*GFUELMAXR
'Group1
3.0D+06
Group2'
3.5D+06
Here the maximum fuel rates are, but need not have been, on a separate line from the group
names.
User's Guide IMEX
Sales Rate Fraction for Group Gas Recycling (Optional)

*GSALESFRAC
PURPOSE:
*GSALESFRAC is used to specify a maximum fraction of the surface gas produced by a

group (less the amount consumed as fuel) which is available for sale. Diversion to sales
reduces the amount of gas available for re-injection by the same or another group. Gas
recycling injection targets are specified for groups using the *GCONI keyword. Numbers
entered under the *GSALESFRAC keyword have no effect unless a group gas recycling
injection target is currently in force.
FORMAT:
*GSALESFRAC group_name_list sales_fractions

DEFINITIONS:
group_name_list
groups named in this list will acquire the sales fractions specified after the
sales_fractions
A set of real values between 0 and 1 inclusive. They may all be zero. If only
a single value is supplied, that value will be applied to all groups in the list.
If more than one value is supplied, the number of sales fractions must equal
the number of group names in the list and the first fraction will be applied to
the first group, etc. These fractions set, or set an upper limit for, the fraction
of the volumetric rate of produced gas, less the fuel consumption rate, which
is re-injected under a gas recycling target. The fractions may be spread over
more than a single line.
DEFAULTS:

*GSALESFRAC, then by default its sales fraction value is zero and no gas is diverted for
sales unless a non-zero fixed sales rate is set with the *GSALESMAXR keyword.
CONDITIONS:
group must be defined before it can appear in the group list following *GSALESFRAC.
Values outside the range [0,1] are rejected as being in error.
User's Guide IMEX
EXPLANATION:
For a full explanation of how sales fractions enter the calculation of a gas recycling injection
target, please see the pages of this manual describing the GCONI keyword. It is important to
note that the sales fraction values are associated with the group producing the re-injected gas,
and not with the group injecting it. Sales will be set at a volumetric rate of
sales_fraction*gas_production_rate
unless a fixed maximum sales rate has been set using the *GSALESMAXR keyword.
Example : Group1 acquires a sales fraction of 0.20 and Group2 of 0.24.
*GSALESFRAC
'Group1 Group2'
0.20
0.24
Here the sales fraction values are, but need not have been, on a separate line from the group
names. Any group specifying Group1 or Group2 as its recycling production group with the
*GPRODGROUP keyword will have the sales fractions set here used in calculating a gas
recycling injection target set using the *GCONI keyword.
User's Guide IMEX
Maximum Sales Rate for Group Gas Recycling (Optional)

*GSALESMAXR
PURPOSE:
*GSALESMAXR is used to specify a maximum rate at which the surface gas produced by a
group is made available for sales. Diversion to sales reduces the amount of gas available for
re-injection by the same or another group. Gas recycling injection targets are specified for
groups using the *GCONI keyword. The numbers entered under *GSALESMAXR have no
effect unless a group recycling target is currently in force.
FORMAT:
*GSALESMAXR group_name_list max_sales_rates

DEFINITIONS:
group_name_list
groups named in this list will acquire the maximum sales rates specified after
the list. The list need not be contained on a single line.
max_sales_rates
A set of non-negative real values (m3/day | SCF/day cm3/min). Zero is
accepted as a value but is interpreted as signifying that no maximum sales
rate is set. If only a single value is supplied, that value will be applied to all
groups in the list. If more than one value is supplied, the number of
maximum sales rates must equal the number of group names in the list and
the first rate will be applied to the first group, etc. These maximum rates set,
or set an upper limit for, the volumetric rate of gas production diverted to
sales and in this way affect the amount of gas re-injected under a gas
recycling target. The values may be spread over more than a single line.
DEFAULTS:

*GSALESMAXR, then by default its maximum sales rate is zero and no gas is sold unless a
non-zero sales fraction is set with the *GSALESFRAC keyword.
CONDITIONS:
group must be defined before it can appear in the group list following *GSALESMAXR.
User's Guide IMEX
EXPLANATION:
For a full explanation of how the maximum sales rate enters the calculation of a gas recycling
important to note that the maximum sales rate values are associated with the group producing
the re-injected gas, and not with the group injecting it. Sales will be set at the entered
maximum rate unless a sales fraction value has been set for the producing group with the
*GFUELFRAC keyword. If both the sales fraction and the maximum sales rate are set to
zero, then there are no sales. If only one of the fraction or maximum rate is entered then the
entered quantity is applied. If both are entered then the smaller of the two sales rates is used.
Example : Group1 acquires a maximum sales rate of 6.0D+06 SCF/day and Group2 of
7.0D+06 scf/day.
*GSALESMAXR
'Group1
6.0D+06
Group2'
7.0D+06
Here the maximum sales rates are, but need not have been, on a separate line from the group
names.
User's Guide IMEX
Water Make-up Target for Group Water Recycling (Optional)

*WMKUPTO
PURPOSE:
*WMKUPTO is used to specify a total group water injection rate which make-up water, in
general up to a certain maximum rate, will supplement the recycled water to meet.
In other words the water rate specified for injection is the rate on the *WMKUPTO keyword.
The rate of make-up water actually injected may be reduced by the rate of water being
recycled and also may account for the maximum make-up water rate specified on the
*WMKMAXR keyword.
The rate specified on the *WMKUPTO keyword will also be referred to as the make_up_to
rate. The maximum make-up rate can be specified either with the *WMKMAXR keyword or
the *RECYCLE *WATER subkeywords of *GCONI. The total injection rate can be entered
only with *WMKUPTO. The rates entered here have an effect only if a water recycling
injection target is in force. Water recycling injection targets are specified for groups using
the *GCONI keyword. The total rate specified here differs from a group STW target in that
the target entered here will not be met if the make-up water would have to exceed a specified
maximum rate in order to meet the total rate.
FORMAT:
*WMKUPTO group_name_list total_water_rates

DEFINITIONS:
group_name_list
groups named in this list will acquire the total water rates specified after the
total_water_rates
A set of non-negative total injected water rate values (m3/day | STB/day |
meaning that the groups water make-up rate is controlled completely the
maximum make-up rate (make_up_to rate) entered using the *WMKMAXR
keyword. If neither a maximum rate nor a make_up_to *WMKUPTO rate is
specified, then the make-up rate is taken to be zero. The total water rate values
can be, but need not be, on the same line as the last name in the group list. The
rate values may be spread over more than a single line.
DEFAULTS:
Optional keyword. If not specified, the water make-up rate is controlled by maximum rates
entered using the *WMKMAXR keyword; if no maximum rate is entered either, then no
make-up water is injected.
User's Guide IMEX
CONDITIONS:
group must be defined before it can appear in the group list following *WMKUPTO.
EXPLANATION:
For a full explanation of how the total water injection rate enters the calculation of a water
keyword. It is important to note that the make up to *WMKUPTO rate is associated with
the group re-injecting the water, and not with the group producing it. A specified total
injected water rate has no effect unless a group water recycling target is currently in force
(see the *GCONI entry in this manual.). The group water injection rate will be set to the total
water rate unless this total rate exceeds the injection rate of recycled water by more than a
maximum make-up rate, in which case the water injection target is set to the recycling rate
plus the maximum make-up rate.
Example : Group1 and Group2 both acquire the single specified total water injection rate.
*WMKUPTO 'Group1 Group2'
300.
Here the total water value is, but need not have been, on the same line as the group names.
User's Guide IMEX
Maximum Water Make-up Rate for Group Water Recycling

(Optional)
*WMKMAXR
PURPOSE:
*WMKMAXR is used to specify the maximum surface rate of make-up water injected as part
of a group water recycling target to supplement the recycled water. This information can also
be specified using the *RECYCLE *WATER subkeywords of *GCONI. The two methods of
entering the make-up rate can over-write each other. Water recycling injection targets are
specified for groups using the *GCONI keyword.
FORMAT:
*WMKMAXR group_name_list max_rates

DEFINITIONS:
group_name_list
max_rates
A set of non-negative maximum water make-up rate values (m3/day |
STB/day | cm3/min). If only one rate value is entered it is applied to all of
the groups listed. If more than one rate value is entered, then the number of
rates entered must equal the number of groups in the list and the first rate is
they are interpreted as meaning that the groups water make-up rate is
controlled completely by the rate entered on the *WMKUPTO keyword. If
neither a maximum rate nor a make_up_to *WMKUPTO rate is specified,
then the make-up rate is taken to be zero. The maximum rate values can be,
but need not be, on the same line as the last name in the group list. The rate
values may be spread over more than a single line.
DEFAULTS:
Optional keyword. If not specified, the water make-up rate is controlled by target rates
entered using the *WMKUPTO keyword; if no target rate is entered either, then no make-up
water is injected. Maximum make-up rates are also specified using the *RECYCLE
*WATER subkeywords of *GCONI.
CONDITIONS:
group must be defined before it can appear in the group list following *WMKMAXR.
User's Guide IMEX
EXPLANATION:
For a full explanation of how the maximum water make-up rate enters the calculation of a
water recycling injection target, please see the pages of this manual describing the *GCONI
keyword. It is important to note that the maximum water make-up rate is associated with the
group re-injecting the water, and not with the group producing it.
Example : Group1 and Group2 both acquire the single specified maximum make-up water
injection rate.
*GMKMAXR 'Group1 Group2'
300.
Here the maximum rate value is, but need not have been, on the same line as the group names.
User's Guide IMEX
Water Producing Group for Group Recycling (Optional)

*WPRODGROUP
PURPOSE:
*WPRODGROUP is used to specify the group from which water production rates will be
taken to calculate a water recycling injection target specified for another group or groups
using the *GCONI keyword.
FORMAT:
*WPRODGROUP group_name_list *FROM 'water_producing_group_name'

DEFINITIONS:
group_name_list
groups named in this list will have water recycling injection targets based on
water production from the named water producing group. The list need not
be contained on a single line.
*FROM
This subkeyword indicates that all groups named in the list preceding
*FROM are to have their water recycling injection targets calculated using
the water production from the single group with the name
water_producing_group_name. *FROM may occur on the same line as the
last group in the list but need not. It is valid for *FROM to be on a new line.
water_producing_group_name
water rates for recycling injection targets imposed upon the listed groups.
This group name may be but need not be on the same line as the *FROM
subkeyword.
DEFAULTS:

*WPRODGROUP, then by default that groups own water production rate is used to compute
water recycling targets imposed on the group.
CONDITIONS:
group must be defined before it can appear in the group list following *WPRODGROUP.
The fraction of water recycled to each group in the group_name_list must sum to no more
than 1.0. It is possible to reinject more water than is produced from the water producing
group if this is not ensured (The fraction of water recycled for each group is entered on its
*GCONI keyword after the *RECYCLE *WATER sub-keyword).
User's Guide IMEX
EXPLANATION:
The imposition of a recycling rate target involves using the water production rates of one
group to determine the water injection rate in the same or another group. Please consult the
calculation of water re-injection targets. Further information is also available from the
manual pages for the other water recycling keywords, which are *WMKMAXR,
*WMKUPTO, and *WRECYMAXR
Example : Group1 and Group2 have their water recycling injection targets calculated using
the gas produced by Group3.
*WPRODGROUP 'Group1 Group2' *FROM Group3
User's Guide IMEX
Maximum Re-injection Rate for Group Water Recycling

(Optional)
*WRECYMAXR
PURPOSE:
*WRECYMAXR is used to specify a maximum rate at which water is re-injected under a

group recycling target. Water recycling injection targets are specified for groups using the
*GCONI keyword. The numbers entered under *WRECYMAXR have no effect unless a
group water recycling injection target is currently in force.
FORMAT:
*WRECYMAXR group_name_list max_recyc_rates

DEFINITIONS:
group_name_list
groups named in this list will acquire the maximum recycling rates specified
max_recyc_rates
A set of non-negative real values (m3/day | STB/day | cm3/min). Zero is
accepted as a value but is interpreted as signifying that no maximum
recycling rate is set. If only a single value is supplied, that value will be
applied to all groups in the list. If more than one value is supplied, the
number of maximum recycling rates must equal the number of group names
in the list and the first rate will be applied to the first group, etc. These
maximum rates set, or set an upper limit for, the volumetric rate at which
produced water is re-injected under a group water recycling target. The
values may be spread over more than a single line.
DEFAULTS:

*WRECYMAXR, then by default its maximum water recycling rate is set to zero, which is
interpreted to mean that there is no fixed maximum imposed upon the recycling injection rate.
CONDITIONS:
group must be defined before it can appear in the group list following *WRECYMAXR.
User's Guide IMEX
EXPLANATION:
For a full explanation of how the maximum recycling rates enter the calculation of a water
the group re-injecting the water, and not with the group producing it. The recycling
volumetric injection rate is set principally by the recycling fraction entered upon the
*RECYCLE *WATER line following *GCONI, but if a maximum has been imposed with
the *WRECYMAXR keyword, the re-injection rate is not allowed to exceed this maximum.
Example: Group1 acquires a maximum recycling injection rate of 400. STB/day and Group2
of 450. STB/day.
*WRECYMAXR
'Group1
400.
Group2'
450.
group names.
User's Guide IMEX
Solvent Make-up Target for Group Solvent Recycling

(Optional)
*SMKUPTO
PURPOSE:
*SMKUPTO is used to specify a total group surface solvent injection rate which make-up
solvent, in general up to a certain maximum rate, will supplement the recycled solvent to meet.
In other words, the solvent rate specified for injection is the rate on the *SMKUPTO
keyword. The rate of make-up solvent actually injected may be reduced by the rate of solvent
being recycled and also may account for the maximum make-up solvent rate specified on the
*SMKMAXR keyword.
The rate specified on the *SMKUPTO keyword This will also be referred to as the
make_up_to rate. The maximum solvent make-up rate can be specified either with the
*SMKMAXR keyword or on the *RECYCLE *SOLVENT line following *GCONI. The
total solvent injection rate can be entered only with *SMKUPTO. The rates entered here
have an effect only if a solvent recycling injection target is in force. Solvent recycling
injection targets are specified for groups using the *GCONI keyword. The total injection rate
specified here differs from a group STS target in that the target entered here is not be met if
the make-up solvent must exceed a specified maximum rate in order to meet the total rate.
FORMAT:
*SMKUPTO group_name_list total_inj_rates

DEFINITIONS:
group_name_list
groups named in this list will acquire the total solvent rates specified after the
total_inj_rates
A set of non-negative total injected solvent rate values (m3/day | SCF/day |
meaning that the groups solvent make-up rate is controlled completely by the
maximum make-up rate (make_up_to rate) entered using the *SMKMAXR
keyword. If neither a maximum rate nor a make up to *SMKUPTO rate is
specified, then the make-up rate is taken to be zero. The total solvent injection
rate values can be, but need not be, on the same line as the last name in the
group list. The rate values may be spread over more than a single line.
DEFAULTS:
Optional keyword. If not specified, the solvent make-up rate is controlled by maximum rates
entered using the *SMKMAXR keyword; if no maximum rate is entered either, then no
make-up solvent is injected.
User's Guide IMEX
CONDITIONS:
group must be defined before it can appear in the group list following *SMKUPTO.
EXPLANATION:
For a full explanation of how the total solvent injection rate enters the calculation of a solvent
keyword. It is important to note that the make up to *SMKUPTO rate is associated with
the group re-injecting the solvent, and not with the group producing it. A specified total
injected solvent rate has no effect unless a group solvent recycling target is currently in force
(see the *GCONI entry in this manual.). The group solvent injection rate will be set to the
total solvent rate unless this total rate exceeds the injection rate of recycled solvent by more
than a maximum make-up rate, in which case the total solvent injection rate is set to the
recycling rate plus the maximum make-up rate.
Example : Group1 and Group2 both acquire the single specified total solvent injection rate.
*SMKUPTO 'Group1 Group2'
5.0D+7
Here the total solvent injection rate value is, but need not have been, on the same line as the
group names.
User's Guide IMEX
Maximum Make-up Solvent Rate for Group Solvent Recycling

(Optional)
*SMKMAXR
PURPOSE:
*SMKMAXR is used to specify the maximum surface rate of make-up solvent injected as
part of a group solvent recycling target to supplement the recycled fluid. This information
can also be specified after the recycling fraction on the *RECYCLE *SOLVENT line after
*GCONI. The two methods of entering the make-up rate can over-write each other. Solvent
recycling injection targets are specified for groups using the *GCONI keyword. The
numbers entered under *SMKMAXR have no effect unless a group solvent recycling
injection target is currently in force.
FORMAT:
*SMKMAXR group_name_list max_rates

DEFINITIONS:
group_name_list
max_rates
A set of non-negative maximum make-up rate values (m3/day | SCF/day |
listed. If more than one rate value is entered, then the number of rates
entered must equal the number of groups in the list and the first rate is
they are interpreted as meaning that the groups make-up rate is controlled by
the rate entered on the *SMKUPTO keyword. If neither a maximum rate nor
a make_up_to *SMKUPTO rate is specified, then the make-up rate is
taken to be zero. The maximum rate values can be, but need not be, on the
same line as the last name in the group list. The rate values may be spread
over more than a single line.
DEFAULTS:
Optional keyword. If not specified, make-up rate is controlled by target rates entered using
the *SMKUPTO keyword; if no target rate is entered either, then no make-up solvent is
injected.
CONDITIONS:
group must be defined before it can appear in the group list following *SMKMAXR.
User's Guide IMEX
EXPLANATION:
For a full explanation of how the maximum make-up rate enters the calculation of a solvent
keyword. It is important to note that the maximum make-up rate is associated with the group
re-injecting the solvent, and not with the group producing it.
Example : Group1 and Group2 both acquire the single specified maximum make-up solvent
injection rate.
*SMKMAXR 'Group1 Group2'
1.0D+7
Here the maximum rate value is, but need not have been, on the same line as the group
names.
User's Guide IMEX
Solvent Producing Group for Group Recycling (Optional)

*SPRODGROUP
PURPOSE:
*SPRODGROUP is used to specify the group from which solvent production rates will be
taken to calculate a solvent recycling injection target specified for another group or groups
using the *GCONI keyword.
FORMAT:
*SPRODGROUP group_name_list *FROM 'solvent_producing_group_name'

DEFINITIONS:
group_name_list
groups named in this list will have solvent recycling injection targets based
on solvent production from the named solvent producing group. The list
need not be contained on a single line.
*FROM
This subkeyword indicates that all groups named in the list preceding *FROM
are to have their solvent recycling injection targets calculated using the solvent
production from the single group with the name
solvent_producing_group_name.
*FROM may occur on the same line as the last group in the list but need not.
It is valid for *FROM to be on a new line.
solvent_producing_group_name
solvent rates for recycling injection targets imposed upon the listed groups.
This group name may be but need not be on the same line as the *FROM
subkeyword.
DEFAULTS:

*SPRODGROUP, then by default that groups own solvent production rate is used to
compute solvent recycling targets imposed on the group.
CONDITIONS:
group must be defined before it can appear in the group list following *SPRODGROUP.
The fraction of solvent recycled to each group in the group_name_list must sum to no more
than 1.0. It is possible to reinject more solvent than is produced from the solvent producing
group if this is not ensured (The fraction of solvent recycled for each group is entered on its
*GCONI keyword after the *RECYCLE *SOLVENT sub-keyword).
User's Guide IMEX
EXPLANATION:
The imposition of a recycling rate target involves using the solvent production rates of one
group to determine the solvent injection rate in the same or another group. Please consult the
calculation of solvent re-injection targets. Further information is also available from the
manual pages for the other solvent recycling keywords, which are *SMKMAXR,
*SMKUPTO, and *SRECYMAXR
Example : Group1 and Group2 have their solvent recycling injection targets calculated using
the gas produced by Group3.
*SPRODGROUP 'Group1 Group2' *FROM Group3
User's Guide IMEX
Maximum Re-injection Rate for Group Solvent Recycling

(Optional)
*SRECYMAXR
PURPOSE:
*SRECYMAXR is used to specify a maximum rate at which solvent is re-injected under a

group recycling target. Solvent recycling injection targets are specified for groups using the
*GCONI keyword. The numbers entered under *SRECYMAXR have no effect unless a
group solvent recycling injection target is currently in force.
FORMAT:
*SRECYMAXR group_name_list max_recyc_rates

DEFINITIONS:
group_name_list
groups named in this list will acquire the maximum solvent recycling rates
specified after the list. The list need not be contained on a single line.
max_recyc_rates
A set of non-negative real values (m3/day | SCF/day | cm3/min). Zero is
accepted as a value but is interpreted as signifying that no maximum recycling
rate is set. If only a single value is supplied, that value will be applied to all
groups in the list. If more than one value is supplied, the number of maximum
recycling rates must equal the number of group names in the list and the first
rate will be applied to the first group, etc. These maximum rates set, or set an
upper limit for, the volumetric rate at which produced solvent is re-injected
under a group solvent recycling target. The values may be spread over more
than a single line.
DEFAULTS:
Optional keyword. If a groups name never appears in a group list following *SRECYMAXR,
then by default its maximum solvent recycling rate is set to zero, which is interpreted to mean
that there is no fixed maximum imposed upon the recycling injection rate.
CONDITIONS:
group must be defined before it can appear in the group list following *SRECYMAXR.
User's Guide IMEX
EXPLANATION:
For a full explanation of how the maximum recycling rates enter the calculation of a solvent
the group re-injecting the solvent, and not with the group producing it. The recycling
volumetric injection rate is set principally by the recycling fraction entered with the
*RECYCLE *SOLVENT line following *GCONI, but if a maximum has been imposed with
the *SRECYMAXR keyword, the re-injection target is not allowed to exceed this maximum.
Example : Group1 acquires a maximum recycling injection rate of 3.0D+07 SCF/day and
Group2 of 3.5D+07 SCF/day.
*SRECYMAXR
'Group1
3.0D+07
Group2'
3.5D+07
group names.
User's Guide IMEX
Monitored Group Constraints (Optional)
*GCONM
PURPOSE:
*GCONM is used to specify monitored group production constraints. Unlike the controls
specified under *GCONP and *GCONI, the quantities specified under *GCONM cannot be
assigned as group targets, and no action resulting in setting the violated value as a target is
possible.
FORMAT:
*GCONM 'group_name_1' 'group_name_2' ... 'group_name_n'

*GOR
*WCUT
*WGR
*MAXGAS
*MAXSTW
value
*MINOIL
*MINGAS
*MINBHF
value
(*STOP)
(*SHUTMOW)
(*SHUTMOL)
(*SHUTMOLDDOWN)
(*SHUTMOLUP)
(*SHUTMOWS)
(*SHUTALL)
(*STOP|*SHUTALL)
DEFINITIONS:

connected to each group have already been specified using the *WELL
keyword. The injection and production targets are met by satisfying either the
instantaneous potential or the guide rates specified using the keyword
*GUIDEP or *GUIDEI
*GOR
This subkeyword identifies a maximum gas-oil ratio (m3/m3 | scf/STB |
cm3/cm3) monitor for group production. The *STOP, *SHUTALL,
*SHUTMOW(S), *SHUTMOL, *SHUTMOLDOWN, and *SHUTMOLUP
actions are valid for this monitor.
*WCUT
This subkeyword identifies a maximum water-cut (fraction) monitor for group
production. The *STOP, *SHUTALL, *SHUTMOW(S), *SHUTMOL,
*WGR
This subkeyword identifies a maximum water-gas ratio (m3/m3 | STB/scf |
cm3/cm3) monitor for group production. The *STOP, *SHUTALL,
actions are valid for this monitor.
User's Guide IMEX
*MAXGAS
This subkeyword identifies a maximum surface gas rate (m3/day | scf/day |
cm3/min) monitor for group production. The *STOP, *SHUTALL,
actions are valid for this monitor. The most offending well or layer for this
constraint is deemed to be the one with the highest GOR rather than the one
with the highest gas rate.
*MAXSTW
This subkeyword identifies a maximum water rate (m3/day | STB/day |
cm3/min) monitor for group production. The *STOP, *SHUTALL,
actions are valid for this monitor. The most offending well or layer for this
constraint is deemed to be the one with the highest WCUT rather than the
one with the highest water rate.
*MINOIL
This subkeyword identifies a minimum oil rate (m3/day | STB/day | cm3/min)
monitor for group production. Only the *STOP and *SHUTALL actions are
allowed for this monitored constraint.
*MINGAS
This subkeyword identifies a minimum gas rate (m3/day | scf/day | cm3/min)
monitor for group production. Only the *STOP and *SHUTALL actions are
*MINBHF
This subkeyword identifies a minimum bottom-hole fluid rate (rm3/day |
rbbl/day | rcm3/min) monitor. Only the *STOP and *SHUTALL actions are
value
Constraint value -- units are given under *GOR, *WCUT, *WGR,
*MAXGAS, *MAXSTW, *MINOIL, *MINGAS and *MINBHF above.
*STOP
Action subkeyword indicating that if the constraint is violated then the
simulation should be stopped. This is the default action for all constraints if
no action is entered explicitly.
*SHUTMOL
This specifies that if a GOR, WCUT, WGR, MAXGAS, or MAXSTW
monitor is violated then the most offending layer (MOL) in the most offending
well (the one with the greatest GOR, WCUT, or WGR) should be shut in.
User's Guide IMEX
*SHUTMOLDOWN
This specifies that if a GOR, WCUT, WGR, MAXGAS, or MAXSTW monitor
is violated then the most offending layer (MOL) and the layers below it in the
most offending well (the one with the greatest GOR, WCUT, or WGR) should
be shut in.
*SHUTMOLUP
monitor is violated then the most offending layer (MOL) and the layers above
it in the most offending well (the one with the greatest GOR, WCUT, or WGR)
should be shut in.
*SHUTMOW
monitor is violated, then the most offending well (MOW the well with the
highest GOR, WCUT, or WGR) should be shut.
*SHUTMOWS
monitor is violated then a list of prioritized most offending wells (MOWS,
the ones with the greater GOR, WCUT, or WGR) should be shut.
*SHUTALL
This specifies that if the monitored constraint is violated for a group, then all
currently open wells in the group should be shut.
DEFAULTS:
Optional keyword. Default is no monitoring on groups. *STOP is the default action for all
constraints.
CONDITIONS:
be defined before it can be given any monitored values.
EXPLANATION:
*GCONM is used to specify monitored constraints. Unlike the constraints which are read
under *GCONP and *GCONI, the values entered under *GCONM cannot be applied as
targets. Note : The *AUTO-OPEN action of the *MINOIL and *MINGAS monitors is no
longer supported. Instead, automatic drilling of wells can be triggered when a group fails to
meet a rate target by using a *GAPPOR *AUTODRILL *ON line for the group and
specifying the well's status as *AUTODRILL. Please see the manual pages describing these
keywords for more information.
User's Guide IMEX
Example:
Automatic termination of simulation when the group oil production rate falls below a
prescribed limit:
*GCONM 'GROUP1'
*MINOIL 100.0 *STOP
User's Guide IMEX
Allow a Cycling Group to be Defined (Optional)

*GCONCYCLE_START, *GCONCYCLE_END
PURPOSE:
*GCONCYCLE_START allows the user to specify data for EOR/IOR processes which rely
on alternating production and/or injection cycles. Generally multiple cycles are required and
each cycle consists of a number of well defined parts. Examples include water alternating gas
injection (WAG) or cyclic steam injection. A number of criteria are available to stipulate the
duration of each cycle part.
FORMAT:
*GCONCYCLE_START group_names
*CYCSTREAMS stream_name_list (total of nstreams) required
*NPARTS nparts required
*MAXRATES (# of streams must be equal to nstreams) required
*TARGETTYPES (# of streams must be equal to nstreams) required
stream_name1 list_of_target_types (# of values must be equal to nparts)
stream_name2 list_of_target_types (# of values must be equal to nparts)
*MAXRATES (# of streams must be equal to nstreams) -- required
stream_name1 maximum_rate, imposed as a group target (# of values must be
equal to nparts)
stream_name2 maximum_rate, imposed as a group target (# of values must be
equal to nparts)
*VREPFR (# of streams must be equal to nstreams) -- required
stream_name1 voidage replacement fraction, imposed as a group target (# of
values must be equal to nparts)
stream_name2 voidage replacement fraction, imposed as a group target (# of
values must be equal to nparts)
*MAXCUMS -- optional
stream_name1 maximum cumulative injection/production (# of values must be
equal to nparts)
stream_name2 maximum cumulative injection/production (# of values must be
equal to nparts)
User's Guide IMEX
*TOTHEAT optional, can be specified for injection streams ONLY, for STARS
stream_name1 maximum cumulative enthalpy injection (# of values must be equal
to nparts)
stream_name2 maximum cumulative enthalpy injection (# of values must be equal
to nparts)
*MAXBHP optional, can be specified for injection streams ONLY
stream_name1 maximum BHP (# of values must be equal to nparts)
stream_name2 maximum BHP (# of values must be equal to nparts)
*MAXTIMES ntime1 ntime2 (# of values must be equal to nparts) optional
*MINTIMES ntime1 ntime2 (# of values must be equal to nparts) optional
*DTWCYC
dtime1 dtime2 (# of values must be equal to nparts) --optional
*NCYCLES ncycles optional

*MINBHP bhp1 bhp2 (# of values must be equal to nparts) optional
*MINQOIL minoil1 minoil2 (# of values must be equal to nparts) --optional
*DEPNDX depndx1 depndx2 ((# of values must be equal to nparts) --optional
*GCONCYCLE_END --required
DEFINITIONS:
*GCONCYCLE_START
This keyword indicates the start of keyword data input for group cycling
control. The complementary keyword *GCONCYCLE_END signals the end
of group cycling keyword data input.
*CYCSTREAMS
This keyword is used to specify the list of streams that participate in the
cycling process. A stream name can be one of the following: Production
stream, or PROD, or injection streams, gas GASI, water WATI or for
the IMEX simulator only solvent SOLI. For each stream a target type must
be entered for every cycle part using the keyword *TARGETTYPES as well
as a maximum rate for every cycle part using the keyword *MAXRATES or
a voidage replacement fraction (for injection streams) using the keyword
*VREPFR. The target type and maximum rate (or voidage replacement
fraction) provide sufficient information to impose a group target rate for each
cycle part for each stream. This keyword is required and is a prerequisite
before any other keyword other than *NPARTS can be specified.
User's Guide IMEX
*NPARTS
This keyword is used to specify the number of cycle parts required to fully
define one complete cycle. A particular cycle part will generally be based on
a period where there is either production or injection of a particular stream or
a period where there is neither production nor injection (soak). However the
user can choose to specify simultaneous injection of one or more streams
with production. This keyword is required and is a prerequisite before any
other keyword with the exception of *CYCSTREAMS can be specified.
*TARGETTYPES
This keyword is used to specify the particular phase associated with a given
stream and whether the entered maximum rates (with keyword
*MAXRATES) are to be interpreted as values at surface or reservoir
conditions. The table below shows the available selections for the target
types for each stream. This keyword is required.
STREAM
IMEX
GEM
STARS
Production (PROD)
Stock tank oil (STO)
Stock tank gas (STG)
Stock tank water (STW)
Bottom hole fluid (BHF)

includes oil, gas and water phase
production at reservoir or bottom
hole conditions as well as
production of the 4th phase, if
defined such as solvent or polymer

hole conditions

hole conditions
Stock tank liquid (STL)

includes production of oil and
water phases at surface or stock
tank conditions
Stock tank liquid (STL) includes

production of oil and water phases
at surface or stock tank conditions
as well as the production of the
intermediate liquid phase if defined
Stock tank liquid (STL)

includes production of oil and
water phases at surface or stock
tank conditions
Intermediate liquid phase (STI)

the intermediate liquid stream has to
be defined first as part of the surface
separation or separator data
Water injection (WATI) Stock tank water (STW) or
bottom hole water (BHW)
Stock tank water (STW) or

Stock tank water (STW) or

Gas injection (GASI)
Stock tank gas (STG) or

bottom hole gas (BHG)


Solvent injection (SOLI) Stock tank solvent (STS) or

bottom hole solvent (BHS)
User's Guide IMEX
*MAXRATES
This keyword is used to specify the maximum rate for each stream and for
each cycle part. The values entered will be interpreted based on the target types
specified using the keyword *TARGETTYPES. The specified maximum rates
are then imposed as group targets. If a voidage replacement group target needs
to be imposed, then enter a negative value for maximum rate for that stream
and a value between 0.0 and 1.0 for the voidage replacement fraction for the
stream using the keyword *VREPFR. For example to specify a voidage
replacement fraction for the water injection stream and a maximum rate group
target for the solvent stream:
*cycstreams 'WATI' SOLI
*maxrates
'WATI' -0.5 -0.5
'SOLI' 0.
250.
*vrepfr
'WATI' 1.0 0.0
'SOLI' -0.5 -0.5
A negative number specified with *MAXRATES indicates that the group target
will be based on voidage replacement fraction. Accordingly a voidage
replacement fraction needs to be specified using the keyword *VREPFR for that
stream. Similarly a negative value specified with *VREPFR indicates that a
value for maximum rate to be imposed as a group target is provided with the
keyword *MAXRATES for that stream. If maximum rate group targets are to be
imposed for all streams then the *VREPFR keyword is not required. Similarly if
voidage replacement fraction is to be specified for all streams then the
*MAXRATES keyword is not required. For the *MAXRATES or *VREPFR
keyword, data for all streams must be specified as shown in the above example.
Please note that group cycling control imposed targets will in general
override any previously specified group targets through data. If, at any time
group cycling control is turned off, any group targets that existed prior to
imposition of group cycling control are not as rule re-applied. However the
user can choose to re-specify any such targets using a special trigger at the
termination of group cycling control.
This keyword and/or *VREPFR is required. For example consider the
following fragment of user data:
cycstreams 'PROD' 'WATI'
nparts 3
targettypes
'WATI' 'STW' 'STW' 'STW'
'PROD' 'STL' 'STL' 'STL'
maxrates
'WATI' 1000. 0. 0.
'PROD' 0.
0. 1000.0
In this case cycle part 1 involves water injection, followed by a soak period
for cycle part 2 and then production for cycle part 3. For cycle part 1, the
group cycling manager will impose a target of 0.0 for STL for the production
User's Guide IMEX
stream and group injection target of 1000 units of STW for the water stream.
For cycle part 2, group production target will be STL value of 0.0 and group
injection target for the water stream will be 0.0 STW. For cycle part 3, the
group production target will be 1000 units of STL and the group injection
target will be 0.0 STW.
INPUT UNIT SYSTEM /
TARGET TYPE
SI & MODI SI UNITS

(IMEX, GEM, STARS)
FIELD UNITS
(IMEX, GEM, STARS)
LAB UNITS
(IMEX, STARS)
STO, STW, STL, STI, BHF, BHW
M3/day
Bbl/day
Cm3/min
WTG, STG, BHG, STS, BHS
M3/day
Ft3/day
Cm3/min
*VREPFR
This keyword is used to specify the voidage replacement fraction in lieu of a
maximum rate for an injection stream (WATI, GASI or SOLI), for each
cycle part. The group control target for that stream will be imposed such that
the specified fraction of total produced volume measured at reservoir
condition is replaced by injection. For a given stream, specify either a
maximum rate or a voidage replacement fraction. Since for keywords
*MAXRATES and *VREPFR values must be entered for all streams, if a
voidage replacement fraction is to be imposed for a given stream, then enter
a negative value for that stream with the keyword *MAXRATES. Please also
refer to the documentation for the *MAXRATES keyword for further details.
Please note that group cycling control imposed targets will in general
override any previously specified group targets through data. If, at any time
group cycling control is turned of, any group targets that existed prior to
imposition of group cycling control are not as rule re-applied. However the
user can choose re-specify any such targets using a special trigger at the
termination of group cycling control. This keyword is optional.
*MAXCUMS
This keyword is used to specify the maximum cumulative production or
injection for each stream and for each cycle part. The values entered will be
interpreted based on the target types specified using the keyword
*TARGETTYPES. The values specified are used to determine cycle part
duration. At the end of every time step the entered maximum cumulative will
be compared with the actual volume of fluid injected for the particular stream.
If the actual volume produced/injected is greater, then the current cycle is
deemed to be completed and the specified group targets for the next cycle in
sequence will be imposed starting with the next time step. If time step sizes are
large it is possible that actual production/injection may exceed considerably the
specified cumulative as cycling group control does not control time step size to
meet imposed cumulative exactly. This keyword is optional.
User's Guide IMEX
INPUT UNIT SYSTEM /

TARGET TYPE
SI & MODI SI UNITS

(IMEX, GEM, STARS)
FIELD UNITS
(IMEX, GEM, STARS)
LAB UNITS
(IMEX, STARS)
STO, STW, STL, STI, BHF, BHW
M3
Bbl
Cm3
WTG, STG, BHG, STS, BHS
M3
Ft3
Cm3
*TOTHEAT
This keyword is used to specify the maximum cumulative heat injection for
any gas or water injection streams for each cycle part. The values specified are
used to determine cycle part duration. At the end of every time step the entered
maximum heat injection will be compared with the actual heat injected for the
particular gas or water injection stream. If the actual heat injected is greater,
then the current cycle is deemed to be completed and the specified group
targets for the next cycle in sequence will be imposed starting with the next
time step. If time step sizes are large it is possible that actual heat injection
may exceed considerably the specified heat injection as cycling group control
does not control time step size to meet imposed targets exactly. This criterion
for determining cycle duration can only be specified for gas or water injection
streams and for the STARS simulator. This keyword is optional.
INPUT UNIT SYSTEM /
TARGET TYPE
STW, BHW, STG, BHG
SI & MODI SI UNITS

(STARS)
FIELD UNITS
(STARS)
LAB UNITS
(STARS)
Joule
Btu
Joule
*MAXBHP
This keyword is used to specify the maximum bottom hole pressure for any
injection streams specified for each cycle part. The values specified are used to
determine cycle part duration. At the end of every time step the entered
maximum bottom hole pressure will be compared with the actual bottom hole
pressure for all wells that are injecting fluid of the same type as the specified
stream. For example if the stream is WATI then the BHP of all water injection
wells belonging to the group will be tested against specified value. If the BHP of
any one well meets the criterion, then the current cycle part is deemed to be
completed and the specified group targets for the next cycle in sequence will be
imposed starting with the next time step. This criterion for determining cycle
duration can only be specified for injection streams. This keyword is optional.
INPUT UNIT SYSTEM
BHP
User's Guide IMEX
SI & MODI SI UNITS

(IMEX, GEM, STARS)
FIELD UNITS
(IMEX, GEM, STARS)
LAB UNITS
(IMEX, STARS)
kPa (SI)
kg/cm2 (MODSI)
psi
kPa
*MAXTIMES
This keyword is used to specify the maximum cycle part duration in terms of
time elapsed. The time step size will also be modified to ensure that in the
event that the cycle part duration is controlled by the specified times, then the
specified times are honored exactly. This keyword is optional.
INPUT UNIT SYSTEM
SI & MODI SI UNITS

(IMEX, GEM, STARS)
FIELD UNITS
(IMEX, GEM, STARS)
LAB UNITS
(IMEX, STARS)
days
days
minutes
MAX TIME
*MINTIMES
This keyword is used to specify the minimum cycle part duration. Therefore
the cycle part cannot end even if some other criterion for determining cycle
duration such as maximum cumulative is met until the minimum cycle
duration time is also met. The time step size is not modified when the cycle
part duration is controlled by the minimum specified time. This keyword is
optional. If the minimum cycle part time is not specified then by default a
value of zero is assumed.
INPUT UNIT SYSTEM
SI & MODI SI UNITS

(IMEX, GEM, STARS)
FIELD UNITS
(IMEX, GEM, STARS)
LAB UNITS
(IMEX, STARS)
days
days
minutes
MIN TIME
*DTWCYC
This keyword is used to specify the starting time step size for each cycle part.
If this keyword is missing the starting time step size will be selected according
to values entered with the *DTWELL and *DTMIN keywords. This keyword
is optional.
INPUT UNIT SYSTEM
SI & MODI SI UNITS

(IMEX, GEM, STARS)
FIELD UNITS
(IMEX, GEM, STARS)
LAB UNITS
(IMEX, STARS)
days
days
minutes
TIMESTEP SIZE
*NCYCLES
This keyword is used to specify the total number of cycles that must be
completed to turn off group cycling control. This keyword is optional. If the
number of cycles is not specified then group cycling control will continue to be
imposed throughout the simulation unless turned off using *GCONCYCLE
*OFF. The user also has the choice of specifying the starting cycle part using the
keyword *CYCPRT_START and the ending cycle part with *CYCPRT_END.
The actual number of total cycle parts completed will be at least equal to the
product of nparts*ncycles. For example if nparts = 3, cycprt_start = 2 and
cycprt_end = 1 and ncycles = 3, then the actual number of parts completed will
be 9:
2, 3
for cycle #1
1,2,3
for cycle #2
User's Guide IMEX
1,2,3
for cycle #3
for cycle #4
Once all the specified cycle parts are completed or the user manually turns of
group cycling control using *GCONCYCLE *OFF then any wells that are
open as a result of group cycling control and any wells that were shut-in as a
result of group cycling control will maintain their status, that is any wells
open will remain OPEN and any wells shut will remain shut. In addition any
group target or specified maximum constraints that became group targets will
NOT be reinstated as group targets at the end of group cycling control. The
user has been provided a means through the use of a special trigger to respecify any group targets at the end of group cycling control. For the above
example for cycle ending part number 1, as an illustration, involves
production, then all producers will remain open at the conclusion of group
cycling and any injection wells that were shut due to operation of group
cycling control in earlier cycle parts 2 and 3 will remain shut.
*MINBHP
This keyword is used to specify the minimum bottom hole pressure as a
criterion for switching cycle part for production wells. This criterion will
only be applied if stream PROD is specified and if the maximum rate for
the PROD stream is greater than zero for the cycle part. The values
specified are used to determine cycle part duration. At the end of every time
step the entered minimum bottom hole pressure will be compared with the
actual bottom hole pressure for all production wells belonging to the group.
If the BHP of any one well meets the criterion, then the current cycle part is
deemed to be completed and the specified group targets for the next cycle in
sequence will be imposed starting with the next time step. This criterion for
determining cycle duration can only be specified for the production stream.
This keyword is optional.
INPUT UNIT SYSTEM
MINIMUM BHP
SI & MODI SI UNITS

(IMEX, GEM, STARS)
FIELD UNITS
(IMEX, GEM, STARS)
LAB UNITS
(IMEX, STARS)
kPa (SI)
kg/cm2 (MODSI)
psi
kPa
*DEPNDX
This keyword is used to specify the minimum depletion index as the criterion
for switching cycle part for production wells. The depletion index is the ratio
of total liquid production for the current cycle divided by total water or steam
injected in a previous injection cycle part. This criterion will only be applied
if stream PROD is specified and if the maximum rate for the PROD
stream is greater than zero for the current cycle part. This criterion is only
meaningful is a production cycle part follows an injection cycle part where
water or steam is injected. The values specified are used to determine cycle
part duration. This criterion for determining cycle duration can only be
specified for the production stream. This keyword is optional.
User's Guide IMEX
*MINQOIL
This keyword is used to specify the minimum oil rate as a criterion for
switching cycle part for production wells. This criterion will only be applied
if stream PROD is specified and if the maximum rate for the PROD
stream is greater than zero for the cycle part. The values specified are used to
minimum oil rate for the group will be compared with the actual oil rate for
the group. For the criterion to be satisfied the oil rate must first transit from a
value higher than the specified minimum to a value less than the specified
minimum. This criterion for determining cycle duration can only be specified
for the production stream. This keyword is optional.
INPUT UNIT SYSTEM
MINIMUM OIL RATE measured at
surface conditions
SI & MODI SI UNITS

(IMEX, GEM, STARS)
FIELD UNITS
(IMEX, GEM, STARS)
LAB UNITS
(IMEX, STARS)
M3/day
Bbl/d
Cm3/min
*MININJ
This keyword is used to specify the minimum injection rate as a criterion for
switching cycle part for injection streams. This criterion will only be applied
if an injection stream is specified and if the maximum rate for the injection
stream is greater than zero for the cycle part. The values specified are used to
minimum injection rate for the group will be compared with the actual oil
rate for the group. For the criterion to be satisfied the injection rate must first
transit from a value higher than the specified minimum to a value less than
the specified minimum. This criterion for determining cycle duration can
only be specified for injection streams. This keyword is optional.
INPUT UNIT SYSTEM
SI & MODI SI UNITS

(IMEX, GEM, STARS)
FIELD UNITS
(IMEX, GEM, STARS)
LAB UNITS
(IMEX, STARS)
WATI (STW/BHW)
M3/day
Bbl/d
Cm3/min
GASI (STG/BHG)
M3/day
Ft3/day
Cm3/min
SOLI (STS/BHS) IMEX only
M3/day
Ft3/day
Cm3/min
DEFAULTS:
There default is to continue applying values originally specified with *GCONCYCLE_START.

CONDITIONS:
A group must be defined before it can be additionally declared as a cycling group using the
keyword *GCONCYCLE_START. Group hierarchy is specified using the keyword
*GROUP. It is important to specify all potential wells that can participate in meeting group
cycling targets as initially OPEN. The group cycling manager will not open a well if it is
specified as SHUT through data. Likewise any wells that are shut in due to an action
specified with a MONITOR will not be reopened to meet a group cycling target. Similarly
any wells on the drill queue with the status of *AUTODRILL will not be opened by the group
cycling manager. All wells of a given type associated with a specific stream that have the
status of OPEN will automatically be included in meeting cyclic group targets for that stream.
User's Guide IMEX
For example for stream WATI, water injection, all wells that inject water that are declared
as OPEN will be included in meeting the target specified for the WATI stream.
EXPLANATION:
Once all specified cycle parts are completed or alternatively a group previously under cycling
control is taken off cycling control by specification of *GCONCYCLE *OFF, then all wells
involved in group cyclic control will retain the status in effect at the end of the selected cycle
part. In addition any group targets imposed by the group cycling manager will remain in
effect as of the last cycle part. The user has the option of changing the status of any of the
wells for a given group or modifying/appending to the existing group target constraints by
specifying the actions through data using a trigger specifically provided to mark the event
corresponding to the lifting of group cyclic control. The syntax of the trigger is:
*TRIGGER trigger_name *ON_CYCGRP group_name CYC-END > 0
Actions
*END_TRIGGER
The trigger condition will be satisfied when a group previously under group cyclic control is
removed from group cyclic control. This can happen in a number of ways:
i)
The group was under cyclic control and the maximum number of cycle parts as
stipulated by *CYCPRT_START, *CYCPRT_END and nparts*ncycles have been
completed
ii) The group was under group cyclic control and *GCONCYCLE *OFF is specified
iii) The group is redefined with another *GCONCYCLE_START keyword
The user can choose to turn off or suspend group cycling control at any time and turn on
group cyclic control back on at any time. If group cycling control is turned back on then
group cyclic control will resume with the cycle part following the cycle part at which group
cyclic control was turned off regardless of whether the criteria for completing or switching to
the next cycle part was fulfilled or not at the time that group cyclic control was turned off.
The cycle part number is saved or preserved when group cycling control is turned off.
Consider the following example with ncycles = 4, nparts = 2
Time = 0.0
*GCONCYCLE wfgroup1 *ON
Time T1
GCONCYCLE wfgroup1 *OFF ** at this time cyclic group control
is in the middle of total cycle part 5 (cycle #3, part #1)
Time T2
*GCONCYCLE wfgroup1 *ON
From time T1 to time T2 group cyclic control is suspended. Group targets and well status in
effect at the start of total cycle part #5 are imposed.
From time T2 group cyclic control resumes with total cycle part #6 (cycle #3, part #2)
regardless of whether the criteria for switching from part 5 to part 6 were met or not.
The user can modify any of the criteria for determining cycle duration at any time using the
keyword block GCONCYR_START and GCONCYCLE_END, provided that the cyclic
group is defined first using the keyword *GCONCYCLE_START. However one cannot
modify the number of cycle parts or streams with GCONCYCR_START. Any changes
introduced with *GCONCYCR_START take effect immediately.
User's Guide IMEX
The user can specify the starting and/or ending cycle part with the keywords
*CYCPRT_START and CYCPRT_END as well as the total number of cycles with
*NCYCLES. The actual number of total cycle parts completed will be at least equal to the
product of nparts*ncycles. The default for CYCPRT_START is 1 and CYCPRT_END is
nparts or the last part. The default for ncycles is infinity. The cycles will start with the
specified part and end after completing the cycle part specified with CYCPRT_END. The
formula for the actual number of cycle parts completed is
Ncycles * nparts + (cycprt_end cycprt_start + 1)
Example 1
Ncycles =13; nparts = 2, cycprt_start = 2, cycprt_end = 1
Ncycles *nparts = 13*2 = 26
Start with cycle #1, part #2 (ipart = 2)
End with cycle #14, part #1 (ipart = 27)
Actual # of parts completed (27-2+1) = 26
Example 2
Ncycles = 4, nparts = 3, cycprt_start = 3, cycprt_end = 2
Ncycles *nparts = 4*3 = 12
Start with cycle #1, part #3 (ipart = 3)
Complete 3 additional full cycles (3*3 = 9 cycle parts
ipart = 12)
Cycle #4, complete part 1&2 ipart = 14
Actual number of cycle parts completed 14-3+1 = 12
User's Guide IMEX
Allow Cycling Group Data to be Revised (Optional)

*GCONCYCR_START, *GCONCYCR_END
PURPOSE:

FORMAT:
*GCONCYCR_START group_name1 group_name2

(Any of *TARGETTYPES, *MAXCUM, *MAXTIMES, *NCYCLES,
MAXBHP, *MININJ, *MINBHP, *MINOIL, *DEPNDX, *TOTHEAT
*GCONCYCR_END
DEFAULTS:
The default is to continue applying values originally specified with *GCONCYCLE_START.

CONDITIONS:
This block of keywords can only be specified for a cycling group if that group is previously
defined using the keyword *GCONCYCLE_START. Any or all of the group cycle part
switching criteria specified earlier with *GCONCYCLE_START can be revised or new
criteria added. However this keyword block cannot be used to change the number of cycle
parts (keyword *NPARTS) or the number and type of streams (keyword CYCSTREAMS).
Number of parts or streams can only be changed by redefining the group using
*GCONCYCLE_START.
EXPLANATION:
See keyword *GCONCYCLE_START

Once a cycling group is defined using *GCONCYCLE_START this keyword block can be
used to modify some of the data used to test for cycle part duration. It is also possible to add
new criteria to determine cycle part end that were not previously specified with
*GCONCYCLE_START. The data entered will take effect immediately. Consider the
following example:
Time 0.0
GCONCYCLE_START 'cyc_group1'
nparts 3
ncycles 40
targettypes
maxrates
'WATI' 1000. 0. 0.
'PROD' 0.
0. 1000.0
User's Guide IMEX
maxcums
'PROD' 0. 0. 600000.0
tothet
'WATI' 5.0E+09 0. 0.
maxbhp
'WATI' 1100.0 0.0 0.0
maxtimes 10.0 3.0 348.0
dtwcyc 0.02 1.0E+5 1.0
GCONCYCLE_END
time 365.0 ** modify BHP criterion
GCONCYCR_START 'cyc_group1'
*maxbhp
'WATI' 400.0 0.0 0.0
GCONCYCLE_END
User's Guide IMEX
Allow Cycling Group Data to be Revised (Optional)

*GCONCYCLE
PURPOSE:
FORMAT:
*GCONCYCLE group_name1 group_name2 *ON|OFF

DEFAULTS:
Once data is specified is specified with *GCONCYCLE_START, the default is

*GCONCYCLE *ON.
CONDITIONS:
This block of keywords can only be specified for a cycling group if that group is previously
defined using the keyword *GCONCYCLE_START. Group cycling control can be specified
at any time once the group is defined using *GCONCYCLE_START.
EXPLANATION:

Once a cycling group is defined using *GCONCYCLE_START this keyword can be used to
turn off group cycling or to impose group cycling once more if group cycling was previously
turned off. The desired state of group cycling control specified with this keyword will take
effect immediately. If GCONCYCLE *OFF is specified for a group previously under group
control, then group cycling control will be terminated immediately and the group target(s)
and well state existing when GCONCYCLE *OFF is specified will continue to be in effect. If
actions are specified as part of the special trigger provided to monitor the event of group
cycling control being turned off, then that trigger condition will be satisfied and the stipulated
actions will take effect at the top of the next time step. If GCONCYCLE *ON is specified at a
time when group control is turned off then group cycling control will resume with the next
cycle part following the part when group cycling was turned off. The count of number of
cycle parts completed is preserved between applications of GCONCYCLE *ON/*OFF.
Consider the following example:
Time 0.0
nparts 3
ncycles 40
targettypes
maxrates
'WATI' 1000. 0. 0.
'PROD' 0.
0. 1000.0
User's Guide IMEX
maxcums
'PROD' 0. 0. 600000.0
tothet
'WATI' 5.0E+09 0. 0.
maxbhp
'WATI' 1100.0 0.0 0.0
maxtimes 10.0 3.0 348.0
dtwcyc 0.02 1.0E+5 1.0
GCONCYCLE_END
Time 100
*GCONCYCLE cyc_group1 *OFF ** was on part 3 when turned of
GCONCYCLE 'cyc_group1' *ON ** will resume with cycle part 1
User's Guide IMEX
Select Cycling Group Cycle Part to Start and End Cycling

Group Control (Optional)
*CYCPRT_START, *CYCPRT_END
PURPOSE:
FORMAT:
*CYCPRT_START group_name1 group_name2 cycle_part_start

*CYCPRT_END group_name1 group_name2 cycle_part_end
DEFAULTS:
Once data is specified is specified with *GCONCYCLE_START, the default for group cycle
starting part is 1 and for group cycle ending part is the value specified with *NPARTS, that is
the last cycle part.
CONDITIONS:
These can only be specified for a cycling group if that group is previously defined using the
keyword *GCONCYCLE_START.
EXPLANATION:

Once a cycling group is defined using *GCONCYCLE_START these keywords can be used
to select the starting and ending cycle parts. The values specified with these keywords will
take effect immediately. If cycling group control is in effect when *CYCPRT_START is
specified then the value specified with *CYCPRT_START will be ignored. If
*CYCPRT_START is specified following *GCONCYLE *ON then starting cycle part will
be the value specified with *CYCPRT_START plus 1. For example
Time 0.0
nparts 3
ncycles 40
targettypes
maxrates
'WATI' 1000. 0. 0.
'PROD' 0.
0. 1000.0
User's Guide IMEX
maxcums
'PROD' 0. 0. 600000.0
tothet
'WATI' 5.0E+09 0. 0.
maxbhp
'WATI' 1100.0 0.0 0.0
maxtimes 10.0 3.0 348.0
dtwcyc 0.02 1.0E+5 1.0
GCONCYCLE_END
Time 100
*GCONCYCLE cyc_group1 *OFF ** was on part 3 when turned of
*GCONCYCLE 'cyc_group1' *ON ** will resume with cycle part 1
*CYCPRT_START 'cyc_group1' 1 ** will now resume with cycle part 2
User's Guide IMEX
Defining Group Production or Injection as Going Through a

Manifold (Optional)
*MANIFOLD
PURPOSE:
*MANIFOLD is used to specify that group production or injection of a stream is treated as

going through a manifold. In most applications this refers to an undersea manifold, but the
defining aspect of the manifold treatment is that all wells of one type (all producers, or all
water injectors, etc.) in the group have the same pressure at the upper end of the tubing. A
surface, or compressor, pressure may be defined for a manifold group. Manifold groups may
operate under manifold and surface pressure constraints, but even when running on rate
constraints the enforcement of equal manifold pressures upon the wells alters the groups
performance relative to a non-manifold group.
Only a group having wells directly attached to it can have *MANIFOLD specified. Groups
having other groups attached cannot have *MANIFOLD specified.
FORMAT:

DEFINITIONS:
PROD
Indicates that group production should be treated as going through a
manifold.
GASI
Indicates that group gas injection should be treated as going through a
manifold.
WATI
Indicates that group water injection should be treated as going through a
manifold.
SOLI
Indicates that group solvent injection should be treated as going through a
manifold.
group_list
Names of groups to which the current manifold specification is to apply.
Each group name is up to 16 characters long and enclosed in single quotes.
All names in group_list must have been previously defined using the
*GROUP keyword.
*ON
Specifies that the manifold treatment is to be imposed.
User's Guide IMEX
*OFF
Specifies that the manifold treatment is not to be imposed (or is to be
removed).
DEFAULTS:
Optional keyword. Default for all groups and target types is *OFF.
CONDITIONS:
This keyword must be located in the Well and Recurrent Data keyword group. The spacing
of the subkeywords and group names is not important they can all be on one line, or lines
may be broken arbitrarily. For example, *ON or *OFF may be on the same line as are the
group names. Only one of the four possible target types may be specified in any one
occurrence of the *MANIFOLD keyword. All wells operating through manifolds must have
a means specified for computing tubing-head pressure, as the tubing-head pressure for these
wells is interpreted as the manifold pressure. *MANIFOLD can only be specified for groups
having wells attached directly to them.
EXAMPLES:
*MANIFOLD
*PROD Group-1 *ON
This specifies that the production for group Group-1 is to be treated as going through a
manifold. All producers attached to Group-1 must have a means specified for computing
tubing head pressure through use of the *PWELLBORE keyword.
*MANIFOLD *GASI Group-1 Group-2
Group3 *ON
This specifies that gas injection for Group-1, Group-2, and Group-3 should be treated as
going through a manifold. All gas injectors in these groups must have a means specified for
computing tubing head pressure through use of the *IWELLBORE keyword.
User's Guide IMEX
Pressure Constraint Translation for Manifolds (Optional)

*PCON-MANIFOLD
PURPOSE:
*PCON-MANIFOLD applies only to group streams treated as going through a manifold (see
the pages for the *MANIFOLD keyword). When *PCON-MANIFOLD is specified, if the
group is operating on a manifold or surface pressure constraint, that constraint is translated to
a tubing-head pressure constraint and equal tubing-head pressure constraints are assigned to
all wells attached to the manifold. Otherwise, the wells are assigned rates constraints which
give the correct manifold or surface pressure at the start of the timestep. When *PCONMANIFOLD is specified, group pressure constraints are usually honored more accurately,
and higher-level rate constraints less accurately, than if it is not specified.
FORMAT:
*PCON-MANIFOLD (*PROD | *GASI | *WATI | *SOLI)

group_list (*ON | *OFF)
DEFINITIONS:
PROD
Indicates that pressure translation should be applied for pressure constraints
upon the production manifolds of the listed groups.
GASI
Indicates that pressure translation should be applied for manifold group gas
injection for the listed groups.
WATI
Indicates that pressure translation should be applied for manifold group water
injection for the listed groups.
SOLI
Indicates that pressure translation should be applied to manifold group
solvent injection for the listed groups.
group_list
Names of groups to which the current pressure translation specification is to
apply. Each group name is up to 16 characters long and enclosed in single
quotes. All names in group_list must have been previously defined using the
*GROUP keyword.
*ON
Specifies that pressure translation is to be imposed.
*OFF
Specifies that pressure translation is not to be imposed (or is to be removed).
User's Guide IMEX
DEFAULTS:
Optional keyword. Default for all groups and target types is *OFF.
CONDITIONS:
This keyword must be located in the Well and Recurrent Data keyword group. *PCONMANIFOLD can have an effect only if the specified group streams have been designated to
be treated as going through a manifold through use of the *MANIFOLD keyword. It is not
considered an error to specify *PCON-MANIFOLD when *MANIFOLD has not been
specified, but pressure translation would then have no effect because a non-manifold group
can never operate on a pressure constraint. The spacing of the subkeywords and group names
is not important they can all be on one line, or lines may be broken arbitrarily. For
example, the group names may be on the same line as the specification of stream type. Only
one of the four possible target types may be specified in any one occurrence of the *PCONMANIFOLD keyword. When a manifold for which *PCON-MANIFOLD is specified is
running on a rate constraint rather than upon a pressure constraint, *PCON-MANIFOLD has
no effect.
EXAMPLES:
*PCON-MANIFOLD
*PROD Group-1 *ON
This specifies that pressure constraints for the production manifold for group Group-1 are to
be translated into well tubing head pressure (manifold pressure) constraints.
*PCON-MANIFOLD *GASI Group-1 Group-2
Group3 *ON
This specifies that pressure constraints for the gas injection manifolds for Group-1, Group2, and Group-3 should be translated into well tubing-head pressure (manifold pressure)
constraints.
User's Guide IMEX
Specification of Hydraulics Tables for Calculating Pressure

Difference Between Manifold and Surface (Optional)
*GPTABLE
PURPOSE:
*GPTABLE (Group Pressure Table) applies only to group targets treated as going through a
manifold (see the pages for the *MANIFOLD keyword). *GPTABLE specifies the hydraulics
table number to be used to compute the pressure drop between manifold and surface, for
production and injection manifolds. The hydraulics table can be entered using the *PTUBE1
keyword for production manifolds or *ITUBE1 for injection manifolds.
FORMAT:

table_numbers
DEFINITIONS:
PROD
Indicates that the pressure table numbers are being specified for production
manifolds.
GASI
Indicates that the pressure table numbers are being specified for gas injection
manifolds.
WATI
Indicates that the pressure table numbers are being specified for water
injection manifolds.
SOLI
Indicates that the pressure table numbers are being specified for solvent
injection manifolds.
group_list
Names of groups to which the current hydraulic table number specification is
to apply. Each group name is up to 16 characters long and enclosed in single
quotes. All names in group_list must have been previously defined using the
*GROUP keyword.
User's Guide IMEX
table_numbers
List of positive integers giving the hydraulics table numbers. The list may
consist either of a single integer, in which case all of the listed groups are
assigned this single hydraulics table number, or table_list must contain
exactly as many entries as does the group_list, in which case the table
numbers are assigned to the groups in order.
DEFAULTS:
Optional keyword. Default for all groups is no assigned hydraulics table number; in which
case the surface (compressor) pressure is reported to be equal to the manifold pressure and no
surface pressure constraint can be specified.
CONDITIONS:
This keyword must be located in the Well and Recurrent Data keyword group Only one of
the four possible target types may be specified in any one occurrence of the *GPTABLE
keyword. The spacing of the subkeywords, group names, and table numbers is not important
they can all be on one line, or lines may be broken arbitrarily.
EXAMPLES:
*GPTABLE
1
*PROD Group-1 Group-2
This specifies that pressure drops between the production manifolds and surface for Group1 and Group-2 will be calculated using the first hydraulics table.
*GPTABLE *GASI Group-1 Group-2
Group-3 1 7 4
This specifies that the gas injection manifolds of Group-1, Group-2, and Group-3 should
have pressure drops between compressor and manifold calculated using hydraulics tables 1,
7, and 4 respectively.
User's Guide IMEX
Manifold Depth for Calculating Pressure Difference Between

Manifold and Surface (Optional)
*GPHYDDEP
PURPOSE:
*GPHYDDEP applies only to group targets treated as going through a manifold (see the
pages for the *MANIFOLD keyword). *GPHYDDEP specifies the depth used to scale the
pressure drop linearly relative to the depth specified with the particular hydraulics table. The
hydraulics table to be used is identified using the *GPTABLE keyword.
FORMAT:
*GPHYDDEP (*PROD | *GASI | *WATI | *SOLI) group_list

depths
DEFINITIONS:
PROD
Indicates that the depths are being specified for production manifolds.
GASI
Indicates that the depths are being specified for gas injection manifolds.
WATI
Indicates that the depths are being specified for water injection manifolds.
SOLI
Indicates that the depths are being specified for solvent injection manifolds.
group_list
Names of groups to which the current depth specification is to apply. Each
group name is up to 16 characters long and enclosed in single quotes. All
names in group_list must have been previously defined using the *GROUP
keyword.
depths
List of real numbers (m | ft | cm | m) giving the manifold depths. The list may
consist either of a single number, in which case all of the manifolds for the
listed groups are assigned this single depth, or depths must contain exactly as
many entries as does the group_list, in which case the depths are assigned to
the groups in order.
DEFAULTS:
Optional keyword. Default for all manifolds is to use the depth specified with the hydraulics
table used for the pressure drop calculation.
User's Guide IMEX
CONDITIONS:
the four possible target types may be specified in any one occurrence of the *GPHYDDEP
keyword. The spacing of the subkeywords , group names, and depth values is not important
EXAMPLES:
*GPHYDDEP *PROD Group-1 Group-2
3512.
This specifies that the production manifolds for Group-1 and Group-2 will taken to be
3512 units below the surface when performing the pressure drop calculation between
manifold and surface.
*GPHYDDEP *GASI Group-1 Group-2
Group-3 4564. 3256. 3987.
This specifies that the gas injection manifolds of Group-1, Group-2, and Group-3 are
assigned depths of 4564, 3256, and 3987 units respectively to be used in scaling tabular
calculations of the pressure drop between manifold and compressor.
User's Guide IMEX
Group Artificial Lift Quantity Value (Optional)
*GROUPALQ
PURPOSE:
*GROUPALQ applies only to group targets treated as going through a manifold (see the
pages for the *MANIFOLD keyword) and for which a manifold-to-surface hydraulics table
has been specified using the *GPTABLE keyword. It introduces the value which will be
used for the ALQ quantity in interpolations in a hydraulics table entered using the *PTUBE1
table format. For some choices of the ALQ variable, the variable value is arbitrary and can be
specified freely in the data. In these situations the units of the ALQ quantity are not
important so long as the assigned values and the tabular values are consistent.
FORMAT:

alq_values
DEFINITIONS:
PROD
Indicates that the ALQ values are being specified for production manifolds.
GASI
Indicates that the ALQ values are being specified for gas injection manifolds.
WATI
Indicates that the ALQ values are being specified water injection manifolds.
SOLI
Indicates that the ALQ values are being specified for solvent injection
manifolds.
group_list
Names of groups to which the current ALQ specification is to apply. Each
group name is up to 16 characters long and enclosed in single quotes. All
names in group_list must have been previously defined using the *GROUP
keyword.
alq_values
List of real numbers giving the ALQ values. The list may consist either of a
single number, in which case all of the manifolds for the listed groups are
assigned this single value, or alq_values must contain exactly as many entries as
does the group_list, in which case the ALQ values are assigned to the groups in
order. The units are arbitrary but must correspond to the tabular units.
User's Guide IMEX
DEFAULTS:
Optional keyword. Default for all manifolds is an ALQ value of zero.

CONDITIONS:
the four possible target types may be specified in any one occurrence of the *GROUPALQ
keyword. The spacing of the subkeywords, group names, and ALQ values is not important
EXAMPLES:
*GROUPALQ *PROD Group-1 Group-2
1.
This assigns ALQ values of 1 for the production manifolds of Group-1 and Group-2.
*GROUPALQ *GASI Group-1 Group-2
Group-3 1. 0.5 0.3
This specifies that the gas injection manifolds of Group-1, Group-2, and Group-3 are
assigned ALQ values of 1., 0.5, and 0.3 respectively to be used in tabular calculations of the
pressure drop between manifold and compressor.
User's Guide IMEX
Well Artificial Lift Quantity Value (Optional)
*WELLALQ
PURPOSE:
*WELLALQ specifies an artificial lift quantity value to be used in the tabular tubing head
pressure calculation for producers. It introduces the value which will be used for the ALQ
quantity in interpolations in a hydraulics table entered using the *PTUBE1 table format. For
some choices of the ALQ variable, the variable value is arbitrary and can be specified freely
in the data. In these situations the units of the ALQ quantity are not important so long as the
assigned values and the tabular values are consistent.
FORMAT:
*WELLALQ well_list
alq_values
DEFINITIONS:
well_list
Names or numbers of wells to which the current ALQ specification is to
apply. Each well name is up to 40 characters long and enclosed in single
quotes. All wells included in well_list must have been previously defined
using the *WELL keyword. Limited wildcarding is available for a list of
well names; please see the manual page for the *SHUTIN keyword for an
explanation of the wildcard facility.
alq_values
List of real numbers giving the ALQ values. The list may consist either of a
single number, in which case all of the wells in well_list are assigned this
single value, or alq_values must contain exactly as many entries as does the
well_list, in which case the ALQ values are assigned to the wells in order.
The units are arbitrary but must correspond to the tabular units.
DEFAULTS:
Optional keyword. Default for all wells is an ALQ value of zero.

CONDITIONS:
This keyword must be located in the Well and Recurrent Data keyword group. The spacing of
the well names and ALQ values is not important they can all be on one line, or lines may be
broken arbitrarily.
User's Guide IMEX
EXAMPLES:
*WELLALQ Well-1 Well-2
1.
This assigns ALQ values of 1 for wells Well-1 and Well-2, in arbitrary units.
*WELLALQ Well-1 Well-2
Well-3 1097. 976.5 1137.3
This assigns ALQ values of 1097, 976.5, and 1137.3, in arbitrary units, to Well-1, Well-2,
and Well-3 respectively.
User's Guide IMEX
Priority List for Automatic Drilling of Wells (Optional)
*DRILLQ
PURPOSE:
*DRILLQ allows the specification of the order in which wells must be drilled automatically.
FORMAT:
*DRILLQ
*DRILLQ
*DRILLQ
*IPP
*IIP
well_list
DEFINITIONS:
*IPP
Specifies that the instantaneous production potential of wells with status
*AUTODRILL will be used to determine the order in which these wells will
be opened. *IPP applies to producers only.
*IIP
Specifies that the instantaneous injection potential of injection wells with
status *AUTODRILL will be used to determine the order in which these
wells will be opened. *IIP applies to injectors only.
well_list
A list of 'well_names' or well_numbers that give the priority in which wells
will be drilled. The first well on the list will be drilled first and so on.
Producers and injectors may be specified in the same list.
Note: wildcards may be used as in the 'well_names' string as follows:
'?ell1').
DEFAULTS:
Optional keywords. Default is to use the instantaneous injection / production potential to

determine drilling priority.
CONDITIONS:
*DRILLQ must be located in the Well and Recurrent Data keyword group, and must follow
the *GCONI or *GCONP keywords.
User's Guide IMEX
EXPLANATION:
This optional keyword is used to specify the order in which wells that have *AUTODRILL as
their status will be drilled. One may either use the instantaneous injection / production
potentials or supply a list to give a drilling priority order.
The opening of *AUTODRILL wells is triggered during the apportionment of group rate
targets if a group fails to be able to meet its rate target and if the group has had the
*AUTODRILL feature turned on through the pair of lines
*GAPPOR 'group_name'
*AUTODRILL *ON
Please consult the documentation for the *GAPPOR keyword for more information.
If a well_list is given then wells will be opened in the order in which they appear on the list.
If not all wells are on the list then those on the list will be opened first and the rest will be
opened according to their IIP or IPP.
Example:
*GCONP 'Group1'
*MAX *STO 1000.0
*GUIDEP *STO 'well-1' 'well-2' 'well-3'
200.0
150.0
650.0
*DRILLQ 'well-4' 'well-5' 'well-6'
User's Guide IMEX
Group Apportionment Options (Optional)
*GAPPOR
PURPOSE:
*GAPPOR introduces subkeywords which control how the apportionment of group target
among contributing wells or subgroups is to be done. Currently *AUTODRILL is the only
subkeyword supported.
FORMAT:
FORMAT:
*GAPPOR
*AUTODRILL
'group_name_1' 'group_name_n'
(*ON | *OFF| stream identifier)
DEFINITIONS:
'group_name_1', , group_name_n
Are the groups to which the following constraints , apply. The wells that are
connected to each group have already been specified using the *WELL
keyword. The injection and production targets are apportioned according to
either the instantaneous potential or the guide rates specified using the
keywords *GUIDEP or *GUIDEI
*AUTODRILL (*ON | *OFF | stream identifier)
The *AUTODRILL apportionment option is turned on or off. *ON specifies
that if one of the listed groups is attempting to meet a rate target, and the
group apportionment routine predicts that the contributing wells have
insufficient injection or production potential to meet this target, any
potentially contributing wells which are shut but have *AUTODRILL status
may be opened automatically to meet the target. *OFF disables the option.
The stream identifier takes the form of *STO, *STG, *STW, *STS, *STL
and *BHF. They are the available group target streams as defined by group
control keyword *GCONP under *TARGET (refer to the explanation therein
for details). The stream identifier has the same effect as *ON if the group is
currently targeted to that stream; and has the same effect as *OFF if the
group is running on a different targeted stream.
DEFAULTS:
Optional keyword. Default is no *AUTODRILL action in the apportionment. If

*AUTODRILL appears with no following *ON or *OFF directive, *ON is assumed.
CONDITIONS:
be defined before it can be assigned options under *GAPPOR.
User's Guide IMEX
EXPLANATION:
*GAPPOR is used to specify options used in the course of apportioning a group's rate target
among the contributing wells.
Example:
The following line directs that if group GROUP1 is unable to meet its current rate target (if
there is one), any wells connected directly or indirectly to GROUP1 which have
AUTODRILL status should be opened in order until either no more AUTODRILL wells are
left or until the target can be met. The order in which the wells are to be opened is
determined by the *DRILLQ keyword; the default is to open the wells in decreasing order of
instantaneous injection/production potential.
**GAPPOR 'GROUP1' *AUTODRILL *ON
User's Guide IMEX
Apportionment Method for Meeting Group Targets (Optional)

*APPOR-METHOD
PURPOSE:
*APPOR-METHOD specifies the apportionment method to distribute the production or

injection rates among contributing wells or groups for meeting the group target(s) defined by
*TARGET under *GCONP or *GCONI.
FORMAT:
*APPOR-METHOD
(*PROD | *GASI | *WATI | *SOLI) 'group_names'
(*IP | *GUIDE | *INGUIDE | *PRIOR)
DEFINITIONS:
*PROD
Indicates that group production targets should be apportioned with the
specified method.
*GASI
Indicates that group gas injection targets should be apportioned with the
specified method.
*WATI
Indicates that group water injection targets should be apportioned with the
specified method.
*SOLI
Indicates that group solvent injection targets should be apportioned with the
specified method (valid for IMEX only).
'group_names'
A list of 'group_names' to which the apportionment method is applied. No
wildcard characters may be used in the group names.
*IP
This sub-keyword specifies the use of instantaneous potential (IP), which is
computed internally by the simulator, to determine the apportionment of
production or injection targets among wells. This is default. Each
contributing well is assigned a rate in proportion to its maximum rate for the
surface stream for which the target is referenced. This has the effect that
either the target can be met with all contributors being assigned rates in the
same fraction of their maximum rate, or the target cannot be met and all
contributors are assigned exactly their maximum rates.
User's Guide IMEX
*GUIDE
This sub-keyword specifies the use of user-supplied guide rates to determine
the apportionment of production or injection targets among contributing
wells or groups. Each contributing well or group receives a fraction of the
target in proportion to its guide rate. If the assignment violates its maximum
rate under the most restrictive constraint, such a well or group will be left out
the apportionment with rate being cut back to the maximum rate. The
reminder of the target is still distributed proportional to the guide rates
among the remaining wells or groups whose assignments do not exceed the
maximum rates. The guide rates of a proper surface reference stream are
specified via keyword *GUIDEP for production or *GUIDEI for injection
(refer to the explanation therein for details).
*INGUIDE
This sub-keyword specifies the use of internal guide rates to determine the
apportionment of production or injection targets among contributing wells.
The idea is very similar to the above *GUIDE, except that the guide rates are
now generated internally by the simulator according to the user-supplied
priority formula defined by keyword *PRIOR-FORM (refer to the
explanation therein for details). The default is to use the instantaneous
productivity or injectivity potential under a bottom hole pressure constraint:
1 atm for production and 3500 atm for injection, if the priority formula does
not present. Unlike the user-supplied guide rate (*GUIDE), the internallygenerated guide rates are time dependent. However, any well violating the
constraint is still cut back to its maximum rate as in *GUIDE.
*PRIOR
This sub-keyword specifies the use of priority ranking to determine the
apportionment of group targets among contributing wells. Wells are ranked
and opened in order of decreasing priority until reaching a swing well. The
user-supplied priority formulae and numerical controls are specified by
keyword *PRIOR-FORM (refer to the explanation therein for details).
DEFAULTS:
Optional keyword. Default is to use the instantaneous potential (*IP) to determine rate
distributions among wells to meet the group target rate.
CONDITIONS:
This keyword must be located in the WELL AND RECURRENT DATA keyword group. It
should appear after the group hierarchy is available. Defining an apportionment method for
non-targeted groups or non-targeted injection streams does NOT have an effect. Priority
ranking apportionment currently does not support the manifold groups.
User's Guide IMEX
EXPLANATION:
groups and the wells directly attached. Groups and wells can only contribute to one particular
target. They will be excluded from the pool created for a higher-level target if they are
currently contributing to a subordinate target. The idea can be best demonstrated with the
following group hierarchy sketch: if only the FIELD target is defined, all the groups and
wells listed will contribute to it; however, if GROUP-2 is targeted additionally, then its child
groups (GROUP-201, GROUP-202 and GROUP-203) and their wells (W05W10) will
contribute directly to the GROUP-2 target instead of the top-level FIELD target. The pool
created originally for FIELD now consists of only GROUP-1 and its wells (W01W04).
FIELD
GROUP-1
W01 W02 W03 W04
GROUP-2
GROUP-201
GROUP-202
GROUP-203
W05 W06 W07
W08 W09
W10
For a given group target, a proper apportionment method needs to be specified. Different
apportionment method can in principal be applied\mixed to different group targets in a single
run. CMGs multi-level well management module currently supports the following methods:
For the above example, the FIELD target can be apportioned, say, with the instantaneous
potential method (*IP) while the priority ranking method (*PRIOR) may be used for the
targeted GROUP-2 in a single run.
Example:
*GCONP FIELD
*TARGET *STO 1000.
*GCONP GROUP-2 *TARGET *STO 600.
*APPOR-METHOD *PROD FIELD
*IP
*APPOR-METHOD *PROD GROUP-2 *PRIOR
*PRIOR-FORM
'Group-2'
*PRIOR-NUMER
0.
1.
0.
0.
0.
*PRIOR-DENOM
0.
0.
1.
0.
0.
0.
0.
0.
0.
0.
0.
In case that the GROUP-2 target is to be apportioned using the guide rates method
(*GUIDE), the group rates for contributing groups GROUP-201, GROUP-202 and GROUPUser's Guide IMEX
203 and/or well rates for contributing wells W05 W10 shall be supplied by *GUIDEP.
Guide rates supplied for non-contributing groups\wells are ignored. One the other hand,
instantaneous potential (IP) is defaulted as the guide rate if it has not been supplied by the
user for a contributing group or well that is under the guide rates control.
*APPOR-METHOD
*GUIDEP *STO
**and/or
*GUIDEP *STO
*PROD GROUP-2
*GUIDE
GROUP-201 GROUP-202 GROUP-203
300.
200.
100.
W05
100.
*W06
W07
W08
W09
W10
Backward Compatibility Notes:
The sub-keywords *IPP under *GCONP and *IIP under *GCONI in the previous versions,
which set the *IP apportionment flag globally for all the group production or injection targets,
now become obsolete. They will be ignored with reminding message if *APPOR-METHOD
is encountered ahead in the datasets. However, the backward compatibility is still maintained
and any old dataset containing *IPP or *IIP will run exactly the same since it shall not have
the new keyword *APPOR-METHOD.
*GCONP *IPP = *APPOR-METHOD *PROD all_groups *IP
*GCONI *IIP = *APPOR-METHOD *GASI all_groups *IP
*APPOR-METHOD *WATI all_groups *IP
*APPOR-METHOD *SOLI all_groups *IP (for IMEX only)
The group and/or well guide rates used in the guide-rate apportionment are supplied
following keywords *GUIDEP and *GUIDEI as in the previous versions. However, they do
not set the apportionment flag anymore. The guide-rate apportionment has to be invoked by
entering the keyword *GUIDE under *APPOR-METHOD for either production or injection
targets. The backward compatibility is still maintained by interpreting *GUIDEP and
*GUIDEI as the old ones if *APPOR-METHOD does not appear ahead in a dataset.
Therefore, old datasets containing *GUIDEP and *GUIDEI will run exactly the same.
*GUIDEPold ref stream guide_rates
= *APPOR-METHOD *PROD all_groups *GUIDE
*GUIDEP ref stream guide_rates
*GUIDEIold *STG guide_rates
= *APPOR-METHOD *GASI all_groups *GUIDE
*GUIDEI *STG guide_rates
*GUIDEIold *STW guide_rates
= *APPOR-METHOD *WATI all_groups *GUIDE
*GUIDEI *STW guide_rates
*GUIDEIold *STS guide_rates
(for IMEX only)
= *APPOR-METHOD *SOLI all_groups *GUIDE
*GUIDEI *STS guide_rates
User's Guide IMEX
Priority Formulae for Apportionment (Conditional)

*PRIOR-FORM
PURPOSE:
*PRIOR-FORM defines the priority formulae and numerical control parameters for the
priority ranking apportionment method or the internal guide rate method set by *APPORMETHOD *PRIOR / *INGUIDE to meet group targets. It is aimed to produce a desired
stream while minimizing the production of other/nuisance streams. It can also be used to meet
injection target by which only the most injectable wells operate.
FORMAT:
*PRIOR-FORM (*PROD|*GASI|*WATI|*SOLI) 'group_names'

(*PRIOR-RATE (*MRC|*BHP(bhp_val)))
(*PRIOR-CTRL freq tcr_min tcr_max)
(*PRIOR-NUMER A0 A1 ... Anph)
(*PRIOR-DENOM B0 B1 ... Bnph)
DEFINITIONS:
*PROD
Indicates that the priority formula is applied to the group production targets
with the priority ranking method.
*GASI
Indicates that the priority formula is applied to the group gas injection targets
with the priority ranking method.
*WATI
Indicates that the priority formula is applied to the group water injection
targets with the priority ranking method.
*SOLI
Indicates that the priority formula is applied to the group solvent injection
targets with the priority ranking method (valid for IMEX only).
'group_names'
A list of 'group_names' to which the priority formula is applied. No wildcard
characters may be used in the group names.
*PRIOR-RATE
This sub-keyword defines what type of rate is to be used in calculating the
well priorities. *MRC indicates that the well priorities are to be calculated
using the stream rates under the most restrictive condition. *BHP indicates
that the well priorities are to be calculated using the production or injection
potentials at a specified bottom hole pressure, bhp_val. If bhp_value does
not appear after *BHP, the default value will be applied: 1 atmosphere for
production targets and 3500 atmosphere for injection targets.
User's Guide IMEX
*PRIOR-CTRL
This sub-keyword introduces the numerical parameters that control the
priority ranking apportionment. freq is the minimum time interval in days
between well priority calculations. tcr_min is the lower limiting value of
the ratio between the pool target and its capacity, above which the priority
ranking apportionment will be applied. tcr_max is the upper limiting value
of the ratio between the pool target and its capacity, below which the priority
ranking apportionment will be applied. *PRIOR-CTRL is no necessary to
and thus has no effect on the internal guide rate apportionment.
*PRIOR-NUMER
This sub-keyword introduces the priority weighting coefficients for the
numerator Ai (i = 1, nph). nph is the number of surface production or
injection streams allowed for group targets.
*PRIOR-DENOM
This sub-keyword introduces the priority weighting coefficients for the
denominator Bi (i = 1, nph).
DEFAULTS:
Conditional keyword. If any of the sub-keywords (*PRIOR-CTRL, *PRIOR-RATE,

*PRIOR-NUMER, *PRIOR-DENOM) is absent after *PRIOR-FORM, its default value(s)
will be applied:
*PRIOR-RATE
*PRIOR-CTRL
*PRIOR-NUMER
*PRIOR-DENOM
*MRC
0.0 0.0
1.0 0.0
1.0 0.0
1.0
0.0 ...
0.0 ...
0.0
0.0
Otherwise, all the required values by that sub-keyword must be filled in. The default for
internal guide rate apportionment (*INGUIDE) is to use the instantaneous potentials as the
guide rates if *PRIOR-FORM does not present.
CONDITIONS:
This keyword must be located in the WELL AND RECURRENT DATA keyword group after
the group hierarchy is available. If a group target is apportioned with the priority ranking
method set by *APPOR-METHOD, a priority formula for that targeted group must be
supplied by *PRIOR-FORM. Defining priority formula for non-targeted groups or nontargeted injection streams does NOT have an effect.
EXPLANATION:
*PRIOR-FORM is used for either the priority ranking apportionment or the internal guide
rate apportionment. In the priority ranking apportionment, wells contributing to a target are
ranked in order of decreasing priority. When the pool target-capacity ratio falls between the
threshold values [tcr_min, tcr_max], wells with highest priorities are assigned to operate at
tcr_max fraction (upper limit) of their maximum rates according to the most restrictive
constraints. If they have sufficient capacity to meet the group target, one 'swing' well will be
identified such that the target will be exceeded if this well were also assigned tcr_max
fraction of its maximum rate. Wells with lower priorities than that of the swing well are
assigned tcr_min fraction (lower limit) of their max rates. The swing well is then assigned a
User's Guide IMEX
makeup rate exactly equal to the deficit. Wells with lower priority, including the swing well,
are flagged so that they are not shut-in by well management in case of violating the minimum
rate constraints.
The priority index for an individual well iw contributing to a targeted group ig is defined as
nph
Priority(iw ) =
A 0 (ig ) + A i (ig )Q i (iw )

i =1
nph
B 0 (ig ) + B i (ig )Q i (iw )

i =1
where Ai and Bi (i = 0, nph) are the weighting coefficients for the numerator and denominator,
respectively, introduced by sub-keywords *PRIOR-NUMER and *PRIOR-DENOM. It is
assumed that all wells contributing to a common group target should have the same priority
definition. However, different targeted groups or targeted injection streams may have different
specifications of well priorities. Coefficients A0 and B0 are constants whilst the rest are the
stream weighting coefficients to be multiplied correspondingly by Qi, the stream rate under the
most restrictive constraints (*MRC) or the production/injection potential at the given bottom
hole pressure (*BHP) distinguished by sub-keyword *PRIOR-RATE . The allowable surface
reference streams for the simulator are outlined in the table below. They are defined in detail
elsewhere on the manual page for keywords *GUIDEP and *GUIDEI. All the weighting
coefficients are non-negative real numbers and at least one Ai and one Bi must be non-zero.
IMEX (Production: nph = 6)
i
CONST
*STO
*STG
*STW
*STS
*STL
*BHF
NUMER
DEMON
A0
B0
A1
B1
A2
B2
A3
B3
A4
B4
A5
B5
A6
B6
IMEX (Injection: nph = 3)

i
CONST
*STG
*STW
*STS
NUMER
DEMON
A0
B0
A2
B2
A3
B3
A4
B4
The first value read by sub-keyword *PRIOR-CTRL, freq, indicates the elapsed minimum
time in days between priority calculations. The default is to update the well priority at every
timestep (freq = 0.0). It is designed to reduce in some cases the frequent shifting of the swing
well among wells that may have very comparable priority values.
The last two values after sub-keyword *PRIOR-CTRL, tcr_min and tcr_max, are the
threshold values for the pool target-capacity ratio (between 0-1), within which the ranking
apportionment will be applied. The instantaneous potential (IP) apportionment will otherwise
be adopted if the said ratio is out of the range [tcr_min, tcr_max]. Default is [0, 1], meaning
that the priority ranking apportionment will always be in place regardless of the pool targetcapacity ratio.
The upper limit tcr_max is designed to avoid the constant violation (and therefore switching)
of the most restrictive constraints that might happen to some wells if they were forced to
produce or inject at their maximum rates (tcr_max = 1). Another consideration is that if the
pool has a very low capacity to meet a certain target (i.e. high pool target-capacity ratio), all
User's Guide IMEX
contributing wells are required to operate at their maxima. The ranking apportionment may
take the least effect because there is no swing well to choose.
The lower limit tcr_min, on the other hand, provides a flexible control over the wells with
lower priorities than the swing well. These wells can operate at the tcr_min fraction of their
maximum rates allowable instead of at the zero rate.
Examples:
Supply the priority formula for targeted groups Group-1 and Group-2. As invoked by
*APPOR-METHOD for the priority ranking method, wells will be opened in order of
decreasing GOR when the pool target-capacity ratios fall between [0.0, 0.8]. Well priorities
are evaluated using the *MRC rates (default) and recalculated at least every 30 days.
*APPOR-METHOD *PROD 'Group-1' 'Group-2' *PRIOR
*PRIOR-FORM
'Group-1' 'Group-2'
**freq
tcr_min
tcr_max
*PRIOR-CTRL
30.
0.0
0.8
**CONST STO STG STW STS STL
*PRIOR-NUMER
0.
1.
0.
0.
0.
0.
*PRIOR-DENOM
0.
0.
1.
0.
0.
0.
BHF
0.
0.
Wells can also be opened in order of decreasing WCUT:

*PRIOR-NUMER
*PRIOR-DENOM
0.
0.
1.
0.
0.
0.
1.
1.
0.
0.
0.
0.
0.
0.
or in order to minimize gas and water production as a whole. Proper weight for the gas is
needed since rate values of gas production are generally higher by magnitudes than those of
water:
*PRIOR-NUMER
*PRIOR-DENOM
0.
0.
1.
0.
0.
0.
0.01 1.
0.
0.
0.
0.
0.
0.
In conjunction with the trigger action (*TRIGGER), appropriate priority formulae can also be
used alternatively to minimize more than one stream. The following example shows the
flexibility to reduce the gas production using the priority formula based on gas oil ratio. One
can switch to that based on the water cut when *WCUT > 0.5 and *GOR < 200.0 through
nested trigger actions.
*GCONP 'FIELD' *TARGET *STO 500.0
*APPOR-METHOD *PROD 'FIELD' *PRIOR
*PRIOR-FORM 'FIELD'
*PRIOR-CTRL 30. 0. 1.
*PRIOR-NUMER
0.
1.
*PRIOR-DENOM
0.
0.
0.
1.
0.
0.
0.
0.
0.
0.
0.
0.
*TRIGGER 'REDUCE_WCUT1' *ON_GROUP 'FIELD' *WCUT > 0.5 APPLY_TIMES 5

*TRIGGER 'REDUCE_WCUT2' *ON_GROUP '@' *GOR < 200.0
*PRIOR-FORM
'@'
*PRIOR-NUMER
0.
1.
0.
1.
0.
0.
0.
*PRIOR-DENOM
0.
0.
0.
1.
0.
0.
0.
*END_TRIGGER
*END_TRIGGER
User's Guide IMEX
Guide Rates for Groups or Wells
*GUIDEP, *GUIDEI
PURPOSE:
*GUIDEP specifies the use of guide rates to distribute the production rates to groups or wells
so as to meet the production target.
*GUIDEI specifies the use of guide rates to distribute the injection rates to groups or wells so
as to meet the injection target.
FORMAT:
*GUIDEP
*GUIDEI
*STO
*STG
*STW
*STS
*STL
*BHF
(group_names)
(well_names)
guide_rates
guide_rates
DEFINITIONS:
*STO
This subkeyword indicates that the oil stream is the reference stream to
which the guide rates apply; i.e., that the guide rate values entered should be
interpreted as oil rates or as being proportional to oil rates. Not valid for use
with *GUIDEI. Rates should be entered as (M3/day | STB/day | cm3/min).
*STG
This subkeyword indicates that the gas stream is the reference stream to
interpreted as gas rates or as being proportional to gas rates. Rates should be
entered as (M3/day | SCF/day | cm3/min).
*STW
This subkeyword indicates that the water stream is the reference stream to
interpreted as water rates or as being proportional to water rates. Rates
should be entered as (M3/day | STB/day | cm3/min).
*STS
This subkeyword identifies that the solvent stream is the reference stream to
interpreted as solvent rates or as being proportional to solvent rates. Rates
should be entered as (M3/day | SCF/day | cm3/min).
User's Guide IMEX
*STL
This subkeyword indicates that the guide rates apply to the total of the liquid
stream rates (oil+water); i.e., that the guide rate values entered should be
interpreted as total liquid rates or as being proportional to total liquid rates.
Not valid for use with *GUIDEI. Rates should be entered as (M3/day |
STB/day | cm3/min).
*BHF
Identifies that guide rate values are bottom-hole fluid rates. Rates should be
entered in (rm3/day | rbbl/day | rcm3/min). Not valid for use with *GUIDEI.
group_names
A list of 'group_names' to which the guide rates are applied. No wildcard
characters may be used in the group names.
well_names
A list of 'well_names' to which the guide rates are applied.
'?ell1').
guide_rates
A list of guide rates for the list of groups or wells. See the entries for the
stream designation strings for the proper units.
If only one value of 'guide_rates' is input, all wells or groups defined in the
well_names or group_names list will be set to that single value.
DEFAULTS:
Optional keywords. Default is to use the instantaneous injection / production potential (IIP or
IPP) to determine rate distributions among groups or wells to meet the target rate.
CONDITIONS:
*GUIDEP AND *GUIDEI must be located in the Well and Recurrent Data keyword group,
and must follow the *GCONI, *GCONP or *GCONM keywords.
User's Guide IMEX
EXPLANATION:
Guide rates for each of the wells specified in 'well_names' or each group specified by the
'group_names'. If guide rates are being used then all producers or injectors connected to that
group must have their guide_rates specified. If no guide rates at all have been specified for
the apportionment of the target of a particular group, then instantaneous injection/production
potentials are used as the guide rates.
When guide rates are used, the contributing wells or groups receive a fraction of the target
which is proportional to its guide rate. If this assignment exceeds the well's or the group's
maximum rate, then the maximum rate is assigned and the remainder of the target is
apportioned among the other contributors, with the contributions still proportional to the
guide rates but with the contributors set at maximum left out of the apportionment.
When *GUIDEI is used, the guide rates apply only for the target specified with the stream
identifier; e.g., water injectors have their guide rates specified under *GUIDEI *STW.
When *GUIDEP is used, the guide rates apply for the listed wells or groups to all target
types, with the proper conversion made for the ratios at which the wells produce the different
streams. For example, if the following lines are entered when SI units are used,
*GUIDEP
*STO
Well 1 Well 2 Well 3

100.
200.
300.
and an STG group target is to be apportioned among these producers, then if Well 1 has a
GOR of 600., Well 2 a GOR of 300., and Well 3 a GOR of 200., then the STG guide rates
used for the wells would be 60000, 60000, and 60000 respectively and the STG target would
be distributed equally among the wells if no other constraints were violated in the process.
Example:
*GUIDEP
*GUIDEP
*GUIDEP
*GUIDEP
*GUIDEI
User's Guide IMEX
*STO 'GR-PA' 'GR-PB' 'GR-PC'

300.00 100.00 400.00
*STO 'PA1' 'PA2' 'PA3'
100.0 100.0 100.0
*STL 'PB1' 'PB2' 'PB3'
100.0 100.0 100.0
*STG 'PC1' 'PC2'
100.0 200.0
*STW 'IA1' 'IB1' 'IC1'
100.0 200.0 100.0
Flag for Accompanying Groups or Wells Not Under Group

Control (Optional)
*GCPOFF, *GCIOFF
PURPOSE:
*GCPOFF specifies the accompanying groups or wells are not under group production
constraint from higher level groups.
*GCIOFF specifies the accompanying groups or wells are not under group injection
constraint from higher level groups.
FORMAT:
*GCPOFF
-or*GCIOFF
(group_names)
(well_names)
*GAS
*WATER
(group_names)
(well_names)
DEFINITIONS:
*GAS
Indicates not under group control for gas injection calculations.
*WATER
Indicates not under group control for water injection calculations.
group_names
A list of 'group_names' to which the group constraint from higher level
should not be applied. No wildcarding is allowed for group names.
well_names
A list of 'well_names' to which the group constraint from higher level should
not be applied.
'?ell1').
User's Guide IMEX
DEFAULTS:
Optional keywords. Default is to apply the group constraint from higher level.
CONDITIONS:
*GCPOFF AND *GCIOFF must be located in the Well and Recurrent Data keyword group,
and must follow the *GCONI, *GCONP or *GCONM keywords.
EXPLANATION:
This keyword isolates the groups or wells so that they can produce or inject according to their
own rate and pressure constraints, without being controlled by higher level constraints.
Example:
*GCPOFF
*GCPOFF
*GCIOFF
*GCIOFF
User's Guide IMEX
'GR-PA' 'GR-PB' 'GR-PC'

'PA1' 'PB2' 'PC3'
*GAS
'GR-PA'
*WATER 'IB2' 'IC1'
Allows for a Set of Keywords to be Processed When a

Specified Condition (Trigger) is Satisfied (Optional) *TRIGGER
PURPOSE:
*TRIGGER allows the user to specify certain actions which are implemented when a specific
condition or trigger is satisfied during the simulation.
FORMAT:
The entire *TRIGGER keyword must appear on one line followed by a list of actions
and then by the *END_TRIGGER.
*TRIGGER trig_name trig_def *APPLY_TIMES napt *INCREMENT rinc
*TEST_TIMES ntestt *TEST_AFTER_TIMER rtimedr *TEST_AFTER_TIMEA rtimeda
{list of actions}
*END_TRIGGER
Where trig_def is one of

(*ON_WELL 'well_names' well_condition operator condition_value)
-or(*ON_GROUP 'group_names' group_condition operator condition_value)
-or(*ON_LAYER 'well_name' layer_UBA layer_condition operator condition_value)
-or(*ON_SECTOR 'sector_name' sector_condition operator condition_value)
-or(*ON_FIELD 'FIELD' field condition operator condition_value)
-or(*ON_ELAPSED TIME time condition operator condition_value)
And where {list of actions} represents valid well and recurrent data keyword lines.
DEFINITIONS:
trig_name
Enter a string (in single quotes) of less than 40 characters to uniquely identify
the trigger. The name must immediately follow the *TRIGGER keyword.
This token is required.
*ON_WELL
This subkeyword indicates that the test condition is to be applied to a well or
list of wells (list, if a wildcard is used in the well name or a list of wells is
specified). One of *ON_WELL or *ON_GROUP or *ON_SECTOR or
*ON_LAYER or *ON_FIELD is a required token immediately following the
trigger name string.
well_names
trigger applies. The well(s) must be previously defined. The well names must
all be specified on a single line.
User's Guide IMEX

on its own to replace all wells (e.g. *TRIGGER trig1 *ON_WELL '*' or
*TRIGGER trig1 *ON_WELL 'wel*').
? Replaces any single character anywhere in the well name (e.g. *TRIGGER
trig1 *ON_WELL '?ell1').
If *ON_WELL is specified, then at least one well name must
immediately follow the *ON_WELL keyword.
Well_condition
Enter a single keyword identifying one of the following: a well stream rate or
cumulative or composition or well bottom hole or tubing head pressure or
backflow. The valid lists of conditions for a well are shown in Table 1 below:
This is a required token and must immediately follow the well name or well list.
Table 1: Well quantities:
Subkeyword
STO-RP
STO-CP
STO-RI
STO-CI
STW-RP
STW-CP
STW-CI
STG-RP
STG-CP
STG-RI
STG-CI
STL-RP
STL-CP
BHF-RP
BHF-CP
BHF-RI
BHF-CI
STS-RP
STS-CP
STS-RI
STS-CI
STI-RP
STI-CP
User's Guide IMEX
Meaning
Stock Tank Oil - Rate of Production
Stock Tank Oil - Cumulative Production
Stock Tank Oil - Rate of Injection
Stock Tank Oil - Cumulative Injection
Stock Tank Water - Rate of Production
Stock Tank Water - Cumulative Production
Stock Tank Water - Cumulative Injection
Stock Tank Gas - Rate of Production
Stock Tank Gas - Cumulative Production
Stock Tank Gas - Rate of Injection
Stock Tank Gas - Cumulative Injection
Stock Tank Liquid - Rate of Production
Stock Tank Liquid - Cumulative Production
Bottom Hole Fluid - Rate of Production
Bottom Hole Fluid - Cumulative Production
Bottom Hole Fluid - Rate of Injection
Bottom Hole Fluid - Cumulative Injection
Stock Tank Solvent - Rate of Production (IMEX only)
Stock Tank Solvent - Cumulative Production (IMEX only)
Stock Tank Solvent - Rate of Injection (IMEX only)
Stock Tank Solvent - Cumulative Injection (IMEX only)
Stock Tank Intermediate liquid - Rate of Production (GEM only)
Stock Tank Intermediate liquid - Cumulative Production (GEM only)
WTG-RP
WTG-CP
BHP
WHP
GOR
WCUT
WGR
GLR
MXX
BKFLOW
Wet Gas - Rate of Production (GEM only)

Wet Gas - Cumulative Production (GEM only)
Bottom Hole Pressure
Well Head Pressure
Gas-Oil Ratio (production)
Water Cut (production)
Gas Liquid Ratio (production)
Mole percent of component XX in produced well stream hc (GEM only)
Back flow (true if any layer is back flowing)
*ON_GROUP
This subkeyword indicates that the test condition is to be applied to a group
or list of groups (if a wildcard is used in the group name or a list of groups is
specified). The group hierarchy must be previously defined to enable group
based triggers to be used. One of *ON_WELL or *ON_GROUP or
*ON_SECTOR or *ON_LAYER or *ON_FIELD is a required token
immediately following the trigger name string.
group_names
Any number of group names (in quotes) to specify the groups to which this
trigger applies. The groups(s) must be previously defined. There are no
groups created by default. The names must all be specified on a single line.
Note: wildcards may be used in the 'group_names' string as follows:
* replaces any number of characters at the end of a group name or can be used
on its own to replace all groups (e.g. *TRIGGER trig1 *ON_GROUP '*' or
*TRIGGER trig1 *ON_GROUP 'grp*').
? Replaces any single character anywhere in the group name (e.g. *TRIGGER
trig1 *ON_GROUP '?rp1').
used the group list generated is printed out for the user to check.
At least one group name must immediately follow the *ON_GROUP
keyword. Required token.
Group_condition
The valid list of conditions for groups and for the field is shown in Table 2
below:
User's Guide IMEX
Table 2: Group and Field quantities:

Subkeyword
STO-RP
STO-CP
STW-RP
STW-CP
STW-RI
STW-CI
STG-RP
STG-CP
STG-RI
STG-CI
STL-RP
STL-CP
BHF-RP
BHF-CP
BHF-RI
BHF-CI
STS-RP
STS-CP
STS-RI
STS-CI
STI-RP
STI-CP
WTG-RP
WTG-CP
GOR
WCUT
WGR
GLR
MPWS MXX
GWGR
WWGR
RECYSTG
VOIDRPG
VOIDRPW
VOIDRPS
VOIDRPV
VOIDRPT
User's Guide IMEX
Meaning
Stock Tank Oil - Rate of Production
Stock Tank Oil - Cumulative Production
Stock Tank Water Rate of Production
Stock Tank Water Cumulative Production
Stock Tank Water Rate of Injection
Stock Tank Water Cumulative Injection
Stock Tank Gas Rate of Production
Stock Tank Gas Cumulative Production
Stock Tank Gas Rate of Injection
Stock Tank Gas Cumulative Injection
Stock Tank Liquid Rate of Production
Stock Tank Liquid Cumulative Production
Bottom Hole Fluid Rate of Production
Bottom Hole Fluid Cumulative Production
Bottom Hole Fluid Rate of Injection
Bottom Hole Fluid Cumulative Injection
Stock Tank Solvent Rate of Production (IMEX only)
Stock Tank Solvent Cumulative Production (IMEX only)
Stock Tank Solvent Rate of Injection (IMEX only)
Stock Tank Solvent Cumulative Injection (IMEX only)
Stock Tank Intermediate liquid Rate of Production (GEM only)
Stock Tank Intermediate liquid Cumulative Production (GEM only)
Wet Gas Rate of Production (GEM only)
Wet Gas Cumulative Production (GEM only)
Water Gas Ratio (production)
Mole percent of component XX in group production (GEM only)
Gas-Wet Gas Ratio (production GEM only)
Water-Wet Gas Ratio (production GEM only)
Group Recycled Gas injection rate
Group voidage replacement ratio by gas injection
Group voidage replacement ratio by water injection
Group voidage replacement ratio by solvent injection (IMEX only)
Group voidage replacement ratio by gas and solvent injection (IMEX
only)
Group voidage replacement ratio by all injection streams
*ON_LAYER
This subkeyword identicates that the test condition is to be applied to a well
layer. The well must be fully defined previously for a trigger based on a layer
condition to be specified.
well_name
A single string (in quotes) of less than 40 characters representing a well name
must immediately follow the *ON_LAYER keyword. The name identifies
the well to which the layer belongs.
Layer_UBA
To identify the layer, enter the layer user block address immediately following the
well name. The user block address is specified in the following general format: i1
j1 k1 / i2 j2 k2 / please review the manual pages on the *PERF keyword for
more information. Do NOT encapsulate the user block address in quotes.
Layer_condition
The valid list of conditions for layers is shown in table 3 below:
Table 3 : Layer quantities:
Subkeyword
STO-R
STW-R
STG-R
STS-R
STI-R
GOR
WCUT
WGR
GLR
BHP
DWM
Meaning
Stock Tank Oil - Rate (sign tells whether Prod or Inj)
Stock Tank Water - Rate
Stock Tank Gas - Rate
Stock Tank Solvent - Rate (IMEX)
Stock Tank Intermediate Liquid - Rate (GEM)
Water Gas Ratio (production)
Layer bottom-hole pressure
Layer drawdown - absolute value of block P - layer P
*ON_SECTOR
This subkeyword identicates that the test condition is to be applied to a
sector. GEM creates a default sector named FIELD which includes all grid
blocks and all wells. IMEX creates a default sector named (Entire Field).
Please note that there are 2 spaces between the word Entire and Field.
Any other sector name used with a trigger must be previously defined.
sector_name
A single string representing a sector name (in quotes) of less than 16
characters to specify the sector to which this trigger applies. The sector must
be previously defined. Only the FIELD (Entire Field) sector is created by
default. The name must be on the same line as the *TRIGGER keyword.
User's Guide IMEX
Table 4 : Sector quantities:

Subkeyword
PAVE
SOAVE
SWAVE
SGAVE
STOIP
STWIP
STFGIP
STGIP
Meaning
Pore-volume weighted pressure
Pore-volume weighted oil saturation
Pore-volume weighted water saturation
Pore-volume weighted gas saturation
Stock tank oil in place
Stock tank water in place
Stock tank free gas in place
Stock tank gas in place, including free and dissolved
Sector_condition
The valid list of conditions for sectors is shown in table 4 above:
*ON_FIELD
This subkeyword identicates that the test condition is based on a field level
quantity. Note: The string FIELD in single quotes must immediately follow
the *ON-FIELD keyword. The ON_FIELD keyword on its own adequately
identifies a trigger on the entire field. However to maintain consistency in
format with other keywords such as ON_WELL and ON_GROUPS, the field
name string has been retained for ON_FIELD as well. The user should enter
the string FIELD for the field name token even if the actual field level
group or sector name assigned is different from FIELD. If a string other
than FIELD is entered following ON_FIELD, a warning message will be
generated and the simulation will proceed normally.
Field_condition
The valid list of conditions for field is identical to that of groups as shown in
Table 2:
Operator
The operator for the triggering condition must be one of:
< Less than
> Greater than
One of < or > (not in quotes) is required immediately following the
specification of the test quantity using keywords such as *STO-RP.
condition_value
The value of the trigger condition. Enter a value based on the trigger
condition and unit system selected for the simulation.
User's Guide IMEX
*ON_ELAPSED
This subkeyword identicates that the test condition is based on a time elapsed
since the beginning of the simulation OR time elapsed after the time the trigger
is defined (that is, when the trigger keyword lines are parsed by the reader). The
trigger definition time could simply be the time specified by the DATE or
TIME card. For an inner trigger of a nested trigger, it is the time that the
immediate outer trigger condition is satisfied. Consider the following examples:
Example 1:
*TIME 10.0
trigger 'trig1' on_elapsed
apply_times 1
*TARGET *BHP
'P1'
200
end_trigger
'time' treltd > 4.99
In this case the trigger will be read in after exactly 10 days have elapsed
trigger definition time is therefore 10 days. The time specified for the trigger
condition is relative to the time the trigger is created therefore at the bottom
of the first time step that results in simulation time greater than 10+4.99 =
14.99 days the trigger condition will be met.
Example 2:
*TIME 10.0
apply_times 1
*TARGET *BHP
'P1'
200
apply_times 1
*TARGET *BHP
'P1'
100
apply_times 1
*TARGET *BHP
'P1'
50
end_trigger
end_trigger
end_trigger
In this case the outermost trigger named TRIG1 is defined at 10 days based
on the time card entry. The inner trigger TRIG2 will be defined or comes into
existence when the outer trigger TRIG1 condition is satisfied (some time
after 14.99 days because the time step sizes are not known apriori it is not
possible to state exactly when this time will be). The inner TRIG3 will be
defined when the outer trigger TRIG2 condition is satisfied, this can occur
after at least 10+4.99+5.99 = 20.98 days have elapsed.
User's Guide IMEX
Note: The string TIME in single quotes must immediately follow the *ONELAPSED keyword. If a string other than TIME is entered following
ON_ELAPSED, a warning message will be generated and the simulation will
proceed normally.
Time_condition
Either the time elapsed from the start of the simulation can be entered or the
time elapsed relative to the time when the trigger is first defined can be
specified. To specify the absolute time use the sub keyword *TIMSIM
followed by a value. To specify elapsed time relative to the time when the
trigger is parsed or defined use the sub keyword *TRELTD followed by a
value. The time should be in days for FIELD/SI/MODSI units and in minutes
for laboratory (LAB) units.
TIMSIM
Indicates that the time value entered is time elapsed from the start of the
simulation, also referred to as absolute time.
TRELTD
Indicates that the time value entered is relative to the time the trigger is
defined, also referred to as relative time.
Operator
The operator for the triggering condition must be one of:
< Less than
> Greater than
One of < or > (not in quotes) is required immediately following the
specification of the test quantity using keywords such as *TIMSIM.
condition_value
The value of the trigger condition. Enter a value based on the time elapsed
condition selected in units of days for field/SI/MODSI and in units of
minutes for laboratory units (LAB).
*APPLY_TIMES
Subkeyword used to specify the maximum number of times that the actions
specified with the trigger can be taken. An integer number must immediately
follow this subkeyword. This subkeyword is optional.
User's Guide IMEX
napt
Enter a single integer representing the maximum number of times that the
specified list of actions can be executed. If no value is entered, then the
default is 1. With the default of 1 the trigger condition is tested at the end of
every timestep. As soon as the trigger condition is satisfied the list of actions
is implemented and the trigger is removed from the list of active triggers. If
more than 1 (say n times) is selected then the trigger remains active until
the trigger condition is satisfied (n) times.
*INCREMENT
Subkeyword used to specify the increment to the trigger value. A single real
number must follow this subkeyword. This subkeyword is optional.
Trigger Increment
Enter a single real value representing an increment to be applied to the
trigger value, each time the trigger condition is satisfied. The trigger
increment can be a negative number. The trigger increment does not need to
be entered. The trigger increment can only be entered if the preceding
number for the repetition times is also entered. The trigger increment is used
only if a value greater than 1 is specified for the number of times the trigger
condition can be satisfied. Once a trigger condition is satisfied the trigger
value is recalculated as value_new = existing_value + increment. The new
value is then used in testing the trigger condition for subsequent times until
the trigger condition is once again satisfied.
*TEST_TIMES
Subkeyword used to specify the maximum number of times that the trigger
can be tested to ascertain if the trigger condition is satisfied. A single integer
number must follow this subkeyword. This subkeyword is optional.
ntestt
Enter a single integer representing the maximum number of times that the
trigger can be tested to ascertain if the trigger condition is satisfied. If no
value is entered, then the default is to test the trigger every timestep. If a
value of 1 is entered then the trigger condition is tested only once at the end
of the timestep during which the trigger is defined. The trigger is then
removed from the active trigger list whether or not the trigger condition itself
is satisfied. If more than 1 (say n times) is selected then the trigger
condition is tested for n timesteps after the trigger is defined.
*TEST_AFTER_TIMER
Subkeyword used to specify the time delay which must elapse before the
trigger condition will be tested. This time delay is relative to the time that the
trigger comes into existence or is defined. A single real number must follow
this subkeyword. This subkeyword is optional.
User's Guide IMEX
rtimedr
Enter a single real number representing the delay in time after the time that
the trigger is defined or parsed that the user wishes the trigger condition to be
tested. If no value is entered, then the default is to assume a time delay of
zero. In days for FIELD/SI/MODSI units and in minutes for laboratory
(LAB) unit system.
*TEST_AFTER_TIMEA
Subkeyword used to specify the time delay in days for FIELD/SI/MODSI
units and in minutes for laboratory (LAB) units, which must elapse before
the trigger condition will begin to be tested. This time delay is relative to the
start of the simulation or absolute time elapsed. A single real number must
follow this subkeyword. This subkeyword is optional.
rtimeda
Enter a single real number representing the delay in time after the start of the
simulation that the user wishes the trigger condition to be tested going
forward. If no value is entered, then the default is to assume a time delay of
zero. In days for FIELD/SI/MODSI units and in minutes for laboratory
(LAB) unit system.
{action_list}
The list of actions in the form of valid well and recurrent data keyword lines may
be specified following the *TRIGGER keyword and its subkeywords. The action
list must start on a new line. The action list is optional. Following the action lines
specify the keyword *END_TRIGGER on a new line to signal the end of the
trigger definition. It is okay to not specify any action lines within a trigger.
*END_TRIGGER
This keyword marks the end of the list of actions or keyword lines associated
with a given trigger. It must be on a new line. The *TRIGGER and
*END_TRIGGER must be used as a pair. For each *TRIGGER keyword, an
*END_TRIGGER keyword is required.
*STO-RP
Oil production rate at surface conditions. For SI units enter a value in m3/D
and for field units in STB/D.
*STO-R
Oil phase rate at surface conditions for a layer. A positive number implies
fluid is flowing in a normal direction. Specifically flow is expected to be
from reservoir to well bore for a layer belonging to a producer and from well
bore to reservoir if the layer belongs to an injection well. On the other hand if
a negative number is entered, then the user is testing for a back flowing layer.
The magnitude of the number indicated the severity of the back flow. For SI
units enter a value in m3/D and for field units in STB/D.
User's Guide IMEX
Example 1:
*TRIGGER trig1 *ON_LAYER well1 1 1 1 / 2 2 2 *STO-R > 300.0
If well1 is an injector, then the condition will be satisfied if flow for the layer
is from the well to the reservoir (injection) and the flow rate is greater than
300. However if well1 is a producer, then the condition will be satisfied if
flow is from reservoir to well bore (production) and the oil flow rate is
greater the 300.0.
Example 2:
*TRIGGER trig2 *ON_LAYER well1 1 1 1 / 2 2 2 *STO-R > -300.0
The condition will be satisfied if the layer is back flowing, for an injector flow
is from the reservoir to the well bore and the magnitude of the back flow rate is
greater than 300.0. For a producer, the condition will be satisfied if flow is
from well bore to reservoir and the magnitude of the back flow rate is greater
than 300.0. Therefore if the back flow rate is 350, then the condition is
satisfied, if the back flow rate is 250 the condition is not satisfied.
*STW-R
Water phase rate at surface conditions for a layer. A positive number entered
for a layer belonging to an injection well implies flow is expected to be from
well bore to reservoir, that is the well layer is injecting into the reservoir. A
positive number entered for a layer belonging to a production well implies
flow is from reservoir to well bore. A negative number implies a back flowing
layer. For SI units enter a value in m3/D and for field units in STB/D.
*STG-R
Gas phase rate at surface conditions for a layer. A positive number entered
flow is from reservoir to well bore. A negative number implies back flowing
layer. For SI units enter a value in m3/D and for field units in scf/D.
*STS-R
Solvent phase rate at surface conditions for a layer. A positive number entered
layer. For SI units enter a value in m3/D and for field units in scf/D.
User's Guide IMEX
*STI-R
Intermediate stream rate at surface conditions for a layer. A positive number
entered for a layer belonging to an injection well implies flow is expected to be
from well bore to reservoir, that is the well layer is injecting into the reservoir.
A positive number entered for a layer belonging to a production well implies
layer. For SI units enter a value in m3/D and for field units in STB/D. Not
valid for the IMEX simulator.
*STO-CP
Oil cumulative production at surface conditions. For SI units enter a value in
m3 and for field units in STB.
*STW-RP
Water production rate at surface conditions. For SI units enter a value in
m3/D and for field units in STB/D.
*STW-CP
Water cumulative production at surface conditions. For SI units enter a value
in m3 and for field units in STB.
*STW-RI
Water injection rate at surface conditions. For SI units enter a value in m3/D
and for field units in STB/D.
*STW-CI
Water cumulative injection at surface conditions. For SI units enter a value
in m3 and for field units in STB.
*STG-RP
Gas production rate at surface conditions. For SI units enter a value in m3/D
and for field units in scf/D.
*STG-CP
Gas cumulative production at surface conditions. For SI units enter a value
in m3 and for field units in scf.
*STG-RI
Gas injection rate at surface conditions. For SI units enter a value in m3/D
and for field units in scf/D.
*STG-CI
Gas cumulative injection at surface conditions. For SI units enter a value in
m3 and for field units in scf.
User's Guide IMEX
*STI-RP
Intermediate liquid stream production rate at surface conditions. Valid for GEM
simulator only. For SI units enter a value in m3/D and for field units in STB/D.
*STI-CP
Intermediate liquid stream cumulative production at surface conditions. Valid
for the GEM simulator only. For SI units enter a value in m3 and for field
units in STB.
*WTG-RP
Wet gas stream production rate at surface conditions. Valid for the GEM
simulator only. For SI units enter a value in m3/D and for field units in scf/D.
*WTG-CP
Wet gas stream cumulative production at surface conditions. Valid for the
GEM simulator only. For SI units enter a value in m3 and for field units in scf.
*STL-RP
Liquid (oil + water) stream production rate at surface conditions. For SI units
enter a value in m3/D and for field units in STB/D.
*STL-CP
Liquid (oil + water) stream cumulative production at surface conditions. For
SI units enter a value in m3 and for field units in STB.
*BHF-RP
For GEM and STARS, the oil plus water plus gas phase production rate at
bottom hole or reservoir conditions. For IMEX, the oil plus water plus gas
plus solvent phase production rate at reservoir conditions. For SI units enter a
value in reservoir m3/D and for field units in reservoir barrels BBL/D.
*BHF-CP
For GEM and STARS, the oil plus water plus gas phase production
cumulative at bottom hole or reservoir conditions. For IMEX, the oil plus
water plus gas plus solvent phase production cumulative at reservoir
conditions. For SI units enter a value in reservoir m3 and for field units in
reservoir barrels BBL.
*BHF-RI
For GEM and STARS, the oil plus water plus gas phase and for IMEX, the
oil plus water plus gas plus solvent phase injection rate at bottom hole or
reservoir conditions. For SI units enter a value in reservoir m3/D and for
field units in reservoir barrels BBL/D.
User's Guide IMEX
*BHF-CI
For GEM and STARS, the oil plus water plus gas phase and for IMEX, the
oil plus water plus gas plus solvent phase injection cumulative at bottom hole
or reservoir conditions. For SI units enter a value in reservoir m3 and for
field units in reservoir barrels BBL.
*STS-RP
Solvent production rate at surface conditions. For SI units enter a value in
m3/D and in field units in scf/D. Valid for the IMEX simulator only.
*STS-CP
Solvent cumulative production at surface conditions. For SI units enter a
value in m3 and in field units in scf. Valid for the IMEX simulator only.
*STS-RI
Solvent injection rate at surface conditions. For SI units enter a value in
m3/D and in field units in scf/D. Valid for the IMEX simulator only.
*STS-CI
Solvent cumulative injection at surface conditions. For SI units enter a value
in m3 and in field units in scf. Valid for the IMEX simulator only.
*GOR
Gas oil ratio at surface conditions. For SI units enter a value as surface m3/D
of gas production per m3/D of surface oil production and for field units in scf
of gas per STB of oil.
*WCUT
Water cut at surface conditions. The water cut is the ratio of surface production
of water divided by the total surface liquid or water + oil production. For SI
units enter a value as surface water production rate (m3/D) divided by the sum
of surface production rates of water and oil (m3/D) and field units in STB/D of
water per STB/D of water + oil production.
*WGR
Ratio of water production rate at surface conditions divided by the surface
gas production rate. For SI units enter a value as surface water production
rate (m3/D) divided by the surface production rates of gas (m3/D) and field
units in STB/D of water per scf/D of gas production.
*GLR
Ratio of gas production rate at surface conditions divided by the surface
liquid (sum of water + oil phases) production rate. For SI units enter a value
as gas production rate (m3/D) divided by the sum of surface production rates
of water and oil (m3/D) and field units in scf/D of gas per STB/D of water +
oil production.
User's Guide IMEX
*Mxx
Mole percent of component m in the well stream. Mole percent is calculated
as the molar rate of component m flowing into the well divided by the total
molar hydrocarbon flow rate multiplied by 100. The total rate is the sum of all
hydrocarbon components (does not include water). Enter the component
number after the letter M. For example for component number 5 in the
components list (as determined by the order that components are specified with
*COMPNAME keyword) use *M5 <value>. Not valid for the IMEX simulator.
Note: Currently *MPWS *Mxx
*GASMxx
Mole percent of component m in the surface gas stream. Mole percent is
calculated as the molar rate of component m in the separator gas stream
divided by the total molar hydrocarbon flow rate of the separator gas stream
multiplied by 100. The total rate is the sum of all hydrocarbon components (does
not include water). For example for component number 5 in the components list
(as determined by the order that components are specified with *COMPNAME
keyword) use *GASM5 <value>. Not valid for the IMEX simulator.
Note: Currently implemented (wmtrin) *MPVS *Mxx
*LIQMxx
Mole percent of component m in the separator liquid stream. If there is no
intermediate liquid stream, then the liquid stream equals the surface oil stream,
otherwise the liquid stream includes the oil and the intermediate liquid
streams. Mole percent is calculated as the molar rate of component m in the
separator liquid stream divided by the total molar flow rate of the separator
liquid stream multiplied by 100. The total rate is the sum of all hydrocarbon
components (does not include water). Enter the component number after the
letter M. For example for component number 5 in the components list (as
determined by the order that components are specified with *COMPNAME
keyword) use *LIQM5 <value>. Not valid for the IMEX simulator.
Note: Currently implemented (wmtrin) *MPLS *Mxx
*BHP
Bottom hole pressure of the well. Enter a value in kPa for SI units and psi for
field units.
*WHP
Tubing head pressure of the well. Enter a value in kPa for SI units and psi for
field units.
User's Guide IMEX
*GWGR
Ratio of gas production rate at surface conditions divided by the wet gas
production rate at surface conditions. For SI units enter a value as surface gas
production rate (m3/D) divided by the surface production rate of wet gas
(m3/D) and field units in scf/D of gas per scf/D of wet gas production. Not
*WWGR
Ratio of water production rate at surface conditions divided by the wet gas
production rate at surface conditions. For SI units enter a value as water
production rate (m3/D) divided by the wet gas surface production rate
(m3/D) and field units in STB/D of water per scf/D of wet gas production.
Not valid for the IMEX simulator.
*BKFLOW
If any layer of the well is back flowing then this condition is satisfied. Note:
To specify the case where action(s) are to be taken if the well IS backflowing
use the *BKFLOW keyword followed by the greater than operator and a
value of 0. Alternatively to specify the case where the action(s) are to be
taken if the well IS NOT back flowing use the *BKFLOW keyword followed
by the less than operator and a value of 0. If an increment is specified with
the *INCREMENT keyword, then that value will be ignored.
*BKFLOW > 0 take actions if the well IS BACK FLOWING
*BKFLOW < 0 take actions if the well IS NOT BACK FLOWING
*GLR
Ratio of gas production rate at surface conditions divided by the surface
liquid (sum of water + oil phases) production rate. For SI units enter a value
as gas production rate (m3/D) divided by the sum of surface production rates
of water and oil (m3/D) and in field units in scf/D of gas per STB/D of water
plus oil production.
*O2CONC
Maximum oxygen concentration of all grid blocks where a production well
has been completed. For SI units enter a value as oxygen
concentration/density as kg/m3 and in field units in lbm/ft3. Valid for the
STARS simulator alone.
*TEMP
Maximum temperature of all grid blocks where a well has been completed.
For SI units enter a value in degrees Celsius and in field units in degrees
Fahrenheit. Valid for the STARS simulator alone.
User's Guide IMEX
*RECYSTG
Minimum group gas recycling rate. Group injection rate is calculated by
summing the injection rates at surface conditions of all cycling gas injectors
that belong to the group. For SI units enter a value in m3/D and in field units
in scf/D.
*RECYSTW
Minimum group water recycling rate. Group injection rate is calculated by
summing the injection rates at surface conditions of all cycling water
injectors that belong to the group. For SI units enter a value in m3/D and in
field units in STB/D.
*DWN
Difference in absolute value well layer pressure minus the pressure of well
block where well layer is completed. Enter a value in kPa for SI units and psi
for field units.
*PAVE
Sector average pressure calculated on the basis of total pore volume. For SI
units enter a value in kPa and for field units in psi.
*SOAVE
Sector average oil saturation calculated on the basis of total pore volume.
*SWAVE
Sector average water saturation calculated on the basis of total pore volume.
*SGAVE
Sector average gas saturation calculated on the basis of total pore volume.
*STOIP
Sector oil in place at standard conditions. For SI units enter a value in m3
and in field units in STB.
*STWIP
Sector water in place at standard conditions. For SI units enter a value in m3
and in field units in STB.
*STFGIP
Sector gas in place based on reservoir gas phase alone at standard conditions.
For SI units enter a value in m3 and in field units in scf.
*STGIP
Sector gas in place based on both reservoir gas and reservoir oil phase at
standard conditions. For SI units enter a value in m3 and in field units in scf.
User's Guide IMEX
*TIMSIM
Condition based on time elapsed since the beginning of the simulation in
days.
*TRELTD
Condition based on time elapsed relative to the time when the trigger is
defined, in days.
DEFAULTS:
*TRIGGER is an optional keyword.

CONDITIONS:
The entire *TRIGGER keyword must appear on one line followed by a list of actions
and then by the *END_TRIGGER.
EXPLANATION:
This keyword allows for certain actions to be implemented when a specified condition or trigger
is satisfied. The actions are specified in the form of a block of valid keywords encapsulated
within the *TRIGGER and *END_TRIGGER keywords. With the exception of the following
keywords: *DATE, *TIME, *REFINE, *AMALGAMATE, *DEALMAGAMATE,
*DEREFINE, *REREFINE, *AMALGAMATE-C, *DEREFINE-C, *AIMSET, *TRANSI,
*TRANSJ, *TRANSK, *TRANLI, *TRANLJ and *TRANLK, *RANGE, *TRANSIJ+,
*TRANSIJ-, *TRANSIK-, *TRANSIK+, *TRANSIENT, *THTYPE, any other valid keywords
that can otherwise be used in the WELL and RECURRENT data section of a GEM or IMEX
input data file can be used with triggers.
Note: some of the above keywords exceptions are not valid with the IMEX simulator.
The trigger condition of each active trigger is tested at the bottom (end) of the timestep (that
is after the timestep is considered to be completed, the cumulatives have been updated and
the simulation is ready to proceed to the next timestep). If the trigger condition is satisfied
then the list of associated actions are processed at the top (beginning) of the next timestep.
triggers are defined. The triggers are not sorted in any other way. If nesting is used (that is a
trigger is defined within another trigger) then the inner trigger is defined or becomes active
only if the outer trigger condition is satisfied such nesting may impact on the order that
triggers appear in the trigger list.
Once a trigger condition is met, the count of the number of times remaining that the trigger
can be satisfied is reduced by 1. If for example the default value of 1 is used for the
maximum number of times the trigger can be satisfied, then the count remaining is 0 and the
trigger is removed from the list of active triggers. The trigger is also removed from the active
trigger list once the trigger condition has been tested for the maximum number of times as
specified with the *TEST_TIMES keyword.
If a well change time (as specified with *DATE or *TIME card) coincides with the time that
the trigger actions are to be processed, then the trigger actions are implemented FIRST
followed by the list of actions or keywords following the *DATE or *TIME card.
User's Guide IMEX
keyword of the outer trigger). This offers great flexibility, however such nesting should be
properly thought out to avoid unintended consequences. For example it is possible to define a
trigger recursively using nesting. That is the inner and outer triggers can be assigned the
same name. In this case when the outer trigger condition is first satisfied, the trigger redefines
itself based on the inner trigger definition. For example consider the following block of
keywords in the well and recurrent data section of a GEM or IMEX data file.
*TIME 0.0
*WELL well1
*OPERATE *STO 750.0
*WELL well2
*OPERATE *STO 250.0
*WELL well3
*OPERATE **STO 250.0
*SHUTIN well2 well3
*TIME 50.0 **Time when Trigger is defined
*TRIGGER trig1 *ON_WELL well1 *STO-RP < 500.0 *APPLY_TIMES 3
*INCREMENT -25.0
** outer trigger
*WTMULT well2 *STO 1.25
*TRIGGER trig1 *ON_WELL well1 *GOR > 4500.0
**inner trigger, same name
*OPEN well3
*END_TRIGGER
*END_TRIGGER
At time equal to 50 days, the outer trigger is defined. At the end of the next and subsequent
timesteps the trigger condition is tested until it is satisfied or the simulation steps. Suppose at
time equal to 65 days the oil production rate for well well1 drops to below 500 STB/D. The
trigger condition is satisfied, well 2 is opened with an initial oil rate target of 250 *1.25 or
312.50. However because the inner trigger has the same name as the outer trigger, the trigger
trig1 is also redefined at this time according to the inner definition. The maximum repeat
count of 3 as well as the increment to the trigger condition of -25 STB/D specified with the
outer trigger definition is redundant and can never be used. After 65 days the new trigger
condition is based on GOR and if satisfied will open well 3 with an initial target of 250 STB/D.
The list of actions (keywords) appearing within a trigger definition are not validated in any
way (that is checked for syntax or consistency with previous well and recurrent data etc) until
the trigger condition is satisfied. In *CHECKONLY mode however each line of the well and
recurrent data including keywords (actions) within the trigger definition are processed.
User's Guide IMEX
With the implementation of trigger actions GEM restart runs are now handled differently.
With previous versions of GEM, well management information required for a restart run was
synthesized from both data saved to a restart file and from parsing the recurrent data to the
last date/ time card prior to restart time. With the new version of GEM all information
required for restart is written and read from the restart file. No recurrent data is read prior to
restart time. Therefore whereas it was previously possible to have certain keywords (actions)
take effect immediately upon restart by changing the recurrent data prior to the restart time,
this is no longer possible.
The @ symbol in quotes may be used as a place holder for a list of wells or groups that have
satisfied the trigger condition. This list will be a subset of the list of wells or groups specified
as part of the trigger definition. For example consider the following trigger:
*TIME 50.0
*TRIGGER trig1 *ON_WELL pr* *GOR > 5000
*SHUTIN @
*END TRIGGER
Suppose production wells named pr1, pr2, pr3, pr4 that match the wild card well list pr* exist
prior to the time that trigger trig1 is defined, that is at 50 days. Then when 50 days of
simulation time has elapsed the GOR of each well pr1, pr2, pr3 and pr4 respectively will be
tested against the trigger value of 5000. Suppose that at some subsequent time, say 70 days,
wells pr1 and pr3 have GOR > 5000. In that case the trigger condition is satisfied and wells
pr1 and pr3 will be shut-in as directed, that is the *SHUTIN @ statement is equivalent to
the user having specified *SHUTIN pr1 pr3. **NOTE: When the @ place holder is used
ALL keyword(s) that PRECEDE the place holder must on the same line as the place holder.
Therefore for example
*TARGET *STO @
1.5
is valid, but
*TARGET *STO
@
1.5
or
*TARGET
*STO @
1.5
are not, since in the latter cases TARGET and STO and @ are not on the same line. Because
of the special significance of the @ symbol, please do not use this symbol as part of a well,
group, sector or trigger name.
User's Guide IMEX
Alter Well Constraint Value via a Multiplier (Optional)

*WTMULT
PURPOSE:
*WTMULT allows modification of any previously specified well constraint value for well(s)
listed by well_numbers or 'well_names' by applying a multiplying factor.
FORMAT:
*WTMULT
(*STO)
(*STW)
(*STG)
(*STL)
(*STI)
(*BHP)
(*WHP)
(*DWN)
(*DWA)
(*DWB)
(*BHF)
(*BHG)
(*BHW)
(*STR)
(*STS)
(*BHS)
(well_numbers)
Multiplier(s)
or
('well_names')
or
@
place holder
DEFINITIONS:
*STO
This subkeyword identifies a surface oil rate (m3/day | STB/day) constraint.
*STW
This subkeyword identifies a surface water rate (m3/day | STB/day)
constraint.
*STG
This subkeyword identifies a surface gas rate (m3/day | scf/day) constraint.
*STL
This subkeyword identifies a total surface liquid rate (oil+water) (m3/day |
STB/day) constraint.
*STI
This subkeyword identifies a surface intermediate liquid rate (m3/day |
STB/day) constraint. Please see the manual entry for the *WELSEP keyword
for more information about the intermediate liquid stream. Not Valid for the
IMEX simulator.
User's Guide IMEX
*BHF
solvent) (m3/day | STB/day) constraint. This constraint is applicable to
producers only.
*BHW
This subkeyword identifies a reservoir water rate (m3/day | STB/day)
constraint.
*BHP
This subkeyword identifies a bottom hole pressure (kPa | psi) operating
constraint.
*WHP
This subkeyword identifies a well-head pressure (kPa | psi) constraint.
*DWN, *DWA, *DWB
This subkeyword identifies a draw-down pressure (kPa | psi) constraint.
*BHG
This subkeyword identifies a reservoir gas rate (m3/day | ft3/day) constraint.
This keyword is applicable to injectors only.
*STR
This subkeyword identifies a volume or molar percentage recycling rate
constraint. This constraint is applicable to cycling injectors only. Not Valid
for the IMEX simulator.
*STS
This subkeyword identifies a surface solvent rate (m3/day | scf/day)
constraint. Valid for the IMEX simulator only.
*BHS
This subkeyword identifies a reservoir solvent rate (m3/day | ft3/day)
constraint. This keyword is applicable to injectors only. Valid for the IMEX
simulator only.
well_numbers
which this target applies. These well numbers must be on the same line as the
*WTMULT keyword. If more than one line is required to specify all the
wells, then the *WTMULT keyword must be repeated.
User's Guide IMEX
well_names
Any number of 'well_names' in quotes to specify the wells to which this target
applies. These names must be on the same line as the *WTMULT keyword. If
more than one line is required for the well list, then the *WTMULT keyword
must be repeated. Limited wild-carding is available for this list; please see the
explanation on the manual page for the *SHUTIN keyword.
@
When the @ symbol in quotes is specified in place of well numbers or well
list, then it is considered to be a place holder for the list of wells that have
satisfied the trigger condition. This list is necessarily a subset of the well list
input with the trigger keyword. The place holder should only be used in
conjunction with trigger actions. All keywords that precede the @ place
holder must appear on the same line as the @ place holder. Therefore
WTMULT and a keyword specifying a stream, such as STO must be on the
same line as the @ place holder. See the EXPLANATION section of
*TRIGGER keyword for examples of the use of @ place holder.
Multiplier(s)
One number for each well identified by well_numbers or 'well_names'
specifying a multiplying factor to be applied. The new value of the specified
constraint = old value of the constraint * the multiplier. Multipliers must
appear on one or more new lines immediately following the *WTMULT line.
Multipliers must NOT appear on the same line as the list of well names or
well numbers. A single multiplier can be applied to all wells in a well list.
DEFAULTS:

CONDITIONS:
*WTMULT must be located in the WELL AND RECURRENT DATA keyword group, and
may appear anywhere in this keyword group following the initial *OPERATE declaration.
EXPLANATION:
This optional keyword is used to alter a constraint value for wells without having to redefine
all of the additional operating constraints. It is an effective method of altering constraints
when performing a history match.
If a wells constraint is subsequently changed (by the use of an *ALTER keyword for
example), *WTMULT is not reapplied automatically. *WTMULT should be redefined if
required.
*WTMULT also opens a well if the well has been shut in by a previous action or if the well
has been initially defined as a shut in well. When *WTMULT is encountered in a data set, the
simulator checks if the altered constraint with the new value becomes the most restrictive
well constraint. If so, the well is switched to this new constraint type. If not, the new value is
entered for the constraint but the well is switched to (or continues to run on) the currently
most restrictive constraint.
User's Guide IMEX
The constraint type is specified by using one of *STO, *STG, *STW, *STL, *STI, *BHP,
*WHP, *DWN, *DWA, *DWB, *BHF, *BHW, *BHG, *STR *STS or *BHS. One of these
qualifiers must be present.
The *WTMULT keyword should NOT be used to define a new operating constraint. The
multiplier specified with the *WTMULT keyword will ONLY be applied if the constraint of
the type specified with *WTMULT already exists for the well based on earlier input using the
*OPERATE keyword.
Examples:
*PRODUCER 1
:
*WTMULT *STO ** This alters the target constraint of *STO
1
** based on the *OPERATE keyword above, i.e.,
1.5
** from 500 to 500 *1.5 = 750
The *WTMULT keyword may also look like this when several wells have been defined:
:
*PRODUCER 1
*PRODUCER 2
*PRODUCER 3
*INJECTOR 4
*MONITOR *MIN *STW 10.00 *SHUTIN
:
*TIME 1200.
** At a later date, want to adjust the target
**
well_numbers
*WTMULT *STO 1:2
** multipliers
2*1.25
**
well_number
*WTMULT *BHP 3
** multiplier ** BHP_new = 2500*0.25 = 500
0.25
**
well_number
*WTMULT *STW 4
** values
0.50
User's Guide IMEX
Group Production Constraints Multiplier (Optional)

*GCONPMULT
PURPOSE:
*GCONPMULT is used to modify existing group production target controls with the use of
multipliers.
FORMAT:
*GCONPMULT 'group_name_1' 'group_name_2' ... 'group_name_n'

or
*GCONPMULT @
(*STO)
(*STG)
(*STW)
(*STL)
(*STI)
(*WTG)
(*BHF)
(*MNP)
(*CPP)
(*STS)
(*VREP)
(*RECYCLE)
(*PMAINT)
multiplier
multiplier
DEFINITIONS:
'group_name_1', group_name_2, ... , group_name_n

Are the groups to which the following constraint multiplier applies.
*STO
This subkeyword identifies a surface oil rate (m3/day | STB/day) constraint.
Zero multipliers are allowed and will have the same effect as shutting in all
the wells connected to that group.
*STG
This subkeyword identifies a surface gas rate (m3/day | scf/day) constraint.
Zero multipliers are allowed and will have the same effect as shutting in all
the wells connected to that group.
*STW
This subkeyword identifies a surface water rate (m3/day | STB/day)
constraint. Zero rates are allowed and will have the same effect as shutting in
all the wells connected to that group.
User's Guide IMEX
*STL
This subkeyword identifies a surface liquid rate (oil + water + intermediate
liquid) (m3/day | STB/day) constraint. Zero rates are allowed and have the
same effect as shutting all the wells connected to the group.
*STI
This subkeyword identifies a surface intermediate liquid rate (m3/day |
STB/day) constraint. Zero rates are allowed and have the same effect as
shutting all the wells connected to the group. For more information on the
intermediate liquid stream please see the description of the *WELSEP
keyword in this manual. Not valid for the IMEX simulator.
*WTG
This subkeyword identifies a surface wet gas rate (m3/day | scf/day)
constraint. Zero rates are allowed and have the same effect as shutting all the
wells connected to the group. For more information on the wet gas stream
please see the description of the *WELSEP keyword in this manual. Not
*BHF
This subkeyword identifies a bottom hole fluid rate (m3/day | rBBL/day)
constraint. Zero rates are allowed and have the same effect as shutting all the
wells connected to the group.
*MNP
This subkeyword introduces a manifold pressure (kPa | psi) constraint. This
may only be applied if the listed groups have all had production specified as
going through a manifold with the *MANIFOLD keyword.
*CPP
This subkeyword introduces a compressor (surface) pressure (kPa | psi)
constraint. This may only be applied if the listed groups have all had production
specified as going through a manifold with the *MANIFOLD keyword.
*STS
This subkeyword identifies a surface solvent rate (m3/day | scf/day)
constraint. Zero multipliers are allowed and will have the same effect as
shutting in all the wells connected to that group. Valid for the IMEX
simulator only.
*VREP
This subkeyword identifies a voidage replacement production target. This
indicates that the production wells connected to this group produce an
amount of the bottom-hole fluid in proportion to the total bottom-hole fluid
injected into the reservoir by the injection wells connected to this group.
User's Guide IMEX
*RECYCLE
This subkeyword specifies a recycling production target. This indicates that
the production wells connected to this group produce such that the phase
injected by the injection wells connected to this group as specified by *GAS,
*WATER or *SOLVENT is reproduced (recycled) out of the reservoir.
*PMAINT
This subkeyword identifies that the group production rates shall be adjusted
so as to maintain the hydrocarbon volume weighted average pressure in a
particular region/sector at a desired level. The multiplier applies only to the
sector pressure target defined by *PMTARG.
@
When the @ symbol in quotes is specified in place of group names, then it is
considered to be a place holder for the list of groups that have satisfied the
trigger condition. This list is necessarily a subset of the group list input with the
trigger keyword. The place holder should only be used in conjunction with
trigger actions. The keyword *GCONPMULT and the @ place holder must
appear on the same line. See the EXPLANATION section of *TRIGGER
keyword for examples of the use of @ place holder.
Multiplier
Multiplier to be applied to the existing constraint value -- see above for units.
DEFAULTS:
Optional keyword.
CONDITIONS:
group together with the target constraint for the particular stream or control must be defined
using the *GCONP grp_name *TARGET keyword before the target constraint multipliers can
be applied. Do not use the *GCONPMULT keyword to define new group target constraints.
EXPLANATION:
*GCONPMULT is used to modify group production constraints which have previously been
defined using the *GCONP grp_name *TARGET keyword.
Example:
*GCONPMULT
Field
*STW
0.75
This directs the specified Field group stock tank water production to be modified such that
the new constraint value is 75% of the value specified with the last *GCONP Field
*TARGET *STW card.
User's Guide IMEX
Example:
*GCONPMULT Group-1
*MNP 1.25
This sets a manifold pressure target of 125 % of the value specified with the last *GCONP
Group-1 *TARGET *MNP card. This group must have had its production assigned
manifold treatment with the *MANIFOLD keyword. Group-1 must have wells directly
attached to it.
User's Guide IMEX
Group Injection Constraints Multipliers (Optional)
*GCONIMULT
PURPOSE:
*GCONIMULT is used to specify multipliers which modify existing group injection target
controls.
FORMAT:
*GCONIMULT 'group_name_1' 'group_name_2' ... 'group_name_n'

or
*GCONIMULT @
(*STG)
(*STW)
(*BHG)
(*BHW)
(*GMP)
(*WMP)
(*GCP)
(*WCP)
(*STS)
(*SMP)
(*SCP)
Multiplier
(*VREP)
(*GAS)
(*WATER)
(*SOLVENT)
(*GMKUP)
(*WMKUP)
(*SMKUP)
Multiplier
(*PMAINT)
(*GAS)
(*WATER)
(*SOLVENT)
Multiplier
DEFINITIONS:
'group_name_1', group_name_2, ... , group_name_n

Are the groups to which the following constraint multiplier(s) apply.
*STG
This subkeyword identifies a surface gas rate (m3/day | scf/day). Zero rates
are allowed and have the same effect as shutting in all the gas injection wells
connected to that group.
User's Guide IMEX
*STW
This subkeyword identifies a surface water rate (m3/day | STB/day). Zero
rates are allowed and have the same effect as shutting in all the water
injection wells connected to that group.
*BHG
This subkeyword identifies a reservoir gas rate (m3/day | scf/day). Zero rates
are allowed and have the same effect as shutting in all the gas injection wells
connected to that group.
*BHW
This subkeyword identifies a reservoir water rate (m3/day | STB/day). Zero
rates are allowed and have the same effect as shutting in all the water
injection wells connected to that group.
*GMP
This subkeyword identifies a gas manifold pressure (kPa | psi) injection
constraint. This subkeyword can only be entered if all of the listed groups
have had gas injection identified as going through a manifold with the
*MANIFOLD keyword.
*WMP
This subkeyword identifies a water manifold pressure (kPa | psi ) injection
have had water injection identified as going through a manifold with the
*MANIFOLD keyword.
*GCP
This subkeyword identifies a gas compressor (surface) pressure (kPa | psi)
injection constraint. This subkeyword can only be entered if all of the listed
groups have had gas injection identified as going through a manifold with the
*MANIFOLD keyword. Also, a hydraulics table for calculation of the gas
manifold-to-surface pressure drop must have been identified for all of the
listed groups with the *GPTABLE keyword.
*WCP
This subkeyword identifies a water surface pressure (kPa | psi) injection
have had water injection identified as going through a manifold with the
*MANIFOLD keyword. Also, a hydraulics table for calculation of the water
manifold-to-surface pressure drop must have been identified for all of the
listed groups with the *GPTABLE keyword.
*STS
This subkeyword identifies a surface solvent rate (m3/day | scf/day). Zero
rates are allowed and have the same effect as shutting in all the solvent
injection wells connected to that group. Valid for the IMEX simulator only.
User's Guide IMEX
*SMP
This subkeyword introduces a solvent manifold pressure (kPa | psi) injection
have had solvent injection identified as going through a manifold with the
*MANIFOLD keyword. Valid for the IMEX simulator only.
*SCP
This subkeyword introduces a solvent compressor (surface) pressure (kPa |
psi) injection constraint. This subkeyword can only be entered if all of the
listed groups have had solvent injection identified as going through a
calculation of the solvent manifold-to-surface pressure drop must have been
identified for all of the listed groups with the *GPTABLE keyword. Valid for
the IMEX simulator only.
*VREP
This subkeyword identifies a voidage fraction injection target. This indicates
that the injection wells connected to this group inject such that the voidage
created by the producers connected to this group is replaced. In this case
*GAS, *WATER or *SOLVENT specifies which phase is to be injected to
replace the voidage. If more than one phase is being injected to replace the
voidage then there must be one *VREP keyword for each phase. These
primary voidage replacement streams are handled independently. One makeup stream can be supplemented with *GMKUP, *WMKUP or *SMKUP to
meet a total voidage replacement fraction. One of *GAS, *WATER,
*SOLVENT, *GMKUP, *WMKUP or *SMKUP must be present for each
*VREP keyword.
*PMAINT
This subkeyword identifies that the group injection rates shall be adjusted so
as to maintain the hydrocarbon volume weighted average pressure in a
particular region/sector at a desired level. The multiplier applies only to the
sector pressure target defined by *PMTARG.
*GAS
Specifies that the gas phase is to be injected for voidage replacement, recycle
or sector pressure maintenance.
*WATER
Specifies that the water phase is to be injected for voidage replacement,
recycle or sector pressure maintenance.
*SOLVENT
Specifies that the solvent phase is to be injected for voidage replacement,
recycle or sector pressure maintenance. Valid for the IMEX simulator only.
User's Guide IMEX
*GMKUP
Specifies that gas phase is the make-up stream supplemented to meet the
*WMKUP
Specifies that water phase is the make-up stream supplemented to meet the
*SMKUP
Specifies that solvent phase is the make-up stream supplemented to meet the
@
When the @ symbol in quotes is specified in place of group names, then it is
considered to be a place holder for the list of groups that have satisfied the
trigger condition. This list is necessarily a subset of the group list input with
the trigger keyword. The place holder should only be used in conjunction with
trigger actions. The keyword *GCONIMULT and the @ place holder must
appear on the same line. See the EXPLANATION section of *TRIGGER
keyword for examples of the use of @ place holder.
Multiplier
Constraint multiplier value.
DEFAULTS:
Optional keyword.
CONDITIONS:
group must be defined, by appearing in the list directly following *GROUP in a *GROUP
line or after the *ATTACHTO keyword on a *GROUP line. The group target constraint for
the particular constraint type multiplier being specified with GCONIMULT must be also be
previously defined using the keyword *GCONI.
EXPLANATION:
*GCONIMULT is used to specify multipliers which can be used to modify existing injection
group constraint targets. *GCONIMULT can also be used to modify previously specified
voidage replacement targets.
Examples:
*GCONIMULT Group-1 Group-2
*STG 3.0
This resets stock tank gas injection targets to 300 % of the previously specified values for
Group-1 and Group-2. The stock tank gas injection target must be previously specified with a
data line such as:
*GCONI Group-1 Group-2 *TARGET *STG 5.555E+07
User's Guide IMEX
Example: This is an example using voidage replacement. The solvent voidage replacement
fraction is reduced by half whereas the gas voidage replacement fraction is increased by a
factor of 2.
*GCONIMULT 'Group1'
*VREP *SOLVENT 0.5
*GCONIMULT 'Group1'
*VREP *GAS
2.0
The voidage replacement targets must be previously specified with data lines such as:
*GCONI Group-1 Group-2 *VREP *SOLVENT 0.6
*GCONI Group-1 Group-2 *VREP *GAS 0.4
User's Guide IMEX
Define Reporting Group (Optional)
*REPORTING-GROUP
PURPOSE:
*REPORTING-GROUP allows the user to define a set of wells with differing membership
weights which has data reported just as for the hierarchical groups which are the basis for group
control (see the manual entry for *GROUP, *GCONP, and related keywords). No group
controls can be specified for reporting groups, but there are no restrictions upon well
membership in reporting groups. A well may be a member of an arbitrary number of reporting
groups, and a well can have any non-negative membership weight in a reporting group.
FORMAT:
*REPORTING-GROUP weight_list
reporting_group_name
well_list
DEFINITIONS:
reporting_group_name
A character string containing not more than 16 characters. The character
strings Default-Field and Default-Group are not allowed as the name of a
reporting group as they are reserved for internal use. Reporting group names
must be distinct from group names. If reporting_group_name has already
been used as the name of a reporting group, then the current instance of
*REPORTING-GROUP has the effect of redefining the named reporting
group.
well_list
A set of 'well_names' or well_numbers; see below. The reporting group will
include all wells in the list.
well_names
Any number of well names (in quotes) to specify the wells in the reporting
group. Limited wildcarding is available for the list of well names; please see
the manual page for the *SHUTIN keyword for an explanation of the
wildcard facility.
well_numbers
Any number of integers, or ranges of integers specifying the numbers of the
wells in the reporting group.
weight_list
Non-negative real numbers specifying the membership weights in the
reporting group of the wells in well_list. If weight_list contains only a single
number, that weight is applied to all of the wells in the list; otherwise the
number of entries in weight_list must equal the number of wells in the
well_list. The numbers in the weight_list must be real numbers with the
decimal point included, in order that the beginning of the weight list can be
distinguished from a continuation of a list of well numbers. Repeat counts
are allowed, e.g. 6*0.5.
User's Guide IMEX
DEFAULTS:
Optional keyword. If *REPORTING-GROUP does not appear in the data set, no reporting
groups exist during the simulation. All of the indicated elements in the format
(reporting_group_name, well_list, and weight_list) must be present.
CONDITIONS:
*DATE line. All wells appearing in the well list must already have been defined with
*WELL lines. The reporting_group_name must not already have been used (through the
*GROUP keyword) as a group name. The weights must be entered as non-negative real
numbers with the decimal point explicitly included. The number of weight values must either
be one, in which case the single value will be applied to all listed wells, or else must equal
exactly the number of wells listed. The well list and the weight list can be spread over
several lines; the weight list can begin immediately after (and on the same line as) the end of
the well list.
EXPLANATION:
The quantities displayed for the reporting group (e.g. cumulative amounts or rates) are
calculated as
Q(reporting group) = w(well)Q(well),
where the summation is over the wells in the well_list, w(well) is the weight value for the
particular well, and Q(well) is the quantitys value for the well. Note that w(well) is the weight
value as entered by the user; no automatic normalization is performed upon the weights.
Example:
*REPORTING-GROUP Producers PROD1 PROD2
1.
This establishes a reporting group with the name Producers consisting of the two wells
PROD1 and PROD2. The single weight value 1. applies to both wells in the list.
User's Guide IMEX
Well/Group On-time Fraction (Optional)
*ON-TIME
PURPOSE:
*ON-TIME specifies the fraction of time during which a well or group operates. It is useful
for history-match or prediction runs in order to obtain the full-flow conditions which
correspond to the instantaneous rates of the well or group but with cumulative flows
reflecting the down time.
FORMAT:
*ON-TIME
well_names | group_names
OTF_input
DEFINITIONS:
well_names
Any number of quoted well names to specify the wells whose on-time
fractions are defined in this command. These names must be on the same line
as the *ON-TIME keyword. If more well names are to be specified than can
fit on one line, then another *ON-TIME keyword must be entered. Limited
wildcarding is available for the list of well names; please see the manual
page for the *SHUTIN keyword for an explanation of the wildcard facility.
For backward compatibility, well_numbers are still allowed, but they cannot
be mixed with the names in the same list.
group_names
A list of group_names to which the on-time fractions are applied. No
wildcard characters may be used in the group names. These names must be
on the same line as the *ON-TIME keyword. If more group names are to be
specified than can fit on one line, then another *ON-TIME keyword must be
entered. group_names cannot be mixed with well_names in the same list.
OTF_input
The values for the on-time fractions to be assigned to the wells/groups
specified. These values must appear on a (or several) new lines immediately
after the line which begins with the *ON-TIME keyword. One fraction must
be entered for each well/group in the list and the OTF_input are assigned to
the wells/groups in the list in order. The n*value format may be used to enter
multiple occurrences of the same fraction. Values must be between 0.001 and
1.0, inclusive. Values less than 0.001 are modified to 0.001 with notification.
DEFAULTS:
If no *ON-TIME keyword is encountered, on-time fractions of 1 are assigned to each

well/group, and the simulation is not affected. Wells/groups never listed under *ON-TIME
retain the on-time fraction of 1 throughout the simulation.
User's Guide IMEX
CONDITIONS:
This keyword must be located in the Well and Recurrent Data keyword group. A well/group
must be defined before it can be assigned options under *ON-TIME. The on-time fraction for
a given well/group may be altered by an additional *ON-TIME line at a subsequent well
change.
EXPLANATION:
Wells and groups are treated in an entirely analogue manner with wells at the lowest level.
The user input on-time factor (OTF_input) for a particular well/group is the on-time fraction
relative to the on-time of its parent group. For the field (top level) group, it is relative to the
total simulated time. The actual (absolute) on-time factor (OTF_actual), computed internally,
is the product of OTF_input's from itself and all the superior groups along the hierarchy
branch (up to the field). If no group structure ever exists, the well on-time fractions are
relative to the conceptual Default-Field, i.e. the total simulated time (and thus OTF_input =
OTF_actual). In normal cases, the on-time factors are assigned on only one level of the
well/group hierarchy.
The instantaneous rate (Q_instant) for a well refers to the rate when the subject well is
running on its full-flow conditions (bhp and whp, et al.). The instantaneous rate for a group,
by definition, is the summation of the on-time averaged rates from all its child groups or
wells directly attached, which takes into account the lower-level down-time effect. The ontime averaged rate (Q_average) for a well/group is the relative mean rate over the period
while its parent group is operating. The actual rate (Q_actual), resulting directly from the
material balance, is the reported true volume produced from or injected into the reservoir per
unit time. It can also be interpreted as the absolute on-time average, and is used to calculate
all the cumulative flows.
At any well/group level, the following relationships hold:
Q_average = Q_instant * OTF_input
Q_actual = Q_instant * OTF_actual
(relative)
(absolute)
For any well/group, the instantaneous rate and the on-time averaged rate are the same if
OTF_input = 1, since this well/group has no down-time relative to its parent group. This
does not necessarily mean that the actual rate equals the instantaneous rate as well.
Theoretically, these three rates might have different values. In case that all group on-time
fractions are unity (well on-time fractions may not be unity), one expects
OTF_actual = OTF_input
Q_actual = Q_instant * OTF_actual
(absolute)
for all levels. Moreover, Q_actual = Q_instant, for all groups (but not the wells). This is the
settings in the earlier versions of CMGs Well Management before incorporating the group
on-time fraction. It should be noted that these equivalences hold true only at the field (top)
level once the group on-time fractions are set < 1.0.
User's Guide IMEX
Any flow constraints, targets and limits specified for a well (e.g. *OPERATE, *MONITOR)
or group (e.g. *GCONP / *GCONI, *GCONM) will apply to its instantaneous rate. For wells,
values of the bottom-hole pressure, drawdown, well layer pressure, and well head pressures
which correspond to the specified (instantaneous) rate values are reported. When the well is
operating on a bottom-hole pressure or well-head pressure constraint, the rate which results is
the actual rate, which has been reduced by the actual on-time fraction. The group on-time
factors are meant to model situations in which all wells in a group are brought down at the
same time and restored to service at the same time; however, wells are still allowed to have
individually a non-unit on-time factor when the group factors are in force.
Example:
Assign on-time fractions of 0.95 to wells 1, 3, 7, 8, 9, and 10.
*ON-TIME
1 3 7:10
6*0.95
A more sophisticated example is to suppose the following group-well hierarchy:
The field is assigned an oil target (*GCONP/*TARGET) of 500 which is to be apportioned

using the Guide Rate method (*APPOR-METHOD/*GUIDE) with supplied guide rates:
*GUIDEP *STO
3.
'GRP-2' 'GRP-3' 'GRP-201' 'GRP-202' 'GRP-203'

2.
1.
2.
3.
In a very general (although not normal) case, the group on-time fractions are input as
*ON-TIME FIELD GRP-2 GRP-3 GRP-201 GRP-202 GRP-203
0.75
0.8
0.8
0.85
0.85
0.85
which specifies that: Field operates at 75% of the total simulated time; groups GRP-2 and
GRP-3 (level 2) operate at 80% of the time while Field is on-time; groups GRP-201, 202, 203
(level 3) operate at 85% of the time while their parent group GRP-2 is operating. The on-time
factors for all wells default to unity (OFT_input = 1), suggesting that they operate (100%) as
long as their parent groups are operating.
Should all the well constraints have not been violated and the group target has been met, the
actual on-time fractions and various flow rates at different levels computed by the simulator
read:
User's Guide IMEX
On-Time Factor (OTF)

Group Hierarchy
FIELD
GRP-2
GRP-201
GRP-202
GRP-203
GRP-3
Default-Group
Input
0.75
0.80
0.85
0.85
0.85
0.80
1.00
Actual
0.75
0.60
0.51
0.51
0.51
0.60
0.75
Rates (Q)
Instant.
[500.0]
375.0
73.5
147.0
220.5
250.0
/
Average
375.0
300.0
62.5
125.0
187.5
200.0
/
Actual
375.0
225.0
37.5
75.0
112.5
150.0
/
[--]: targeted instantaneous value
All wells under group GRP-2 have the actual on-time fractions of 0.51, and those for wells
under group GRP-3 are 0.6. The field instantaneous rate by which the target is set is the
summation of the on-time averaged rates of its child groups GRP-2 and GRP-3, which can also
be converted directly from the field actual rate (500.0 = 300.0 + 200.0 = 375.0/0.75). Similarly,
summing the on-time averaged rates of groups GRP-201,202,203 yields the instantaneous rate
of their parent group GRP-2, which can also be converted from GRP-2s actual rate by dividing
it with GRP-2s actual on-time fraction (375.0 = 62.5 + 125.0 + 187.5 = 225.0/0.6).
As far as the behaviors of lower level groups (and wells) are concerned, the above dataset
would run exactly the same if the user directly (and only) define OTF_input = 0.51 for group
GRP-2 and OTF_input = 0.6 for group GRP-3. The actual rates and cumulatives for all the
wells/groups remain unchanged. However, since now the Field has a unity on-time fraction,
i.e. the instantaneous rate is indeed the actual rate for the field, one needs to adjust the field
target to 375.0 to render a fair comparison.
On-Time Factor (OTF)
Group Hierarchy
FIELD
GRP-2
GRP-201
GRP-202
GRP-203
GRP-3
Default-Group
Input
1.00
0.51
1.00
1.00
1.00
0.60
1.00
Actual
1.00
0.51
0.51
0.51
0.51
0.60
1.00
Rates (Q)
Instant.
[375.0]
441.0
73.5
147.0
220.5
250.0
/
Average
375.0
225.0
73.5
147.0
220.5
150.0
/
Actual
375.0
225.0
37.5
75.0
112.5
150.0
/
[--]: targeted instantaneous value
User's Guide IMEX
Hydraulic Pressure Table (Conditional)
*PTUBE1, *ITUBE1
PURPOSE:
*PTUBE1 or *ITUBE1 introduces input of tubing-head (or outlet) pressure data tables for a
production or injection well (or manifold group). The tabulated variable is the bottom-hole
(or inlet) pressure, with tubing-head (or outlet) pressure and several independent variables
such as surface rates or rate ratios.
FORMAT:
*PTUBE1 / *ITUBE1
table_number
*DEPTH refdepth
Flo
flo(1) ... flor(nflo)
Gfr
gfrl(1) ... gfrl(ngfr)
Wfr
wfr(1) ... wfr(nwfr)
Add
add(1) ... add(nadd)
*WHP
whp(1) ... whp(nwhp)
*BHP
** iflo
:
igfr
:
iwfr
:
iadd
:
bhp(1)...
:
bhp(nwhp)
:
*DEPTH
This subkeyword introduces input of the reference depth (m | ft | cm).
Flo
Independent Flow Rate variable. All rates are measured at surface conditions.
For *PTUBE1, it is identified by one of the following subkeywords:
User's Guide IMEX
*OIL
Oil production rate (m3/day | stb/day | cm3/min);
*LIQ
Liquid (oil+water) production rate (m3/day | stb/day |

cm3/min);
*GAS
Gas production rate (m3/day | scf/day | cm3/min);
For *ITUBE1, it is identified by one of the following subkeywords:

*GAS
Gas injection rate (m3/days | scf/day | cm3/min);
*WAT
Water injection rate (m3/day | stb/day | cm3/min);
*SOL
Solvent injection rate (m3/day | scf/day | cm3/min);
Gfr
Independent Gas Fraction variable (*PTUBE1 only), identified by one of the
following subkeywords:
*GOR
Gas-oil ratio (m3/m3 | scf/stb | cm3/cm3);
*GLR
Gas-liquid ratio (m3/m3 | scf/stb | cm3/cm3);
*OGR
Oil-gas ratio (m3/m3 | stb/scf | cm3/cm3);
Wfr
Independent Water Fraction variable (*PTUBE1 only), identified by one of
the following subkeywords:
*WOR
Water-oil ratio,
*WCUT
Water cut (water-liquid ratio);
*WGR
Water-gas ratio (m3/m3 | bbl/scf | cm3/cm3);
Add
Independent Additional look-up variable (*PTUBE1 only), identified by one
of the following subkeywords:
*LFG
Lift gas injection rate (m3/day | scf/day | cm3/min);
*LFR
Injection gas-liquid ratio for gas lift (m3/m3 | scf/stb |

cm3/cm3);
*LFT
Total gas-liquid ratio for gas lift (m3/m3 | scf/stb | cm3/cm3);
*DNO
Surface mass density of oil (kg/m3 | lb/ft3 | g/cm3). It takes

the volumetric-weighted average value for a manifold group;
*DNG
Surface mass density of gas (kg/m3 | lb/ft3 | g/cm3). It takes

the volumetric-weighted average value for a manifold group;
*WHP
This keyword specifies the well-head (or manifold outlet) pressures (kPa | psi
| kg/cm2).
*BHP
This keyword introduces input of the bottom-hole (or manifold inlet)
pressure table (kPa | psi | kg/cm2).
refdepth
Non-zero real number specifying the reference depth for the pressure table.
User's Guide IMEX
table_number
Hydraulic table number. Tables must be defined in order starting with table 1
and increasing sequentially.
flo(1) ... flo(nflo)
gfr(1) ... gfr(ngfr)
wfr(1) ... wfr(nwfr)
add(1) ... add(nadd)
whp(1) ... whp(nwhp)
Real numbers specifying the values of the first through the last entry of the
corresponding variable. At least one (1) entry must be entered for each
identified variable. Values must be monotonically increasing, respectively.
Gfr, Wfr and Add cannot be entered for *ITUBE1.
iflo, igfr, iwfr, iadd
Integers specifying the corresponding variable index. They must be in the
fixed order for *PTUBE1 and only iflo shall be entered for *ITUBE1 when
reading the *BHP table.
bhp(1) ... bhp(nwhp)
Real numbers specifying the values of bottom-hole (or manifold inlet) pressure
corresponding to the variable values specified by the integer indices iflo, etc.
and well-head (or manifold outlet) pressure values (1, ..., nwhp). If one or
several of the BHP values is not available (for example, if the program used to
compute the BHP values did not converge for these conditions), the keyword
*EXTP may be entered in place of the value. In this case the missing values in
a row of the table are determined by linear interpolation or extrapolation from
entered values in the same row. At least one BHP value (as opposed to the
string *EXTP) must be entered in each row of the table.
DEFAULTS:
Conditional keyword for wells but optional for manifold groups. No defaults.
CONDITIONS:
Required when *WHP is one of the well constraints or when the well bore model is desired,
and the TABLE is referred to. All the information must be entered before the subkeyword
*BHP for reading the pressure table. All the variables, except for the *LFG, *LFR and *LFT
in the Add variable item, apply to the manifold groups.
EXPLANATION:
*PTUBE1 or *ITUBE1 sets up the independent variable and WHP values for the *BHP tables
for wells (or manifold groups). The bottom-hole (or manifold inlet) pressure table is also
entered here (see related keywords *IWELLBORE, *PWELLBORE, *GLCONTROL and
*GPTABLE). When *GLIFT is in effect, the lift gas injection rate is added to the producers
formation gas rate before calculating the wellbore pressure drop and well liquid production
rates, unless the *LFG, *LFR or *LFT is specified as the Add variable in *PTUBE1.
User's Guide IMEX
The depth required with *PTUBE1/*ITUBE1 is a well (or a manifold) depth for an individual
well (or manifold, see keyword *GPHYDDEP). The depth is used to adjust the tabulated
bottom-hole (or manifold inlet) pressures from the table depth to the actual well (or manifold
outlet) depth using linear scaling. In this way several wells (or groups) with different depths may
use the same hydraulics pressure table. Total well length should be entered for horizontal tables.
The bottom-hole (or manifold) pressures read under *PTUBE1/*ITUBE1 correspond to values
of the defined variables (rates, surface ratios, water-cuts, etc.). The number of bottom-hole (or
manifold inlet) pressures entered for each set of index integers must be equal to the number of
well-head (or manifold outlet) pressures entered under *WHP.
For instance, if three well-head pressures were entered, then there must be four columns (iflo,
igfr, iwfr, iadd) containing bottom-hole pressures for *PTUBE1 or one column (iflo) for
*ITUBE1. The integer indices identify the particular values of the defined variables among
those entered, and must be in the fixed order. The first entered BHP value corresponds to the
first WHP value, and so on.
Example 1: Oil Production
*PTUBE1 1
*DEPTH 5000.0
*OIL
**flo(1)
0.0
*GOR
**gfr(1)
500.0
*WCUT **wfr(1)
0.00
*ALQ
**add(1)
0.0
*WHP
**whp(1)
200.0
*BHP
**iflo igfr iwfr
1
1
1
1
1
2
1
2
1
1
2
2
2
1
1
2
1
2
2
2
1
2
2
2
3
1
1
3
1
2
3
2
1
3
2
2
4
1
1
4
1
2
4
2
1
4
2
2
flo(2)
4000.0
gfr(2)
1000.0
wfr(2)
0.50
flo(3)
8000.0
whp(2)
900.0
whp(3)
1600.0
iadd
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
bhp(1)
2176.1
2646.7
1997.7
2447.7
2135.5
2618.0
1682.7
2189.0
2133.6
2630.9
1463.1
2022.0
2160.1
2696.4
1425.7
2080.0
flo(4)
16000.0
bhp(2)
2873.7
3344.7
2670.9
3124.7
2876.6
3351.2
2674.6
3132.3
2884.2
3368.4
2684.5
3152.8
2912.5
3433.4
2721.3
3231.0
bhp(3)
3573.7
4044.7
3370.9
3824.7
3576.6
4051.2
3374.6
3832.3
3584.2
4068.4
3384.5
3852.8
3612.5
4133.4
3421.3
3931.0
User's Guide IMEX
Example 2: Water Injection

*ITUBE1 1
*DEPTH 1800.0
*WAT
** flo(1) flo(2) flo(3) flo(4) flo(5) flo(6)
0.0
100.0 200.0 350.0 500.0 600.0
*WHP
**whp(1)
whp(2)
whp(3)
whp(4)
whp(5)
101.325 10000.0 20000.0 30000.0 40000.0
*BHP
**iflo bhp(1) bhp(2) bhp(3) bhp(4) bhp(5)
1
17820.0 27794.0 37872.0 47949.0 58026.0
2
17816.0 27791.0 37868.0 47945.0 58022.0
3
17813.0 27788.0 37865.0 47942.0 58019.0
4
17807.0 27782.0 37859.0 47936.0 58013.0
5
17802.0 27777.0 37854.0 47931.0 58008.0
6
17798.0 27773.0 37850.0 47927.0 58004.0
Example 3: Gas Production in a Gas-Water run

*PTUBE1 1
*DEPTH 5000.0
*GAS
** flo(1) flo(2)
flo(3)
flo(4)
0.0
40000.0 80000.0 160000.0
*GOR
** gfr(1)
0.0
*WGR
** wfr(1) wfr(2)
0.00
0.01
*ALQ
** add(1)
0.0
*WHP
** whp(1) whp(2) whp(3)
200.0
900.0
1600.0
*BHP
**iflo igfr iwfr iadd bhp(1) bhp(2)
1
1
1
1
2176.1 2873.7
1
1
2
1
2646.7 3344.7
2
1
1
1
2135.5 2876.6
2
1
2
1
2618.0 3351.2
3
1
1
1
2133.6 2884.2
3
1
2
1
2630.9 3368.4
4
1
1
1
2160.1 2912.5
4
1
2
1
2696.4 3433.4
User's Guide IMEX
bhp(3)
3573.7
4044.7
3576.6
4051.2
3584.2
4068.4
3612.5
4133.4
Example 4: Oil Production with extrapolation

*PTUBE1 1
*DEPTH 5000.0
*OIL
** flo(1) flo(2) flo(3) flo(4)
0.0
4000.0 8000.0 16000.0
*GOR
** gfr(1) gfr(2)
500.0
1000.0
*WCUT
** wfr(1) wfr(2)
0.00
0.50
*ALQ
** add(1)
0.00
*WHP
** whp(1) whp(2) whp(3)
200.0
900.0
1600.0
*BHP
**iflo igfr iwfr iadd bhp(1) bhp(2)
1
1
1
1
2176.1 2873.7
1
1
2
1
2646.7 3344.7
1
2
1
1
*EXTP
2670.9
1
2
2
1
2447.7 3124.7
2
1
1
1
2135.5 2876.6
2
1
2
1
*EXTP
3351.2
2
2
1
1
1682.7 2674.6
2
2
2
1
2189.0 3132.3
3
1
1
1
2133.6 2884.2
3
1
2
1
2630.9 3368.4
3
2
1
1
1463.1 2684.5
3
2
2
1
2022.0 *EXTP
4
1
1
1
2160.1 2912.5
4
1
2
1
2696.4 3433.4
4
2
1
1
1425.7 2721.3
4
2
2
1
2080.0 3231.0
bhp(3)
3573.7
4044.7
3370.9
3824.7
3576.6
*EXTP
3374.6
3832.3
*EXTP
4068.4
3384.5
3852.8
3612.5
4133.4
3421.3
3931.0
Row '2 1 2 1', which has two of three values extrapolated, will have constant BHP values of
3351.2.
User's Guide IMEX
Example 5: Production in a condensate field

*PTUBE1 5
*DEPTH 8000.0
*GAS
** flo(1) flo(2) flo(3) flo(4)
0.0
1.0e+6 3.0e+6 7.0e+6
*OGR
** gfr(1) gfr(2)
0.00001 0.00002
*WGR
** wfr(1) wfr(2)
0.0
0.0001
*ALQ
** add(1)
0.0
*WHP
** whp(1) whp(2) whp(3)
500.0
1500.0 2500.0
*BHP
** iflo igfr iwfr iadd bhp(1) bhp(2)
1
1
1
1
2176.1 2873.7
1
1
2
1
2646.7 3344.7
1
2
1
1
1997.7 2670.9
1
2
2
1
2447.7 3124.7
2
1
1
1
2135.5 2876.6
2
1
2
1
2618.0 3351.2
2
2
1
1
1682.7 2674.6
2
2
2
1
2189.0 3132.3
3
1
1
1
2133.6 2884.2
3
1
2
1
2630.9 3368.4
3
2
1
1
1463.1 2684.5
3
2
2
1
2022.0 3152.8
4
1
1
1
2160.1 2912.5
4
1
2
1
2696.4 3433.4
4
2
1
1
1425.7 2721.3
4
2
2
1
2080.0 3231.0
bhp(3)
3573.7
4044.7
3370.9
3824.7
3576.6
4051.2
3374.6
3832.3
3584.2
4068.4
3384.5
3852.8
3612.5
4133.4
3421.3
3931.0
The acceptable range of values for reference depth is:
Min
Max
User's Guide IMEX
SI
m
-1.0E+4
1.0E+4
Field
ft
-32,808.0
32,808.0
Lab
cm
-1.0E+6
1.0E+6
Backward Compatibility Notes:
The previous tubing data keyword *PTUBE is obsolete. Its conversion to *PTUBE1 is given
as follows:
*PTUBE *OIL
*QO
(iqol)
*GOR
(igor)
*WCUT
(iwcut)
--*BHPTO
*PTUBE *LIQ
*QO/*QLIQ
(iqol)
*GOR
(igor)
*WCUT
(iwcut)
--*BHPTO
*PTUBE *WATER_GAS
*QG
(iqg)
--*WGR
(iwgr)
--*BHPTG
*PTUBE *CONDENSATE
*QG
(iqg)
*OGR
(iogr)
*WGR
(iwgr)
--*BHPTC
*PTUBE1
*OIL
*GOR
*WCUT
*ALQ
*BHP
*PTUBE1
*LIQ
*GLR
*WCUT
*ALQ
*BHP
*PTUBE1
*GAS
*OGR
*WGR
*ALQ
*BHP
*PTUBE1
*GAS
*OGR
*WGR
*ALQ
*BHP
(iflo)
(igfr)
(iwfr)
(iadd)
(iflo)
(igfr)
(iwfr)
(iadd)
(iflo)
(igfr)
(iwfr)
(iadd)
(iflo)
(igfr)
(iwfr)
(iadd)
For blanks in the *PTUBE format (---), add one entry of zero value for the corresponding
variable in the *PTUBE1 format. Subkeywords *DEPTH and *WHP remain the same.
BUILDER will automatically convert *PTUBE to *PTUBE1. Well Management can still
read the obsolete tubing table formats (*PTUBE, *VFPPROD and *VFPINJ), but can convert
them into the new formats (*PTUBE1 and *ITUBE1) by command-line option -newptube.
The new table(s) will be printed to the output file following the echo of the obsolete
keywords.
User's Guide IMEX
Set Number of Points for WHP Root Search (Optional)

*NWHYFS
PURPOSE:
*NWHYFS allows the user to set the number of points used to determine whether a value of
the BHP exists which yields the target well head pressure (WHP). This number defaults to 9
and it is not anticipated that the user would normally alter this value. However in certain
cases in which a well operating on an implicit WHP constraint suddenly shuts for no apparent
reason, invoking NWHYFS for the well with a value exceeding 9 (11 for example) may allow
the well to continue. Larger values of this integer may cause run times to increase.
FORMAT:
*NWHYFS well_list
integer_list
DEFINITIONS:
well_list
A set of 'well_names' or well_numbers; see below. The well list must be on
the same line as *NWHYFS and must end with the end of the line. If more
wells than can be listed on one line must have NWHYFS specified, a new
*NWHYFS line must be entered.
well_names
keyword applies. Limited wildcarding is available for the list of well names;
please see the manual page for the *SHUTIN keyword for an explanation of
the wildcard facility.
well_numbers
Any number of integers, or ranges of integers specifying the well numbers to
which this keyword applies.
integer_list
A list of integers containing either a single integer or exactly the same
number of entries as well_list contains. The values must be odd, must be at
least 5, and must not exceed 17. The default is 9. If a single integer is
entered in the list, it is applied to all wells in well_list.
DEFAULTS:
Optional keyword. If *NWHYFS does not appear in the data set, the value 9 is used for all wells.
User's Guide IMEX
CONDITIONS:
If it appears, this keyword must be located in the WELL AND RECURRENT DATA keyword
group. It must appear AFTER (but not necessarily immediately after) the first *DATE line. All
wells appearing in the well list must already have been defined with *WELL lines. The wells
in the list may be producers or injectors, but it is anticipated that user specification of this
parameter will be warranted only for producers and only in special circumstances. The integers
listed must be odd and between 5 and 17 inclusive. The parameter will have an effect only if at
least one of the specified wells is operating on a WHP constraint.
EXPLANATION:
If too few points are used in the WHP root search, the simulator may erroneously conclude
that no root exists and that the well is unable to operate on the WHP constraint. Specification
of a larger number of points may remedy this. Only extremely rarely is the default value of 9
points insufficient. On the other hand if there is no problem with roots not being located,
execution time may be shortened by decreasing the value, say, to 7. Again, only extremely
rarely would such a decrease shorten the run time significantly.
Example:
*NWHYFS
11 7
WELL1 WELL2
Directs that 11 points be used for WHP root searches for WELL1 and that 7 points be used
for WELL2.
User's Guide IMEX
Terminate Simulation (Required)
*STOP
PURPOSE:
*STOP causes simulation to terminate.

FORMAT:
*STOP
DEFAULTS:

CONDITIONS:
EXPLANATION:
This keyword is generally the last keyword in the input-data-file, and is used in conjunction
with a *DATE or *TIME keyword to indicate when to stop the simulation. However, it may
appear at any point within the Recurrent Data section to terminate the simulation at that point.
Data after the *STOP keyword is ignored.
For example, to complete the input-data-file in the normal manner at a specific date, use:
:
*DATE 1997 10 22
*STOP
For example, to terminate simulation after 7 years,
without deleting the remainder of the input-datafile, use:
:
*TIME 2555.0
*STOP
User's Guide IMEX
Documentation, Templates and

Execution
Introduction
This section is divided into two parts. The first part describes the installation procedure. The
second part describes the actual running of the program.
Documentation
LIST OF FILES REQUIRED FOR IMEX DOCUMENTATION
The following is the complete list of files to document IMEX.
DOCUMENTATION FILES
Mx200810en.chm
Soft copy of the IMEX 2008.10 user's manual in Windows .chm format.
Mx200810en.pdf
Soft copy of the IMEX 2008.10 user's manual in Adobe Acrobat Reader .pdf
format.
template.txt
Listing of headers of all sample data sets supplied in the testbed.
User's Guide IMEX
Documentation, Templates and Execution 1005
Templates
One hundred and sixty-seven template data sets are supplied with IMEX 2008.10. They are
all described in template.txt in the tpl directory. Below is a partial list of the types of data sets
provided.
Run Statistics for each of the 171 data sets are stored in the tpl/output directory. These
statistics compare results on the Windows 32 bit and Windows 64 bit workstations using 1, 2
and 4 CPUs.
SIMULATOR DATASETS
MXSPR002.DAT
Pseudo-miscible with chase gas option
MXSPR003.DAT
Restart from MXSPR002.DAT
MXDRM005.DAT
Field scale demonstration data set
MXSMO001.DAT
Demonstrates the depth-averaged initialization option
MXFRR002.DAT
Dual porosity option with Dual Permeability
MXFRR010.DAT
MXDP1B.DAT with z-coordinate downward
MXFRR003.DAT
Dual porosity option with MINC formulation
MXGRO002.DAT
SPE1 data set with a fault in the reservoir
MXDRM006.DAT
Sample data set for gas recycle
MXWWM001.DAT
A test for the gathering centre option
MXSMO002.DAT
A test for the gas lift option
MXSMO003.DAT
A test for the gas-water option
1006 Documentation, Templates and Execution
User's Guide IMEX
MXHRW004.DAT
Test # 1 for the horizontal well option
MXHRW005.DAT
MXGRO008.DAT
A test for the hybrid grid option
MXSMO005.DAT
A test for the hysteresis on water capillary pressure
MXSMO006.DAT
A test for the hysteresis on gas relative permeability
MXSPR004.DAT
A test for the laboratory unit option
MXSMO007.DAT
A test for the multiple PVT table option
MXWWM002.DAT
Test for multiple wells completed in a single grid block.
MXSMO004.DAT
The MXSMO005.DAT test data without water capillary pressure hysteresis.
Used for comparison with MXSMO005.DAT
MXWWM003.DAT
Test for GOR penalization constraint option.
MXGRO009.DAT
A test for the refined grid option
MXSPE001.DAT
SPE1 - 3d gas injection
MXGRO010.DAT
SPE1 data set with z-coordinate downward
MXSMO008.DAT
Test of net pay and volume modifiers
MXSPE002.DAT
SPE2 - 2d radial 3-phase coning
MXSPE005.DAT
SPE5 - pseudo-miscible
User's Guide IMEX
MXWWM004.DAT
Test of simple wellbore model
MXFRR004.DAT
Test no. 1 for the subdomain dual-porosity option
MXFRR005.DAT
MXGRO011.DAT
5,000 grid block problem based on mxspe1.dat
MXSMO009.DAT
2-phase (oil-water) 3d water injection
MXSPR005.DAT
Polymer option test case
MXSMO010.DAT
Test for Co vs. Pressure option.
MXGRO012.DAT
3d variable thickness, variable depth
MXWWM005.DAT
Voidage replacement, lease lines and sectors
MXNWP004.DAT
Sample data for well test analysis
See template.lis (or template.txt) for a complete list of data sets in the
template directory supplied
USEFUL NOTE
Keep a soft copy of these files somewhere, for comparison later with modified source code
and regenerated testbed outputs.
CHARACTER SET
There is a set of allowed characters that may be used in referring to keywords. Any character
outside this set will be interpreted as a blank. Characters in quotes or comments are not checked.
The purpose of the character set is to detect invisible non-blank characters, such as tab, which
some editors may insert in your data file.
The CMG keywords are composed of the upper and lower case alphabet, numerals 0-9,
keyword indicator (*), and arithmetic operators (=, +, -, /). Extra characters are included in
the set to accommodate the *TRANSLATE facility (see below). You may increase the
character set by expanding a data definition in subroutine RDLINE in the simulator source
code. Of course, the character set is restricted by the operating system being used.
User's Guide IMEX
INPUT AND OUTPUT UNITS, FILES, AND ATTACHMENTS

Identify the preferred method for attaching FORTRAN I/O unit numbers to files in their
computer's operating system. The following is a summary of the files that IMEX 2002.10
may use:
VARIABLE
UNIT
NO
FORMATTED
yes
iwrout
23
yes
iwrstd
yes
irdndx
iwrndx
irdsrf
25
24
27
yes
yes
no
iwrsrf
26
no
irdrwd
iwrrwd
irddat
irdda2
29
28
22
21
no
no
yes
yes
iscrt0
32
yes
User's Guide IMEX
MAX.
COMMENTS
RECL.
(bytes)
80
Used only if open statements are
enabled, to write queries to screen
and read file names.
132
Output print file. If file names are
read in, and a blank is input in
response to the output file name
query, defaults to standard out.
132
If iwrout = iwrstd, for example iwrout
= iwrstd = 6 then no diary file is
written. If iwrout is different than
iwrstd, for example, iwrout=23 and
iwrstd = 6, error messages and a time
step summary will be written to
iwrstd. iwrstd becomes a shorter file
which can be used as a diary or log
file. The default is iwrout = 23 and
iwrstd = 6 and thus this diary file is
written.
132
Input index results file.
132
Output index results file.
8*n
Input main-results-file. Where n is the
number of grid blocks. 8*n is the
record length for systems where each
variable will be represented by 8
bytes (e.g. IBM double precision).
8*n
Output main-results-file. Where n is
the number of grid blocks. 8*n is the
record length for systems where each
variable will be represented by 8
bytes (e.g. IBM double precision).
8*n
Input rewindable-restart-file.
8*n
Output rewindable-restart-file.
132
Input data file.
132
Second input data file for
*INCLUDE (if enabled).
132
Scratch file for internal IMEX use.
COMPARISON OF TESTBED RESULTS

Compare the output files with the corresponding files from the tape. Experience has shown
that different computers may generate slightly different answers resulting from slow 'drift'
from the HP results, especially for problems that use an iterative matrix solution method. This
is normal. Different timestep sizes also have this effect as well as different IMPES/fully
implicit patterns. Therefore, it may be sufficient to check the final answers for 'close'
agreement.
DIARY FILE
The file mxdiary.out includes the diary files in alphabetical order, for all of the test data sets.
Compare the diary files for individual runs with the corresponding case in mxdiary.out. The
number of time steps should be the same. If the timesteps are close, the rates should be close
as well. If the number of timesteps are different, you may check to see whether this is due to a
slow "drift" until, for instance, one run has an extra time step cut while the other manages to
get through.
RESTART RUNS
Try running a few restart runs. For example, run mxspe2.dat for two timesteps by setting
*MAXSTEPS = 2 in the *NUMERICAL keyword group and *WRST 1 in the I/O keyword
group. *WRST 1 will force the restart record writing every time step.
Call the output SR2 files as 'mxspe2.irf' and 'mxspe2.mrf' respectively. Then re-run the
second timestep by setting '*RESTART 1' in the I/O keyword group defining 'mxspe2.irf' and
'mxspe2.mrf' as the input SR2 files. Compare the results. They should be identical.
User's Guide IMEX
Running IMEX
In the explanation below, lines printed by IMEX are prefixed by "(IMEX)". User responses
are prefixed by "(USER)". You do not actually have to type "(IMEX)" or "(USER)".
Running IMEX is very simple and straight forward. After the compiling and linking have
been completed, IMEX may be executed in two different ways:
METHOD (1)
Execute the compiled and linked file for IMEX. The program will print the following on the
screen
SPECIFY FILE NAMES INTERACTIVELY
File names may be specified interactively by responding to IMEX prompts as shown below
(IMEX)
(USER)
(IMEX)
(IMEX)
(USER)
(IMEX)
(USER)
(IMEX)
(USER)
(IMEX)
*** IMEX Input/Output files specification :

*** Enter file name for data file
Q to quit, <C/R> to use default name of imex.dat:
data file name (for example mxspe1.dat)
Opened Scratch file on unit 32
Opened data file on unit 22, filename is mxspe1.dat
*** Enter file name for output file
Q to quit, <C/R> to use default name of mxspe1.out
output file name (for example mxspe1.out)
Opened output file on unit 23, filename is mxspe1.out
*** Enter file name for output index results
Q to quit, <C/R> to use default name of 'mxspe1.irf':
output index results file name (for example mxspe1.irf)
Opened INDEX-OUT on unit 24, filename is 'mxspe1.irf'
*** Enter file name for output main results
Q to quit, <C/R> to use default name of 'mxspe1.mrf':
output main results file name (for example mxspe1.mrf)
Opened MAIN-RESULTS-OUT on unit 26, filename is 'mxspe1.mrf'
METHOD (2)
Execute the compiled and linked file for IMEX. The program will print the following on the
screen
SPECIFY FILE NAMES INSIDE THE DATA FILE
Output, input irf, output irf, input mrf, and output mrf file names may be specified in the data
file itself by using the *FILENAMES keyword.
See Section 4 of this user's manual.
(IMEX)
(USER)
(IMEX)
*** IMEX Input/Output files specification :

*** Enter file name for data file
Q to quit, <C/R> to use default name of imex.dat
data file name (for example mxspe1.dat)
Opened Scratch file on unit 32
Opened data file on unit 22, filename is mxspe1.dat
Opened output file on unit 23, filename is mxspe1.out
Opened INDEX-OUT on unit 24, filename is 'mxspe1.irf'
Opened MAIN-RESULTS-OUT on unit 26, filename is 'mxspe1.mrf'
After the above file names have been given the program will start executing and the
following will appear on the screen.
User's Guide IMEX
===================== SUMMARY (from subroutine: INDATA) ============

Reading of initial data is complete.
Simulation will stop if there were error messages.
0 Warning messages. 0 Error messages.
====================================================================
===================== SUMMARY (from subroutine: INWELL) ============
Reading of well data is complete.
Simulation will stop if there were error messages.
0 Warning messages. 0 Error messages.
====================================================================
If there are any error or warning messages they will appear here. Next a summary will be
printed
I M E X
T I M E
S T E P
S U M M A R Y
First SPE Comparative Solution ...
(See Odeh, A.S., J.P.T., 33, pp.13-?)
vnew
========================================================================
Time Step
Time
Total Production
-------------- ---------------- ---------------------------------------C
Oil
Gas
Water GOR. Wat
Size
U
SCF/ Cu
No. Days IT T Days
yy:mm:dd
STB/D
MCF/D STB/D STB
%
---- ---- -- - ----- -------------------- ------ ---- --1w 1.00
1
0 1.000 1986:04:23 20000.00 25400.00
1270
...
2
1.98
1
0 2.979 1986:04:25 20000.00 25400.00
1270
...
3
2.07
1
0 5.047 1986:04:27 20000.00 25400.00
1270
...
.
.
.
.
.
.
.
.
.
. ...
===================== SUMMARY (from subroutine: INWELL) ===================
Simulation run terminated. Stopping time reached.
0 Warning messages.
0 Error messages.
===========================================================================
Timesteps: 40
Newton Cycles: 112
Cuts: 1
Average Implicitness: 0.226
Material Balances (owgs): 0.999 1.000 1.001
Solver Iteration...
End of Simulation: Normal Termination
If the simulation run is not successful then an ABNORMAL termination message is printed.
User's Guide IMEX
Sample Data Sets
Notes on Sample Data Sets

The following data sets are included in the testbed shipped in soft copy with IMEX. A brief
description of each data set is given below. Of these, some data sets are also attached in hard
copy and are part of this section of the user's manual. If you wish to examine sample data that
are not part of the hard copy please refer to the directory on your computer system where
IMEX was installed. Please see template.lis (or template.txt) for brief descriptions of the
complete test bed.
SIMULATOR DATASETS AND OUTPUT
MXSPR002.DAT
Pseudo-miscible with chase gas option
MXSPR003.DAT
Restart from MXSPR002.DAT
MXDRM005.DAT
Field scale demonstration data set
MXSMO001.DAT
Demonstrates the depth-averaged initialization option
MXFRR002.DAT
Dual porosity option with Dual Permeability
MXFRR010.DAT
MXDP1B.DAT with z-coordinate downward
MXFRR003.DAT
Dual porosity option with MINC formulation
MXGRO002.DAT
SPE1 data set with a fault in the reservoir
MXDRM006.DAT
Sample data set for gas recycle
User's Guide IMEX
Sample Data Sets 1013
MXWWM001.DAT
A test for the gathering centre option
MXSMO002.DAT
A test for the gas lift option
MXSMO003.DAT
A test for the gas-water option
MXHRW004.DAT
MXHRW005.DAT
MXGRO008.DAT
A test for the hybrid grid option
MXSMO005.DAT
A test for the hysteresis on water capillary pressure
MXSMO006.DAT
A test for the hysteresis on gas relative permeability
MXSPR004.DAT
A test for the laboratory unit option
MXSMO007.DAT
A test for the multiple PVT table option
MXWWM002.DAT
Test for multiple wells completed in a single grid block
MXSMO004.DAT
The MXSMO005.DAT test data without water capillary pressure hysteresis
Used for comparison with MXSMO005.DAT
MXWWM003.DAT
Test for GOR penalization constraint option
MXGRO009.DAT
A test for the refined grid option
MXSPE001.DAT
SPE1 - 3d gas injection
1014 Sample Data Sets
User's Guide IMEX
MXSPE001.OUT
Output file for mxspe001.dat
MXGRO010.DAT
SPE1 data set with z-coordinate downward
MXSMO008.DAT
Test of net pay and volume modifiers
MXSPE002.DAT
SPE2 - 2d radial 3-phase coning
MXSPE002.OUT
MXSPE005.DAT
SPE5 - pseudo-miscible
MXSPE005.OUT
MXWWM004.DAT
Test of simple wellbore model
MXFRR004.DAT
MXFRR005.DAT
MXGRO011.DAT
5,000 grid block problem based on mxspe1.dat
MXSMO009.DAT
2-phase (oil-water) 3d water injection
MXSPR005.DAT
Polymer option test case
MXSMO010.DAT
Test for Co vs. Pressure option
MXGRO012.DAT
3d variable thickness, variable depth
MXGRO012.OUT
Output file for mxgro012.dat
User's Guide IMEX
MXWWM005.DAT
Voidage replacement, lease lines and sectors
MXNWP004.DAT
Sample data for well test analysis
User's Guide IMEX
MXSPE001.dat: First SPE Comparative Solution Model

******************************************************************************
** mxspe001.dat:
First SPE Comparative Solution Model
**
******************************************************************************
** FILE:
MXSPE001.DAT
**
** MODEL: 3x3x3
SPE 1
GAS INJECTION **
******************************************************************************
** This is a 10x10x3 Cartesian grid gas injection problem (SPE 1). The gas injector **
** is located in block (1,1,3), and the producer in block (10,10,1).
There is a
**
** large permeability contrast in the vertical direction.
**
**************************************************************************************
** CONTACT: CMG (403)531-1300; 282-6495 (fax); cmg@cmgroup.com (Internet)**
********************************************************************************
** I/O Control Section
**
********************************************************************************
*TITLE1
'First SPE Comparative Solution Project'
*TITLE3
'(See Odeh, A.S., J.P.T., 33, pp.13-25, 1981.)'
*INUNIT *FIELD
*WPRN
*WPRN
*WPRN
***WPRN
*WELL
*GRID
*ITER
*ITER
10
*TIME
*BRIEF
*MATRIX
*OUTPRN
*OUTPRN
*OUTPRN
*OUTPRN
*WELL
*RES
*TABLES
*GRID
*BRIEF
*NONE
*NONE
*SO *SG *PRES *BPP *IMEXMAP
*WSRF
*WSRF
*OUTSRF
*GRID
*WELL
*GRID
*TIME
1
*OILSAT *GASSAT *PRES *BPP
********************************************************************************
** Reservoir Description Section
**
********************************************************************************
*GRID *CART 10
10
3
** 10 blocks in each the I (x) and J (y) directions.
** 3 layers in the K (z) direction.
*DI *CON 1000.0
** I direction block dimensions constant at 1000 ft.
*DJ *CON 1000.0
** J direction block dimensions constant at 1000 ft.
*DK *KVAR
** K direction block dimensions (layer thicknesses)
50.0 30.0 20.0
** are 50 ft., 30 ft. and 20 ft. for the bottommost
** middle and uppermost layers, respectively.
*DEPTH 1 1 1 8400.0
*POR *CON 0.3
** Depth to center of first block, in bottom layer.

** Porosity is constant throughout the reservoir.
*CPOR
3.0E-6
*PRPOR 14.7
** Rock compressibility and

** reference pressure.
*PERMI *KVAR
200.0
50.0
*PERMJ *KVAR
200.0
50.0
*PERMK *KVAR
20.0
40.0
*MODEL *BLACKOIL
User's Guide IMEX
500.0
500.0
60.0
**
**
**
**
**
**
Each of the I, J, and K permeabilities are

constant within each layer of the reservoir,
but vary from layer to layer. Hence the use
of *KVAR to input them layer by layer. Note
that the first K layer is at bottom of the
reservoir.
** Solve three equations.
********************************************************************************
** Component Property Section
**
********************************************************************************
*PVT
** p
rs
bo
eg
viso
visg
14.70000
1.00000
1.06200
6.00000
1.04000 0.0080000
264.700 90.50000
1.15000 82.70000 0.9750000 0.0096000
514.700
180.000
1.20700
159.000 0.9100000 0.0112000
1014.700
371.000
1.29500
313.000 0.8300000 0.0140000
2014.700
636.000
1.43500
620.000 0.6950000 0.0189000
2514.000
775.000
1.50000
773.000 0.6410000 0.0208000
3014.700
930.000
1.56500
926.000 0.5940000 0.0228000
4014.700 1270.000
1.69500 1233.000 0.5100000 0.0268000
5014.700 1618.000
1.82700 1541.000 0.4490000 0.0309000
9014.700 2984.000
2.35700 2591.000 0.2030000 0.0470000
**
**
**
**
Note: The above PVT data violates the total hydrocarbon compressibility check. To
correct, you could either change rs value at p=5014.7 from 1618 to 1600 or to change
the last bo entry from 2.357 to 2.352. Other possibilities could be equally applied.
The above corrections were tested and they have insignificant effects.
*DENSITY *OIL
46.244
*DENSITY *GAS
0.0647
*DENSITY *WATER 62.238
*CO
1.3687E-5
*CVO
4.6000E-5
*BWI
1.0410
*CW
2.9886E-6
*REFPW 14.7
*VWI
0.31
*CVW
0.0
*ROCKFLUID
********************************************************************************
** Rock-Fluid Property Section
**
********************************************************************************
*RPT 1
*SWT
** sw
krw
krow
0.1200000
0.0
1.0
0.8200000
0.0
0.0
*SLT
** sl
0.12
0.15
0.30
0.40
0.50
0.55
0.60
0.70
0.75
0.80
0.88
0.95
0.98
0.999
1.0
krg
0.984
0.98
0.94
0.87
0.72
0.6
0.41
0.19
0.125
0.075
0.025
0.005
0.0
0.0
0.0
krog
0.0
0.0
0.0
1.0000E-4
0.001
0.01
0.021
0.09
0.2
0.35
0.7
0.98
0.997
1.0
1.0
User's Guide IMEX
*INITIAL
********************************************************************************
** Initial Conditions Section
**
********************************************************************************
*VERTICAL *BLOCK_CENTER *WATER_OIL_GAS ** Use vertical equilibrium calculation.
*PB *CON
4014.7
** bubble point pressure
*REFDEPTH
8400.
** Give a reference depth and
*REFPRES
4800.
** associated pressure.
*DWOC
9500.
** Depth to water-oil contact
*DGOC
7000.
** Depth to gas-oil contact
*NUMERICAL
********************************************************************************
** Numerical Control Section
**
********************************************************************************
*DTMAX
1000.
** Maximum time step size
*MAXSTEPS
100
** Maximum number of time steps
*NORM *PRESS 1000.0
** Normal maximum changes per time step
*NORM *SATUR
0.20
*AIM *THRESH
0.25 .25
** Use thresh hold switching criteria
*RUN
********************************************************************************
** Well and Recurrent Data Section
**
********************************************************************************
*DATE 1986 04 22
*DTWELL 1.0
*WELL 1
'Injector'
*INJECTOR
*INCOMP
*OPERATE
*OPERATE
*MONITOR
*PERF 1
** if
1
*WELL 2
*UNWEIGHT 1
*GAS
*MAX *STG 1.0E+8
*MAX *BHP 20000.0
*BACKFLOW
*STOP
jf
1
kf
3
** Define the type of well 1.

** Injecting gas.
** If backflow occurs, stop the simulation.
wi
1.0E+5
'Producer'
*PRODUCER 2
*OPERATE *MAX
*OPERATE *MIN
*STO 20000.0
*BHP 1000.0
*MONITOR *MIN
*STO
*MONITOR *BACKFLOW
*MONITOR *GOR
1000.0
*STOP
20000.0
*STOP
*STOP

**
rad
geofac wfrac
*GEOMETRY *K 0.25
0.34
1.0
*PERF *GEO 2
** if
jf
kf
ff
10
10
1
1.0
*TIME 3650.0
*STOP
User's Guide IMEX
**
**
**
**
**
**
**
**
Define the type of well 2.

Initially, operate on a rate constraint.
If the BHP falls below 1000 psi, switch
to a minimum pressure constraint.
When the oil rate falls below 1000 bbl/D
stop the simulation.
Stop if backflow occurs.
Stop if the GOR exceeds 20000.
skin
0.0
** Time of next well change.

** Stop the simulation.
MXSPE002.dat: Three-Phase Coning Model

******************************************************************************
** mxspe002.dat:
Three-Phase Coning Model
**
******************************************************************************
** FILE:
MXSPE002.DAT
**
** MODEL: 10x1x15
SPE 2
THREE-PHASE CONING MODEL **
**************************************************************************************
** Second Comparative Solution Project: Chappelear,J.E., and Nolen,J.S., "Second **
** Comparative Solution Project: A Three Phase Coning Study", Sixth SPE Symposium on **
** Reservoir Simulation,1982 (SPE 10489). This is a 3-phase,10x1x15, coning model. **
**************************************************************************************
** CONTACT: CMG (403)531-1300; 282-6495 (fax); cmg@cmgroup.com (Internet)**
*************************************************************************************
*TITLE1
'Second SPE Comparative Solution Project'
*TITLE2
'Three Phase Coning Study - Radial coordinates'
*TITLE3
'Chappelear and Nolen, Sixth SPE Symposium, Jan.31-Feb.3, 1982.'
*INUNIT
*FIELD
*WPRN
*WELL 10
*WPRN
*GRID *TIME
*WPRN
*ITER *NONE
*OUTPRN
*GRID *SO *SG *SW *PRES *IMEXMAP
*WSRF
*GRID 0
*WSRF
*WELL 1
**************************************************************************
** Reservoir description Section
**************************************************************************
*GRID *RADIAL
10
1
15
** This is a cylindrical coordinate
** problem, r x theta x z.
*DI *IVAR
2.0
2.32
50.55 109.21
5.01
235.92
10.84
509.68
*DJ *CON
360.000
** J (theta) direction dimensions are constant
*DK *KVAR
100.0 50.0
2*8.0 14.0
*DEPTH 1
20.0
16.0
*POR *KVAR
0.157 0.116
0.13
0.14
0.13
0.16
15
18.0
15.0
19.0
26.0
9010.0
** K (z) direction dimensions vary from block to block

12.0 18.0
15.0 20.0
** Depth to center of first block, in topmost layer.
** Porosities vary per K (z) layer

0.12 0.105 0.12
0.08 2*0.17
0.111 0.097 0.087
*PRPOR 3600.0
*CPOR 4.0E-6
**
**
**
**
** I (r) direction dimensions vary from block to block

23.4
1101.08

** reference pressure
Each of the I, J, and K permeabilities are constant within each layer

of the reservoir but vary from layer to layer. Hence the use of *KVAR
to input them layer by layer. Note that the first K layer is at
*PERMI *KVAR
350.0 191.0
472.0 682.0
90.0 202.0
*PERMJ *KVAR
350.0 191.0
472.0 682.0
90.0 202.0
137.5
60.0
148.0
300.0
775.0
47.5
125.0
418.5
35.0
137.5
60.0
148.0
300.0
775.0
47.5
125.0
418.5
35.
User's Guide IMEX
*PERMK *KVAR
35.0
19.1
47.2
68.2
9.0
20.2
13.75
6.0
14.8
30.0
77.5
4.75
12.5
41.8
3.5
**************************************************************************
** Component Properties Section
**************************************************************************
*MODEL *BLACKOIL
** Three phase - three equation
*PVT
**
p
400.000
800.000
1200.000
1600.000
2000.000
2400.000
2800.000
3200.000
3600.000
4000.000
4400.000
4800.000
5200.000
5600.000
rs
165.000
335.000
500.000
665.000
828.000
985.000
1130.000
1270.000
1390.000
1500.000
1600.000
1676.000
1750.000
1810.000
*DENSITY *OIL
*DENSITY *GAS
*DENSITY *WATER
*CO
*CVO
*BWI
*CW
*REFPW
*VWI
*CVW
bo
1.01200
1.02550
1.03800
1.05100
1.06300
1.07500
1.08700
1.09850
1.11000
1.12000
1.13000
1.14000
1.14800
1.15500
45.0
0.0702
63.02
1.0E-5
0.0
1.01420
3.0E-6
14.7
0.96
0.0
eg
169.490
338.980
510.200
680.270
847.460
1020.400
1190.500
1351.400
1538.500
1694.900
1851.900
2040.800
2222.200
2381.000
**
**
**
**
**
**
**
viso
1.17000
1.14000
1.11000
1.08000
1.06000
1.03000
1.00000
0.9800000
0.9500000
0.9400002
0.9200000
0.9100000
0.9000000
0.8900000
visg
0.0130000
0.0135000
0.0140000
0.0145000
0.0150000
0.0155000
0.0160000
0.0165000
0.0170000
0.0175000
0.0180000
0.0185000
0.0190000
0.0195000
Oil compressibility
Viscosity dependence on oil compressibility
Water formation volume factor
Water compressibility
Reference pressure
Water viscosity at reference pressure
Pressure dependence on water compressibility
**************************************************************************
**Rock-fluid Property Section
**************************************************************************
*ROCKFLUID
*RPT
*SWT
**
sw
0.22
0.30
0.40
0.50
0.60
0.80
0.90
1.00
krw
0.0
0.07
0.15
0.24
0.33
0.65
0.83
1.0
** Water-oil relative permeability table.
*SLT
**
sl
0.22
0.30
0.40
0.50
0.60
0.70
0.80
0.90
0.96
1.0
krg
1.0
0.8125
0.5
0.42
0.34
0.24
0.1
0.022
0.0
0.0
User's Guide IMEX
krow
1.0
0.4
0.125
0.0649
0.0048
0.0
0.0
0.0
krog
0.0
0.0
0.0
0.0
0.0
0.02
0.1
0.33
0.6
1.0
pcow
7.0
4.0
3.0
2.5
2.0
1.0
0.5
0.0
** Liquid-gas relative permeability table.
pcog
3.9
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.2
0.0
**************************************************************************
**************************************************************************
*INITIAL
*VERTICAL *BLOCK_CENTER *WATER_OIL_GAS ** Turn vertical equilibrium option
*PB *CON 0.0
** Set bubble point pressure to 0.
*REFPRES 3600.0
** Set reference pressure
*REFDEPTH 9035.0
** at the reference depth.
*DWOC
9209.0
** Set water-oil contact
*DGOC
9035.0
** Set gas-oil contact
**************************************************************************
**************************************************************************
*NUMERICAL
*DTMAX
365
*MAXSTEPS
100
*NORM *PRESS 250.
*AIM *THRESH .25 .25
**************************************************************************
** Well Data Section
**************************************************************************
*RUN
*DATE 1994 01 01
*DTWELL
0.0250000
*WELL 1 'Producer' *VERT 1 1
*PRODUCER 1
** Primary operating constraint is oil rate
** Subsequent operating constraint is bottomhole pressure.
*OPERATE *MIN *BHP 3000.00 *CONT *REPEAT
**
rad
*GEOMETRY *K 0.25
*PERFV *GEO 1
** kf
ff
8:9
1.0
*AIMWELL *WELLN
*TIME 10.0
*ALTER
1
100.0
*AIMWELL *WELLN
*TIME 50.0
*ALTER
1
1000.0
*AIMWELL *WELLN
*TIME 720.0
*ALTER
1
100.0
*AIMWELL *WELLN
*TIME 900.0
geofac
0.5
wfrac
1.0
skin
0.0
** Well change at this time.

** Change primary constraint value.
** Normal termination should occur at

** this time
*STOP
User's Guide IMEX
MXWWM004.dat WAG with Simple Wellbore Model

******************************************************************************
** mxwwm004.dat:
WAG With Simple Wellbore Model
**
******************************************************************************
** FILE:
MXWWM004.DAT
**
** MODEL: 7x7x3
SPE 5
WAG (PSEUDO-MISCIBLE) WITH SIMPLE WELLBORE MODEL **
**************************************************************************************
** Based on the Fifth SPE Comparative Solution Project, it models WAG process using **
** pseudo-miscible formulation.
A simple wellbore model was applied to the wells **
** for calculating well head pressures, the operating constraints.
**
**************************************************************************************
** CONTACT: CMG (403)531-1300; 282-6495 (fax); cmg@cmgroup.com (Internet)
**
**************************************************************************************
*TITLE1
'SPE 5, With Simple Wellbore Model'
*TITLE2
'Two injectors - one solvent, one water located in (1,1,3)'
*INUNIT *FIELD
*WPRN
*WPRN
*WPRN
*WSRF
*WSRF
*ITER
*WELL
*GRID
*GRID
*WELL
*BRIEF
*TIME
*TIME
0
*TIME
*************************************************************************
*************************************************************************
*GRID *CART 7 7 3
*DI *CON
** All I (x) direction blocks are constant
500.000
** in length.
*DJ *CON
** All J (y) direction blocks are constant
500.000
** in length.
*DK *KVAR
** K (z) direction blocks vary per layer
50.0 30.0 20.0
*DEPTH 1 1 1 2000.
*POR *CON
0.3000000
** Porosity is constant throughout.
*PRPOR 4000.000
*CPOR
5.000E-6

**
**
**
**
Each of the I, J, and K permeabilities are constant within each layer

of the reservoir but vary from layer to layer. Hence the use of *KVAR
to input them layer by layer. Note that the first K layer is at
*PERMI *KVAR
200. 50. 500.
*PERMJ *EQUALSI
*PERMK *KVAR
25.0 50. 50.
** The transmissibility multipliers default to 1.
*MODEL *MISNCG
User's Guide IMEX
** Use the pseudomiscible option with

** no chase gas.
*************************************************************************
** Component Property section
*************************************************************************
*PVT
** PVT table
**
p
rs
bo
eg
viso
visg
14.7
0.0
1.03480
4.730 0.310
0.0107
500.0
117.600
1.10170
168.800 0.295
0.0127
1000.0
222.600
1.14780
350.803 0.274
0.0134
1200.0
267.700
1.16770
426.603 0.264
0.0138
1500.0
341.400
1.19970
541.800 0.249
0.0145
1800.0
421.500
1.23500
657.808 0.234
0.0153
2000.0
479.000
1.26000
735.168 0.224
0.0159
2302.3
572.800
1.30100
850.991 0.208
0.0170
2500.0
634.100
1.32780
907.703 0.200
0.0177
3000.0
789.300
1.39560 1015.022 0.187
0.0195
3500.0
944.400
1.46340 1096.972 0.175
0.0214
4000.0 1099.500
1.53210 1159.958 0.167
0.0232
4500.0 1254.700
1.59910 1215.953 0.159
0.0250
4800.0 1347.800
1.63980 1245.020 0.155
0.0261
** table range extended
5300.0 1503.000
1.70700 1285.000 0.149
0.0279
5800.0 1658.000
1.77400 1320.000 0.144
0.0297
*PVTS
**
p
rss
es
14.7
0.0
4.480
500.0
0.0
178.501
1000.0
0.0
395.100
1200.0
0.0
491.304
1500.0
0.0
641.313
1800.0
0.0
790.077
2000.0
0.0
885.269
2302.3
0.0 1020.096
2500.0
0.0 1100.715
3000.0
0.0 1280.902
3500.0
0.0 1429.797
4000.0
0.0 1555.210
4500.0
0.0 1661.958
4800.0
0.0 1719.100
** table range extended
5300.0
0.0 1815.766
5800.0
0.0 1911.667
*DENSITY *OIL
38.53
*DENSITY *GAS
0.06864
*DENSITY *WATER
62.4
*DENSITY *SOLVENT 0.06248
*CO
*CVO
*BWI
*CW
*REFPW
*VWI
*CVW
2.1850E-5
2.5500E-5
1.0
3.3000E-6
14.7
0.7000000
0.0
*OMEGASG 1.000
*MINSS
0.0
** PVT table for solvent

viss
omega_s
0.011
0.0
0.012
0.0
0.013
0.0
0.014
0.0
0.016
0.0
0.018
0.0
0.019
0.0
0.022
0.0
0.023
0.2833596
0.027
1.0
0.031
1.0
0.034
1.0
0.037
1.0
0.038
1.0
0.040
0.042
**
**
**
**
**
**
**
1.0
1.0
Oil compressibility
Pressure dependence on oil compressibility
Water compressibility
Reference pressure
Water viscosity at reference pressure
Pressure dependence on water compressibility
** Gas and solvent mixing parameter

** Minimum solvent saturation
User's Guide IMEX
*ROCKFLUID
*************************************************************************
** Rock-Fluid Properties
*************************************************************************
*RPT
*SWT
** sw
krw
krow
pcow
sorm
0.2000000
0.0
1.0000
45.00
0.0
0.2899000
0.0022
0.6769
19.03
0.0
0.3778000
0.0180
0.4153
10.07
0.0
0.4667000
0.0607
0.2178
4.90
0.0
0.5556000
0.1438
0.0835
1.80
0.0
0.6444002
0.2809
0.0123
0.50
0.0
0.7000000
0.4089
0.0
0.05
0.0
0.7333000
0.4855
0.0
0.01
0.0
0.8221999
0.7709
0.0
0.0
0.0
0.9111000
1.0000
0.0
0.0
0.0
1.00000
1.0000
0.0
0.0
0.0
*SLT
** sl
0.2000000
0.2889000
0.3500000
0.3778000
0.4667000
0.5556000
0.6444002
0.7333000
0.8221999
0.9111000
0.9500000
1.00000
krg
1.00
0.56
0.39
0.35
0.20
0.10
0.05
0.03
0.01
0.001
0.0
0.0
krog
0.0
0.0
0.0
0.0110
0.0370
0.0878
0.1715
0.2963
0.4705
0.7023
0.8800
1.0000
pcow
30.000
8.000
4.000
3.000
0.800
0.030
0.001
0.001
0.0
0.0
0.0
0.0
*INITIAL
*************************************************************************
*************************************************************************
*USER_INPUT
*PRES *KVAR
** Pressure is constant in each layer
4000.0
3990.3
3984.3
** but varies between the layers.
*PB
*CON
** Initial bubble point pressure is
2302.3
** constant throughout
*PBS *CON
** Initial solvent bubble point pressure
2302.3
** is constant throughout.
*SW *CON
** Water saturations are constant.
0.20
*SO *CON
** Oil saturations are constant.
0.80
*DATUMDEPTH 2000.0 *REFDENSITY *GRIDBLOCK
*NUMERICAL
*************************************************************************
** Numerical Methods Control Section
*************************************************************************
*DTMAX 100.
*MAXSTEPS 999
*NORM *PRESS 500.0
*AIM *THRESH .25 .25
User's Guide IMEX
*RUN
*************************************************************************
*************************************************************************
*DATE 1986 12 1
*DTWELL 1.000
**
**
**
**
**
Two injectors are defined for the same grid block, to model one
injection well in the problem. This is to allow easy switching
between water injection and solvent injection during the WAG
process without having to completely redefine the injector.
At least one of the injectors is shutin at all times.

*WELL 2 'PRODUCER'
*WELL 3 'SOLVENT INJECTOR'
** All wells have the same geometry.
**
rad
geofac wfrac skin
*GEOMETRY *K
0.25
0.34
1.0
0.0
** Initially both injectors are shut in; only the

** producer is operational. True for the first two years.
*IWELLBORE
** wdepth wlength rough whptem
8400.
8400.
0.0001 60.0
*INCOMP *WATER
*OPERATE *MAX *STW
12000.0
*OPERATE *MAX *WHP
3000.0
*OPERATE *MAX *BHP
10000.0
*PERF *GEO 1
** if jf kf ff
1
1
3
1.0
*SHUTIN 1
*PRODUCER 2
**
depth
ibhp
*PWELLBORE 7000.0
1
*OPERATE *MIN *WHP 1600.0
*MONITOR *GOR
10000.0 *STOP
*MONITOR *WCUT 0.8330002 *STOP
*PERF *GEO 2
** if jf kf ff
7
7
1
1.0
bhptem
160.0
rw
0.25
** Primary operating constraint.

** Bottomhole pressure is also monitored.
** Monitor GOR and stop simulation if hit.
** Monitor Water Cut and stop simulation.
*IWELLBORE
** wdepth wlength rough whptem bhptem
rw
8400.
8400.
0.0001 60.0
160.0
0.25
*INCOMP
*SOLVENT *GLOBAL
** c1, c2, c3, ic4, nc4, ic5, nc5, nc6, co2
0.77 0.0 0.20 0.0 0.0 0.0 0.0 0.03 0.0
*OPERATE *MAX *STS
12000.0
*OPERATE *MAX *WHP
12000.0
*OPERATE *MAX *BHP
10000.0
*PERF *GEO 3
** if jf kf ff
1
1
3
1.0
*SHUTIN 3
User's Guide IMEX
*AIMSET *CON 0
*AIMWELL *WELLN
*PTUBE *OIL 1
** tubing data for the producer.
*DEPTH 8400.0
** used with wellbore option.
*GOR
** gor(1)
gor(2)
500.
1000.
*QO
** qo(1)
qo(2)
qo(3)
qo(4)
0.0 4000.0 8000.0 16000.0
*WCUT
** wcut(1) wcut(2)
0.0
0.5
*WHP
** whp(1)
whp(2)
whp(3)
200.0
900.0
1600.0
*BHPT
** ig
iwcut
iq
bhp(1)
bhp(2)
bhp(3)
1
1
1
2176.1
2873.7
3573.7
2
1
1
1997.7
2670.9
3370.9
1
2
1
2646.7
3344.7
4044.7
2
2
1
2447.7
3124.7
3824.7
1
1
2
2135.5
2876.6
3576.6
2
1
2
1682.7
2674.6
3374.6
1
2
2
2618.0
3351.2
4051.2
2
2
2
2189.0
3132.3
3832.3
1
1
3
2133.6
2884.2
3584.2
2
1
3
1463.1
2684.5
3384.5
1
2
3
2630.9
3368.4
4068.4
2
2
3
2022.0
3152.8
3852.8
1
1
4
2160.1
2912.5
3612.5
2
1
4
1425.7
2721.3
3421.3
1
2
4
2696.4
3433.4
4133.4
2
2
4
2080.0
3231.0
3931.0
*TIME 365.000
*TIME 730.000
*OPEN 1
*AIMSET *CON 0
*AIMWELL *WELLN
*TIME 1095.0
*SHUTIN 1
*OPEN 3
*AIMSET *CON 0
*AIMWELL *WELLN
**
**
**
**
**
Forces time step to end at one

Forces time step to end at two
Open water injector, begin WAG
Reset all blocks to explicit.
Set well blocks and neighbours
year.
years.
process.
implicit.
** Shutin water injector.

*TIME 1460.0
*SHUTIN 3
*OPEN 1
*AIMSET *CON 0
*AIMWELL *WELLN
*TIME 1825.0
*SHUTIN 1
*OPEN 3
*AIMSET *CON 0
*AIMWELL *WELLN
*TIME 2190.0
*SHUTIN 3
*OPEN 1
*AIMSET *CON 0
*AIMWELL *WELLN
User's Guide IMEX
*TIME 2555.0
*SHUTIN 1
*OPEN 3
*AIMSET *CON 0
*AIMWELL *WELLN
*TIME 2920.0
*SHUTIN 3
*OPEN 1
*AIMSET *CON 0
*AIMWELL *WELLN
*TIME 3285.0
*SHUTIN 1
*OPEN 3
*AIMSET *CON 0
*AIMWELL *WELLN
*TIME 3650.0
*SHUTIN 3
*OPEN 1
*AIMSET *CON 0
*AIMWELL *WELLN
*TIME 4015.0
*SHUTIN 1
*OPEN 3
*AIMSET *CON 0
*AIMWELL *WELLN
*TIME 4380.0
*SHUTIN 3
*OPEN 1
*AIMSET *CON 0
*AIMWELL *WELLN
*TIME 4745.0
*SHUTIN 1
*OPEN 3
*AIMSET *CON 0
*AIMWELL *WELLN
*TIME 5110.0
*SHUTIN 3
*OPEN 1
*AIMSET *CON 0
*AIMWELL *WELLN
*TIME 5475.0
*SHUTIN 1
*OPEN 3
*AIMSET *CON 0
*AIMWELL *WELLN
*TIME 5840.0
*SHUTIN 3
*OPEN 1
*AIMSET *CON 0
*AIMWELL *WELLN
*TIME 6205.0
*SHUTIN 1
*OPEN 3
*AIMSET *CON 0
*AIMWELL *WELLN
User's Guide IMEX
*TIME 6570.0
*SHUTIN 3
*OPEN 1
*AIMSET *CON 0
*AIMWELL *WELLN
*TIME 6935.0
*SHUTIN 1
*OPEN 3
*AIMSET *CON 0
*AIMWELL *WELLN
*TIME 7300.0
*STOP
User's Guide IMEX
MXWWM005.dat: Voidage Replacement & Lease Line Model

******************************************************************************
** mxwwm005.dat:
Voidage Replacement And Lease Line Model
**
******************************************************************************
** FILE:
MXWWM005.DAT
**
** MODEL: 10x10x3
MODIFIED SPE 1
VOIDAGE REPLACEMENT GAS INJECTION **
**************************************************************************************
** SPE 1 gas injection problem is modified to inject gas at a rate that would **
** replace the voidage at desired fractions.
Meanwhile, Leaseline option is **
** activated to generate production output based on each of the leases.
**
**************************************************************************************
**
**************************************************************************************
*TITLE1
'Voidage replacement, lease-line option'
*TITLE2
'10x10x3 rectangular grid problem'
*INUNIT *FIELD
*WPRN
*WPRN
*WELL 10
*GRID 10
*OUTPRN *GRID *EXCEPT *SS

*WSRF
*GRID 0
*WSRF
*WELL 1
*SSPRES *WINFLUX
*DATUMPRES
****************************************************************************
**************************************************************************
*GRID *CART 10
10
3
** Rectangular grid with 10 grid blocks in
** I direction, 10 grid blocks in J
** direction and 3 layers in the K
** direction.
*DI *CON 1000.0
** Dimension of I direction blocks are
** constant for each block (1000 ft).
*DJ *CON 1000.0
** Dimension of J direction blocks are
** constant for each block (1000 ft).
*DK *KVAR
** K direction dimensions are the same
50.0 30.0 20.0
** within a layer but differ between layers.
*DEPTH 1 1 1 2000.0
*POR *CON 0.30
*CPOR
3.0E-6
*PRPOR 14.7
**
**
**
**
Depth to center of first block, in bottom layer.

Porosity is constant throughout.
Rock compressibility and
reference pressure.
*PERMI *KVAR
200.0
50.0
*PERMJ *KVAR
200.0
50.0
*PERMK *KVAR
20.0 200.0
**
**
**
**
**
**

reservoir.
500.0
500.0
200.0
** By specifying the sector and leaseline options you are requesting

** special output for the areas indicated.
** Each sector is defined here separately
**
name
i1:i2 j1:j2 k1:k2
*SECTOR 'South-West' 1:4
1:4
1:3
'South-East' 1:4
5:10
1:3
*SECTOR 'North-West' 5:10
1:4
1:3
'North-East' 5:10
5:10
1:3
'Total-Field' 1:10
1:10
1:3
** The leaseline is defined here as well.
**
caname i1:i2 j1:j2
k1:k2
*LEASE
'LOO1' 5:5
1:10
1:3
*IDIR
User's Guide IMEX
**************************************************************************
** Component Property section
**************************************************************************
*MODEL *BLACKOIL
** Three phase (three equations solved)
*PVT
**
p
14.7
265.0
515.0
1015.0
2015.0
2515.0
3015.0
4015.0
5015.0
9015.0
rs
1.0
90.5
180.0
371.0
636.0
775.0
930.0
1270.0
1600.0
2984.0
*DENSITY *OIL
*DENSITY *GAS
*DENSITY *WATER
*CO
*CVO
*BWI
*CW
*REFPW
*VWI
*CVW
bo
1.06
1.15
1.21
1.30
1.43
1.50
1.57
1.69
1.83
2.36
eg
6.0
82.7
159.0
313.0
620.0
773.0
926.0
1233.0
1541.0
2591.0
viso
1.04
0.975
0.910
0.830
0.6950002
0.641
0.594
0.510
0.449
0.203
visg
0.0080
0.0096
0.0112
0.0140
0.0189
0.0208
0.0228
0.0268
0.0309
0.0470
46.2
0.0647
62.2
1.37E-5
4.60E-5
1.04
3.04E-6
14.7
0.31
0.0
**************************************************************************
** Rock-fluid Property Section
**************************************************************************
*ROCKFLUID
*RPT
*SWT
** Water-oil relative permeability table.
**
sw
krw
krow
0.12
0.0
1.0
0.82
0.0
0.0
*SLT
**
** Liquid-gas relative permeability table.

sl
0.12
0.15
0.3
0.4
0.5
0.55
0.6
0.7
0.75
0.8
0.88
0.95
0.98
0.999
1.0
krg
0.984
0.98
0.9400002
0.87
0.72
0.6
0.41
0.19
0.125
0.075
0.025
0.005
0.0
0.0
0.0
krog
0.0
0.0
0.0
1.0E-4
0.001
0.01
0.021
0.09
0.2
0.35
0.7
0.98
0.997
1.0
1.0
**************************************************************************
** Initial conditions Section
**************************************************************************
*INITIAL
*USER_INPUT
*PRES *CON 4800.0
** Initial reservoir pressure, bubble
*PB
*CON 4015.0
** point pressure, water and oil saturations
*SO
*CON
0.88
** are all constant throughout.
*SW
*CON
0.12
User's Guide IMEX
**************************************************************************
**************************************************************************
*NUMERICAL
*DTMAX 250.0
*NORM *PRESS 1000.0
*NORM *SATUR
0.2
*AIM *THRESH
**************************************************************************
**************************************************************************
*RUN
*DATE 1986 4 22
*DTWELL 1.00
*GROUP 'g1' *ATTACHTO 'field'
*WELL 1
*WELL 2
'GAS INJECTOR'
'PRODUCER'
*ATTACHTO 'g1'
*ATTACHTO 'g1'
*INCOMP *GAS
*OPERATE *MAX *BHP 27000.
*PRODUCER 2
*OPERATE *MAX
*OPERATE *MIN
*MONITOR *MIN
*MONITOR *GOR
*PERF 1
** if
1
jf
1
**Define gathering centre for voidage replacement.
*STO
*BHP
*STO
20000.0
1000.0
1000.0
20000.0
kf
3
wi
1.0E+5
** Injector 1 matches the voidage from producer 2

** Gas is being injected.
*STOP
*STOP

**
rad
geofac wfrac
*GEOMETRY *K 0.25 0.34 1.0 0.0
*PERF *GEO 2
** if
jf
kf
ff
10
10
1
1.0
*GCONI 'g1'
*VREP *GAS 1.0
*IIP
skin
** Specify voidage replacement.
*TIME 3650.000
*STOP
User's Guide IMEX
MXFRR002.dat: Dual Permeability Radial Coning Model

******************************************************************************
** mxFRR002.dat:
Dual Permeability Radial Coning Model
**
******************************************************************************
** FILE:
MXFRR002.DAT
**
** MODEL: 10x1x15 RADIAL
FRACTURED RESERVOIR
DUAL PERMEABILITY OPTION **
**
GAS CAP
2 ROCK TYPES
UPWARD LAYERING
**
**
FIELD UNIT
**
**************************************************************************************
** A two-layer well is depleting a fractured reservoir with an initial gas cap. **
** Dual permeability is used with matrix connectivity in the vertical direction. **
** Radial coordinate grid is used.
**
**************************************************************************************
**
**************************************************************************************
***********************************************************************
***********************************************************************
*TITLE1
'DUAL POROSITY, PERMEABILITY CONING TEST'
*TITLE2
'Coning Study in Reservoir with Initial Gas Cap'
*INUNIT *FIELD
*WPRN
*WPRN
*WPRN
*WRST
*WSRF
*WSRF
*OUTSRF
*WELL 10
*GRID 10
*ITER *NEWTON
0
*GRID
*TIME
*WELL
1
*GRID
*SO *SG *PRES
*GRID *RADIAL 10 1 15
** Model a radial grid system, 10 x 1 x 15.
***********************************************************************
***********************************************************************
** Grid sizes in I,J, and K directions
*DI *IVAR
** There are 10 grid blocks in the
2.00
2.32
5.01
10.84
23.40 ** I direction; therefore 10 values
50.55
109.21
235.92
509.68 1101.08 ** for I dimension are expected.
*DJ *CON
360.00000
*DK *KVAR
100.0 50.0
2*8.0 14.0
** One value is expected for the J direction.
20.0
16.0
18.0
15.0
19.0
26.0
** 15 values are expected for K dimensions.

12.0 18.0
15.0 20.0
*DEPTH 1 1 1 2000.00
** The depth from a reference block

** (1,1,1) is 2000 ft.
*DUALPERM
*SHAPE *GK
**
**
**
**
**
**
*DIFRAC *CON
50.0
*DJFRAC *CON
50.0
*DKFRAC *CON
50.0
Use the DUAL PERMEABILITY option

Use Gilman & Kazemi equation for shape
factor calculation.
Fracture data, starting with fracture
spacing.
It is constant everywhere.
*POR *MATRIX *KVAR

** Porosity must be the first grid array
0.157 0.116 0.12 0.105 0.12 ** entered. Enter one value for each
0.13
0.14
0.08 2*0.17
** layer.
0.087 0.097 0.111 0.16
0.13
User's Guide IMEX
*POR *FRACTURE *KVAR

2*0.00 13*0.040
** Enter values for the fracture.
*PRPOR *MATRIX 14.7

*CPOR *MATRIX 3.0E-6
** Rock Compressibility
*PRPOR *FRACTURE 14.7

*CPOR *FRACTURE 3.0E-6
** Fracture Compressibility
*PERMI *MATRIX *CON

100.0
*PERMJ *MATRIX *CON
100.0
*PERMK *MATRIX *CON
10.0
** Permeability for the rock matrix.

** Values are constant in all
** directions.
*TRANSI *MATRIX *CON

0.0
*TRANSJ *MATRIX *CON
0.0
*TRANSK *MATRIX *KVAR
2*1.0 0.0 12*1.0
**
**
**
**
**
**
*PERMI *FRACTURE *CON

400.0
*PERMJ *FRACTURE *CON
400.0
*PERMK *FRACTURE *CON
40.0
** The fracture permeability is also

** constant in each direction.
The matrix blocks are connected only in the

vertical direction. Set the transmissibility
multipliers in the horizontal directions to
zero. Vertical will default to 1.0.
multipliers in the horizontal directions to
zero. Vertical will default to 1.0.
*MODEL *BLACKOIL
** Three phase problem.
***********************************************************************
** Component Properties Section
***********************************************************************
*PVT
** p
rs
bo
eg
vo
vg
15.0
5.0 1.050
4.444
3.934
0.0105
115.0
43.0 1.080
40.000
2.776
0.0109
315.0
95.0 1.100 104.998
2.400
0.0113
615.0
160.0 1.126 209.996
2.053
0.0122
915.0
210.0 1.148 320.000
1.810
0.0132
1215.0
265.0 1.170 429.923
1.600
0.0143
1515.0
315.0 1.190 539.957
1.413
0.0153
2015.0
403.0 1.223 739.645
1.196
0.0173
2515.0
493.0 1.257 930.233
0.990
0.0193
3015.0
583.0 1.293 1109.878
0.880
0.0212
3515.0
680.0 1.331 1270.648
0.790
0.0232
4015.0
782.0 1.372 1420.455
0.725
0.0253
4273.0
836.0 1.393 1490.313
0.702
0.0264
4515.0
883.0 1.410 1560.062
0.679
0.0274
5015.0
983.0 1.449 1709.402
0.638
0.0295
5515.0 1083.0 1.487 1988.072
0.602
0.0317
6015.0 1183.0 1.526 2109.705
0.572
0.0338
*DENSITY *OIL
56.70000
*DENSITY *GAS
0.0561600
*DENSITY *WATER
62.40000
*CO
*CVO
*BWI
*CW
*REFPW
*VWI
*CVW
9.2810E-6
4.6000E-5
1.07
3.5000E-6
14.70
0.35
0.0
User's Guide IMEX
*ROCKFLUID
***********************************************************************
** Rock-fluid Properties Section
***********************************************************************
*RPT 1
** Set rock type 1 for the fracture
*SWT
** Sw
krw
krow
0.00
0.00
1.00
0.05
0.05
0.95
0.25
0.25
0.75
0.50
0.50
0.50
0.75
0.75
0.25
0.95
0.95
0.05
1.00
1.00
0.00
*SLT
** Sl
0.00
0.05
0.25
0.50
0.75
0.95
1.00
krg
1.00
0.95
0.75
0.50
0.25
0.05
0.00
krog
0.00
0.05
0.25
0.50
0.75
0.95
1.00
*RPT 2
*SWT
** Sw
0.2
0.25
0.30
0.31
0.35
0.40
0.45
0.50
0.60
0.70
0.80
0.90
1.00
** Set rock type 2 for the matrix.

krw
0.0
0.0004
0.0024
0.0033
0.0075
0.0167
0.0310
0.0515
0.1146
0.2133
0.3542
0.5438
0.7885
krow
1.0
0.6027
0.4490
0.4165
0.3242
0.2253
0.1492
0.0927
0.0265
0.0031
0.0
0.0
0.0
*SLT
** Sl
0.3
0.4
0.5
0.6
0.7
0.8
0.85
0.9
0.95
1.0
krg
0.6345
0.5036
0.3815
0.2695
0.1692
0.0835
0.0477
0.0183
0.0
0.0
krog
0.0
0.00002
0.00096
0.00844
0.03939
0.13010
0.21670
0.34540
0.53020
1.00000
*RTYPE *FRACTURE *CON 1

*RTYPE *MATRIX
*CON 2
Pcow
5.00
4.00
3.00
2.80
2.50
2.00
1.80
1.60
1.40
1.20
1.00
0.50
0.00
Pcog
1.92
1.15
0.77
0.50
0.32
0.22
0.18
0.15
0.12
0.10
** Assign set 1 to the fracture blocks
** and set 2 to the matrix blocks.
*INITIAL
***********************************************************************
***********************************************************************
*PB *FRACTURE *CON 0.0
*PB *MATRIX
*CON 0.0
*REFPRES
3015.0
*REFDEPTH
2000.0
*DWOC
1900.0
*DGOC
1726.0
User's Guide IMEX
*NUMERICAL
***********************************************************************
** Numerical methods Control Section
***********************************************************************
*DTMAX
100.0
*MAXSTEPS
100
*NORM *PRESS
250.
*NORM *SATUR
0.3
*RUN
***********************************************************************
** Well data Section
***********************************************************************
*DATE 1994 01 01
*DTWELL
0.05
*WELL 1 'Producer' *VERT 1 1
*PRODUCER 1
*OPERATE *MAX *STO
500.00
*OPERATE *MIN *BHP 2500.00 *STOP
*MONITOR *GOR
2000.00 *STOP
*MONITOR *BACKFLOW
**
*GEOMETRY
*K
rad
0.25
*STOP
geofac
0.500
**
**
**
**
**
**
**
**
primary operating constraint is

oil rate for an oil producer.
Also monitor bottomhole pressure
and stop when min. of 2500 psi is
reached.
Monitor GOR; stop when reaching
2000 scf/STB.
Stop as soon as backflow occurs.
wfrac
1.00
skin
0.0
*PERFV *GEO 1
** kf
ff
8:9
1.0
*AIMWELL *WELLN
TIME 20.0
*DTWELL 0.25
*ALTER
1
300.0
**
**
**
**
*TIME 50.0
*ALTER
1
500.0
** Well change at 50 days.
*TIME 750.0
*ALTER
1
400.0
** Well change at 750 days.
*TIME 900.0
*STOP
** Stop the simulation at 900 days.
Well change at 20 days.

Size of first timestep after change.
Change the primary (first) operating
constraint for Well 1 to 300 STB/D.
User's Guide IMEX
MXGRO008.dat: Hybrid Refined Grid Model

******************************************************************************
** mxgro008.dat:
Hybrid Refined Grid Model
**
******************************************************************************
** FILE:
MXGRO008.DAT
**
** MODEL: 3x3x3
GAS INJECTION
LOCAL HYBRID REFINED GRID
FIELD UNIT **
**************************************************************************************
** This data set is based on the First SPE Comparative Solution Project. ,J.P.T., **
** 1981(JAN). It is a 3x3x3 grid gas injection problem. The producer is in a hybrid **
** locally refined grid. Note, this results in a different well location than SPE1. **
**************************************************************************************
**
**************************************************************************************
********************************************************************************
**
********************************************************************************
*TITLE1
'Hybrid Grid Test Problem'
*TITLE2
'Modified First SPE Comparative Solution Project'
*TITLE3
'(See Odeh & Aziz, J.P.T., 33, pp.13-25, 1981.)'
*INUNIT *FIELD
*WPRN
*WPRN
*WPRN
*OUTPRN
*OUTPRN
*OUTPRN
*OUTPRN
*WSRF
*WSRF
*WELL 10
*GRID *TIME
*ITER *ALL
*WELL
*BRIEF
*TABLES *NONE
*RES
*NONE
*GRID
*SO *SG *SW *PRES *OILPOT *BPP *IMEXMAP
*GRID 0
*WELL 1
********************************************************************************
**
********************************************************************************
*GRID *CART 3 3 3
** 3 blocks in each the I (x) and J (y) directions.
** 3 layers in the K (z) direction.
*DI *IVAR
** I direction block dimensions
3000.0 4000.0 3000.0
*DJ *JVAR
** J direction block dimensions
3000.0 4000.0 3000.0
*DK *KVAR
** K direction block dimensions (layer thicknesses)
50.0 30.0 20.0
** are 50 ft., 30 ft. and 20 ft. for the bottommost
** middle and uppermost layers, respectively.
*DEPTH 1 1 1 8400.0
******************************
** Input refined grid data
******************************
*REFINE *HYBRID 2 1 1
** Number of refined grids in I direction,
*RW 0.25
** J direction and K direction
*ALPHAI 2256.7600
** default value for rw, set alphai so volume grid
** block equals one-ninth of fundamental grid block
*RANGE 3 3 1:3
** Location of fundamental grid(s) where
** refinement is requested. (around the producer)
User's Guide IMEX
******************************
** These values are the same
** for refined grids and the
** fundamental grids
******************************
*POR *CON 0.3
** Porosity is constant throughout the reservoir.
*CPOR
3.0E-6
*PRPOR 14.7
*PERMI *KVAR
200.0
50.0
*PERMJ *KVAR
200.0
50.0
*PERMK *KVAR
20.0
40.0

500.0
500.0
60.0
**
**
**
**
**
**

reservoir.
*MODEL *BLACKOIL
** Solve three equations.
********************************************************************************
** Component Property Section
**
********************************************************************************
*PVT
** p
rs
bo
eg
viso
visg
14.7000
1.00000
1.06200
6.00000
1.04000 0.0080000
264.700 90.50000
1.15000 82.70000 0.9750000 0.0096000
514.700
180.000
1.20700
159.000 0.9100000 0.0112000
1014.70
371.000
1.29500
313.000 0.8300000 0.0140000
2014.700
636.000
1.43500
620.000 0.6950000 0.0189000
2514.000
775.000
1.50000
773.000 0.6410000 0.0208000
3014.700
930.000
1.56500
926.000 0.5940000 0.0228000
4014.700 1270.000
1.69500 1233.000 0.5100000 0.0268000
5014.700 1600.000
1.82700 1541.000 0.4490000 0.0309000
9014.700 2984.000
2.35700 2591.000 0.2030000 0.0470000
*DENSITY *OIL
46.244
*DENSITY *GAS
0.0647
*DENSITY *WATER 62.238
*CO
1.3687E-5
*CVO
4.6000E-5
*BWI
1.0410
*CW
2.9886E-6
*REFPW 14.7
*VWI
0.31
*CVW
0.0
*ROCKFLUID
********************************************************************************
** Rock-Fluid Property Section
**
********************************************************************************
*RPT 1
*SWT
** sw
krw
krow
0.1200000
0.0
1.0
0.8200000
0.0
0.0
User's Guide IMEX
*SLT
** sl
0.12
0.15
0.30
0.40
0.50
0.55
0.60
0.70
0.75
0.80
0.88
0.95
0.98
0.999
1.0
krg
0.984
0.98
0.94
0.87
0.72
0.6
0.41
0.19
0.125
0.075
0.025
0.005
0.0
0.0
0.0
krog
0.0
0.0
0.0
1.0000E-4
0.001
0.01
0.021
0.09
0.2
0.35
0.7
0.98
0.997
1.0
1.0
*INITIAL
********************************************************************************
**
********************************************************************************
** Use vertical equilibrium calculation.
*PB *CON
4014.7
** bubble point pressure
*REFDEPTH
8400.
** Give a reference depth and
*REFPRES
4800.
** associated pressure.
*DWOC
9500.
** Depth to water-oil contact
*DGOC
7000.
** Depth to gas-oil contact
*NUMERICAL
********************************************************************************
**
********************************************************************************
*DTMAX
1000.
** Maximum time step size
*MAXSTEPS
1000
*NORM *PRESS 1000.0
*NORM *SATUR
0.20
*AIM *THRESH
0.25 0.25 ** Use threshold switching criteria
*PRECC
1.0d-05
** Extra tight convergence of linear solver is
** needed for this problem.
*RUN
********************************************************************************
** Well and Recurrent Data Section
**
********************************************************************************
*DATE 1986 04 00
*DTWELL 1.0
*WELL 1
*WELL 2
'Injector' *VERT 1 1
'Producer' *VERT 3 3
*INJECTOR *UNWEIGHT 1
*INCOMP *GAS
** Define the type of well 1.

** Injecting gas.
*PRODUCER 2
*OPERATE *MAX
*OPERATE *MIN
*STO 20000.0
*BHP 1000.0
*MONITOR *MIN
*STO
**
**
**
**
**
**
**
**
1000.0
*MONITOR *BACKFLOW
*STOP
*MONITOR *GOR 20000 *STOP
User's Guide IMEX
*STOP
** If backflow occurs, stop the simulation.

Define the type of well 2.
Initially, operate on a rate constraint.
If the BHP falls below 1000 psi, switch
to a minimum pressure constraint.
When the oil rate falls below 1000 bbl/D
stop the simulation.
Stop if backflow occurs.
Stop if the GOR exceeds 20000.
*PERF 1
** if
1
jf
1
kf
3
wi
8.5809E+4

**
rad
geofac wfrac
*GEOMETRY *K 0.25
0.34
1.0
*PERFVHY *GEO 2
** kf kr
ff
1
1
1.0
*TIME 3.6500
*STOP
skin
0.0
** Stop simulation at this time
User's Guide IMEX
Appendix A
Theoretical Outline
Four Component Black-Oil Model with Dual Porosity
In this appendix, the governing equations for IMEX are discussed. Section A.1 discusses the
conservation equations used, including those for the dual-porosity model, the polymer option
and the pseudo-miscible option. Section A.2 describes the equations for the dual-porosity,
dual-permeability model. Section A.3 describes the subdomain method for dual-porosity
modelling, and the matrix-fracture coefficient. Finally, in section A.4, the independent and
dependent variables are given.
This version of IMEX is a four-component dual-porosity model which can be used to
simulate miscible flood performances and naturally fractured reservoirs. It is also a three
phase adaptive implicit black-oil simulator.
For the dual-porosity model, IMEX uses either the Multiple-Interacting Continua (MINC)
approach or the subdomain approach. The reservoir is discretized into a primary grid which
represents the interconnecting fractures. The rock mass is represented by secondary grids
discretized in a nested format (MINC) or in a layered format (Subdomain) as shown in
Figure A.1. Besides the MINC approach and the subdomain approach, a dual-permeability
option is also available.
IMEX can be run in two-component (oil-water), three-component (oil-water-gas or oil-waterpolymer), or four-component (oil-water-gas-solvent or oil-water-gas-polymer) mode. The
model solves the mass conservation equations (A.1)-(A.10). Note that the polymer and
solvent equations can not both be solved for the same problem. For the MINC dual porosity
option it is assumed that we have nested matrix blocks from 1 to J as shown in Figure A.1.
Thus, for the outer-most matrix block, k=J, the matrix subblock equations (A.6) to (A.10),
include the matrix-fracture transfer terms. Note that for single-porosity systems, the
equations are the same as the fracture equations, (A.1) to (A.5), without the matrix-fracture
transfer term, and in terms of matrix (reservoir) variables rather than fracture variables.
Conservation of oil in the fracture:
x
n +1
x
x
n +1
n +1
Tof
p of
of
D + Tomf
p om
p of
+ q on +1
User's Guide IMEX
Vb
t
S
o
B o
n +1
S
o
Bo
=0
(A.1)
Appendix A 1041
Conservation of water in the fracture:

x
Twf
p nwf+1 xwf D
x
+ Twmf
+1
p nwm
J
p nwf+1
+ q nw+1
Conservation of gas in the fracture:
x x
n +1
x
x
n +1
x
Tof
R sf p of
of
D + Tgf
p gf
gf
D
V
b
t
n +1
S
w
B w
S
w
Bw
=0
(A.2)
x
x
n +1
n +1
x
n +1
n +1
x n +1
+ Tomf
R smf
p om
p of
+ Tgmf
p gm
p gf
+ R sf
qo
+ q gn +1
V
b
t
R S
fn +1 sf of + E gf S gf
B
of
n +1
Conservation of polymer in the fracture:
x x
x
+1
Twf
C pf p nwf+1 xwf D + Twmf
p nwm
p nwf+1
pS w D e
+
Bw
+ q w C*p
V
b
t
S D
C npf+1 + p w e
B
w
f
R S
fn sf of
Bof
+ E gf S gf
C n +1 C npf+1
pm J
mf
Conservation of solvent in the fracture:
x
x
n +1
x
Twf
R ssf
p nwf+1 xwf D + Tsfx p sf
sf
D
=0
(A.3)
)
)
n +1
n
S C

pS w C p
p w p + A

+ A d = 0
d
B w
B
w
f
(A.4)
x
x
+1
x
n +1
n +1
x
+ Twmf
R smf
p nwm
p nwf+1 + Tsmf
p sm
p sf
+ R ssf
q nw+1
J
J
+ q sn +1
V
b
t
R S
fn +1 ssf wf + E sf Ssf
B
wf
n +1
R S
fn ssf wf
B wf
+ E sf Ssf
=0
(A.5)
Conservation of oil in the matrix subblock k:

x
Tom
k 1 / 2
(p
n +1
om k 1
n +1
n +1 Vb x
p om
+ Tom k +1 / 2 p on +1 p om
k
k
y
t
S
o
Bo
n +1
o
m k Bo
n +1
mk
=0
(A.6)
Conservation of water in the matrix subblock k:

x
Twm
k 1 / 2
(p
n +1
wm k 1
+1
p nwm
k
)+ T
n +1 Vb k S w
n +1
wm k +1 / 2 p w y p wm k
t B w
n +1
w
m k Bw
n +1
mk
=0
(A.7)
Conservation of gas in the matrix subblock k:

x
Tom
k 1 / 2
x
R sm
k 1 / 2
1042 Appendix A
(p
n +1
om k 1
n +1
x n +1
n +1
p om
+ Tom k +1 / 2 R sz
p o y p om k
k
User's Guide IMEX
x
+ Tgm
k 1 / 2
x
R sm
(p
k 1 / 2
n +1
om k 1
n +1
x
p n + 1 p n +1
p gm
+ Tgm
gm k
k
k +1 / 2 g y
n +1
n +1
Vb k n +1 R sS o

n R sS o

=0
+
E
S
+
E
S
g g
g g
mk B
t m k B o
m
m
k
k
Conservation of polymer in the matrix subblock k:

x
Twm
k 1 / 2
C xpm
k 1 / 2
(p
pS w D e
+
Bw
n +1
wm k 1
+1
+1
p nwm
+ Twm k +1 / 2 C xpmz p nwy+1 p nwm
k
k
pS w D e
+1
+1
C npm
C npm
+
k 1
k
m k 1 / 2
Bw
(A.8)
+1
C npy+1 C npm
k
z
n +1
n
Vb k pS w C p
pS w D p

=0
A
d
d
B
t B w
w
mk
m k
)
(A.9)
Conservation of solvent in the matrix subblock k:

x
Twm
k 1 / 2
x
R ssm
x
+ Tsm
k 1 / 2
k 1 / 2
(p
(p
n +1
wm k 1
n +1
sm k 1
+1
x n +1
n +1
p nwm
+ Twm k +1 / 2 R ssz
p w p wm
k
y
k
n +1
x
p n +1 p n + 1
p sm
+ Tsm
sm k
k
k +1 / 2 s y
n +1
n +1
Vb k n +1 R ssS w

n R ssS w
+
E
S
+
E
S
s
s
s
s
=0
mk B
t m k B w
w
mk
mk
(A.10)
In the above expressions, the symbols have the following meanings:

Ad
Bo
Bw
Cp
C* p
D
De
Eg
Es
pl
ql
Rs
Rss
Sl
Tl
t
polymer adsorption
oil formation volume factor
water formation volume factor
polymer concentration
n+1
polymer concentration for well: Cpf for producers, input Cp for injectors depth
depth
dispersion coefficient. Input dispersion coefficient times geometric factor as in
transmissibilities times transmissibility multiplier
gas expansion factor
solvent expansion factor
pressure in phase 1 = oil, gas, water, solvent
well rates for phase 1 = oil, gas, water, solvent
solution gas-oil ratio
solution solvent-water ratio
saturation of phase l = oil, gas, water, solvent
transmissibilities of phase l = oil, gas, water, solvent
time
User's Guide IMEX
Appendix A 1043
specific weight of phase l = oil, gas, water, solvent

porosity
polymer accessible pore volume
Superscripts
n
old time level
n+1 new time level
x
old or new time level
Subscripts
f
fracture
g
gas
J
nested matrices are numbered from 1 to J where J is the outermost matrix
block connected to fracture
k
submatrix block index between 1 to J
m
matrix
o
oil
w
water
s
solvent
y
equal to f when k=J, it is equal to mk+1 when kJ
z
equal to mk, mk+1, or f, depending on which is the upstream
x
of
The fracture transmissibility T
in the direction X for the flow term from i to i+1 is:

x
x
Tof
i +1 / 2
k
Y Z
=
ke
ro
i +1 / 2 o B o f
(A.11)
k
where k is i or i+1 depending on the potential between the two blocks, such that the mobility
is always upwind weighted. ke is thex effective fracture permeability which is kf f. The
matrix-to-fracture transmissibility T omf can be written as:
x
Tom
j+1 / 2
AmJ
k x
=
k m J ro
d m J / 2
o Bo t
(A.12)
where t is equal to f or mJ depending on whether the fracture or the matrix is upstream. The
matrix-to-matrix transmissibility Tom for the flow term from matrix block j to matrix block
j+1 can be written as:
x
x
Tom
j+1 / 2
k
A
ro
= m k m
dm
j +1 / 2 o B o m k
(A.13)
is the matrix-fracture transfer coefficient which will be discussed later. Am and dm are the
surface areas and nodal distances for the nested matrix elements. The nodal distance for the
innermost matrix element is determined by assuming quasi-steady state flow within the block
as described in Warren and Root ("The Behavior of Naturally Fractured Reservoirs," SPEJ,
Sept. 1963, pp. 245-253). km is the matrix permeability.
1044 Appendix A
User's Guide IMEX
The gas saturation is related to the water and oil saturations through the saturation constraint
Sg = 1 - Sw So (A.14)
or when the pseudo-miscible option is used, the saturation constraint becomes
Sg = 1 - Sw - So Ss (A.15)
The individual phase pressures are related through the capillary pressures Pcl:
pw = po pcw (A.16)
pg = po + pcg (A.17)
The equations are discretized into finite-difference form using first order backward
differences in time, and central differences in space with upstream mobilities.
The polymer formulation is further described in Appendix E.
Dual Permeability
The dual-permeability model is an enhancement to the standard dual-porosity model. In this
model, the communication between the matrices (the intergranular void space which is also
referred to as the primary porosity) is not assumed to be negligible. The differences in terms
of interblock communication for the standard dual porosity model, the dual permeability
model, and the MINC model are shown in Figure A.2.
The equations for matrix flow (A.6-A.10) need to be modified to include matrix-matrix flow
terms as follows:
Conservation of oil in the matrix:
V S
x
n +1
x
x
n +1
n +1
Tom
p om
om
D + Tomf
p om
b o
p of
t B o
n +1
o
m
Bo
n

= 0
m
(A.18)
Conservation of water in the matrix:

x
Twm
+1
p nwm
xwm D
x
+ Twmf
n +1
n +1
p om
p of
V
b
t
S
w
B w
Conservation of gas in the matrix:
x
x
n +1
x
x
n +1
x
Tom
R sm
p om
om
D + Tgm
p gm
gm
D
)
)+ R
x
x
n +1
n +1
x
n +1
n +1
+ Tomf
R smf
p of
p om
+ Tgmf
p gf
p gm
n +1
w
m
Bw
n

= 0 (A.19)
m
x n +1
sm q o
n +1
n
R S
R S

V
b nm+1 sm om + E gmS gm
nm sm om + E gmS gm = 0 (A.20)
t
Bom
Bom

Conservation of polymer in the matrix:
x
n +1
x D + T x p n + 1 p n + 1
Twm
C xpm p om
om
omf of
om
pS w D e
+
Bw
User's Guide IMEX
+1 pS w D e
C npm
+
m
Bw
)
(
+1
C npf+1 C npm
mf
)
Appendix A 1045
V
b
t
n +1
n +1
S C
S w C p

w p + A d
+ A d = 0
B
B w
w
m
m
Conservation of solvent in the matrix:
x
x
+1
x
n +1
x
Twm
R ssm
p nwm
xwm D + Tsm
p sm
sm
D
)
)+ R
x
x
+1
x
n +1
n +1
+ Twmf
R ssmf
p nwf+1 p nwm
+ Tsmf
p sf
p sm
(A.21)
x
n +1
ssm q w
n +1
n

Vb n +1 R ssmS wm
n R ssmS wm
m
+
E
S
E
S
sm sm
m
sm sm = 0
B
t
B wm
wm
(A.22)
The fracture equations are the same as (A.1-A.5) except that the subscript J is not required.
x
om
The matrix transmissibility T
in the direction X for the flow term from i to i+1 is

x
x
Tom
i +1 / 2
YZ
km
ro
=
i +1 / 2 o B o m
x
(A.23)
k
where k is i or i+1 depending on the potential between the two blocks such that the mobility
is always upwind weighted. km is the harmonic mean of the matrix absolute permeability.
Subdomain Model
Besides the MINC approach and dual permeability, another enhancement to the standard
dual-porosity model is called the subdomain method. In essence, this model discretizes the
matrix into a layered format to improve the representation of gravity effects in the matrixfracture transfer calculation. More details of this method can be found in Fung and Collins,
"An Evaluation of the Improved Dual-Porosity Model for the Simulation of Gravity Effects
in Naturally Fractured Reservoirs," CIM paper no. 88-39-05 presented at the 39th Annual
Technical Meeting of the Petroleum Society of CIM held in Calgary, June 12-16, 1988.
Matrix-Fracture Transfer Coefficient, sigma

The transfer coefficient, , in equations (A.12) and (A.13) is calculated as follows:
1. All matrix subdivisions of MINC except the innermost subdivision:
= MINC =
XYZ
LxLyLz
(A.24)
where X, Y, and Z are the grid block dimensions and Lx, Ly, and Lz are the
matrix block dimensions.
2. For the innermost subdivision, or when one-matrix one-fracture system is defined,
either (a) Gilman and Kazemi or (b) Warren and Root's approach is used:
a) Gilman and Kazemi (1983) used the following matrix-fracture transfer
coefficient, , ("Improvements in Simulation of Naturally Fractured
Reservoirs," SPEJ, August 1983, pp. 695-707)
1046 Appendix A
User's Guide IMEX
k
ky
k
k KG = 4 x2 + 2 + z2 XYZ
Lx
Ly
Lz
(A.25)
when substituted into (A.12) and (A.13) give:

k
ky
k
x
= 4 x +
+ z
Tomf
2
2
Lx
Ly
Lz 2
x
Tom
j+1 / 2
XYZ k ro
o Bo t
k
ky
k
= 4 x2 + 2 + z2
dx
dy
dz
dxdydz MINC k ro
B
o o t
b) Warren and Root (1963) assumed quasi-steady state flow and arrived at the
following form of shape factors for 3 orthogonal sets of plane fractures with
constant matrix absolute permeability in all direction ("The Behavior of
Naturally Fractured Reservoirs," SPEJ, Sept. 1963, pp. 245-255).
2
WR
20 1 1 1
=
+ + XYZ (A.26)
3 Lx Ly Lz
when substituted into (A.12) and (A.13) give:

x
x
Tomf
=
k
20 1
1
1
XYZk m ro
+
+
3 Lx Ly Lz
o Bo t
2
x
Tom
j+1 / 2
k
20 1
1
1
=
+
+ dxdydz MINC k m ro
3 dx dy dz
o Bo t
where dx, dy, and dz are the dimensions of the innermost subdivision of
MINC.
Independent and Dependent Variables

The independent variables at each grid node are oil pressure po, water saturation Sw, oil
saturation So in the presence of free gas when the gas option is on, polymer concentration Cp
when the polymer option is on, and solvent saturation Ss in the presence of solvent when the
pseudo-miscible option is on. When no free gas is present pB, saturation (bubble point)
pressure, replaces So as an independent variable. Similarly, when no free solvent is present
pBs, saturation pressure of solvent in water, replaces Ss as an independent variable. All other
variables are functions of these two, three, or four independent variables. The gas saturation
is a dependent variable, as can be seen from equation (A.11). The gas, solvent, and polymer
options are set using the *MODEL keyword in the component properties section.
Each independent variable is associated with one of the conservation equations. In the twocomponent mode (oil-water) the oil and water equations are used with po and So as primary
variables. For the three-component modes (oil-water-gas, or oil-water-polymer) the equations are
oil, water and gas; or oil, water and polymer with primary variables po, Sw, So or pb; or po, So, Cp.
In the four-component modes, the equations are oil, water, gas and solvent or oil, water, gas and
polymer with primary variables po, Sw, So or pb, and Ss or pbs; or po, Sw, So or pb, and Cp.
User's Guide IMEX
Appendix A 1047
In dual-porosity simulation, the fracture equations are (A.1)-(A.5) and the corresponding
variables are pof, Swf, Sof or pBf, Cpf and Ssf or pBsf. The matrix equations are (A.5)-(A.8) and
the corresponding variables are pom, Swm, Som or pBm, Cpm and Ssm or pBsm. If nested matrix
blocks are used, each matrix node will be represented by a set of equations (A.6)-(A.10).
(a): MINC-partitioning
Matrix
Fracture
(b): Subdomain partitioning

Figure A-1
1048 Appendix A
User's Guide IMEX
(a) Standard dual-porosity model

Fractures
Matrices
(b) Dual-permeability model

Fractures
Matrices
(c) MINC model

Fractures
Matrices
Figure A-2
Schematic diagrams of connectivity for the standard dual-porosity model,
the dual-permeability model, and the MINC model
User's Guide IMEX
Appendix A 1049
Appendix B
Well Model
There are two types of injection wells available. Each is capable of injecting oil, (free) gas,
(free) water, and solvent. An injector well may be mobility weighted or unweighted. There
is also a production well model. This accounting leads to nine distinct well types, indicated
by the choice of keywords available: *PRODUCER, *INJECTOR, and *INCOMP.
Total Mobility Weighted Injectors

*INJECTOR *MOBWEIGHT
The well rate for the single injected phase is calculated as follows, where 1
denotes oil, water, gas, or solvent.
q l = [WI T (p BH + head p i ) / B l ]
j = o, w , g , s
ily
Unweighted Injectors
*INJECTOR *UNWEIGHT
The well rate for the single injected phase is calculated as follows, where l
denotes oil, water, gas, or solvent.
q l = [WI(p BH + head p i )]
ily
Well Injectivity
The well injectivity is user set with any of the well completion keywords, *GEOMETRY,
*PERF, *PERFRG, and *PERFV and has the units of (fluid volume)/(time)(pressure). The
equation used to calculate well injectivity is:
WI = 2k h f fh / (ln (re / rw ) + S)
*GEOMETRY keywords are used with mobility weighted injectors and producers only. Well
indices are entered directly by the user if unweighted injectors are used.
User's Guide IMEX
Appendix B 1051
Producers
*PRODUCER
The well rate for the four phases oil (o), gas (g), water (w), solvent (s) are
calculated as follows:
q o = [WP o (p BH + head p i ) / B o ]
ily
q g = WP g (p BH + head p i )E g + R s q o
ily
q w = [WP w (p BH + head p i ) / B w ]
ily
q s = [WP s (p BH + head p i )E s + R ss q w ]
ily
The well productivity is calculated using the formula:

WP = 2k h f fh / (ln (re / rw ) + S)
In the above,
ql
surface flow rate of phase l=o,g,w

sum over blocks in the well completion
ily
pBH
bottom hole well flowing pressure
head
wellbore head
pi
block pressure
l = (k rl / l )
mobility of phase l=o,w,g,s in the block
T = o + g + w + s
total block mobility
Rs
solution gas/oil ratio
Rss
solution solvent/water ratio
Once a well type has been set, the blocks in the completion must be determined along with
the well injectivities and productivities. You may either allow IMEX to calculate these
internally by using *GEOMETRY to enter the well geometry elements and use any of the
well completion keywords with *GEO, or you use the well completion keywords by
themselves and enter your own injectivities and productivities.
Wellhead Pressure Wellbore Model

A simplified wellbore model is included in order to apply wellhead pressure constraints. A
single phase wellbore model is used for injectors similar to Aziz et al (1972). For producers a
table look-up approach is used.
1052 Appendix B
User's Guide IMEX
INJECTORS
A single phase wellbore model is used for injectors.

The equation governing the change in pressure in the direction of flow used for the wellbore
model is
p = PH - pKE - pF
where
p H = g z
p F =
(Hydrostatic head over the length z)
2fv
z (Friction gain over a length z)
D
2
p
p KE = v 2 ln 2
p1
(Kinetic energy gain)
Here
g
f
v
D
p = p2 - p1
=
=
=
=
=
=
density of the flowing " in situ" mixture

acceleration due to gravity
fanning friction factor
average velocity of mixture
inside pipe diameter
pressure drop over the length z
PRODUCERS
For producers, a table look-up procedure is used. Aziz et al (1972) was not used for
producers as the phase behavior and fluid properties in that method are inconsistent with
those used within the simulator. Aziz et al used simple analytic correlations while, in
general, black oil simulators use isothermal tables to calculate fluid properties and phase
behavior. Within the wellbore, however, thermal effects may be very important. Thus for
producers, a table look-up is used where bottom hole pressure is calculated as a function of
wellhead pressure, oil rate, GOR and water cut. Also, a reference depth is given for each
table and a head adjustment is made in order to account for the difference between table
depth and the bottom hole well depth.
REFERENCES
Aziz, K., Govier, G.W., and Fogarasi, M., "Pressure Drop in Wells Producing Oil and Gas,"
JCPT, July-Sept., 1972.
User's Guide IMEX
Appendix B 1053
Appendix C
Calculation of Well Parameters

x
m
Assume incompressible, single-phase, steady state flow, with no skin effects.
If it is assumed that the pressure calculated at the points j, k, l and m are exact then these
pressures may be related to the pressure in the wellbore by the radial flow equation (derived
from radial form of Darcy's law).
p j p w = [q / 2kh ] ln (x / rad ) (C.1)
where
pj
pw
q
k
h
rad
=
=
=
=
=
=
=
pressure at node j
pressure at wellbore
flow rate
viscosity
permeability
height of grid block
wellbore radius
The finite-difference equation for inflow into grid block i (derived from the linear form of
Darcy's law) can be written:
(pj - pi) + (pk - pi) + (pl - pi) + (pm - pi) = q / kh
(C.2)
Noting that from symmetry considerations (pj - pi) = (pk - pi) = (pl - pi) = (pm - pi), then the
above can be written as:
(pj - pi) = q / 4 kh
User's Guide IMEX
(C.3)
Appendix C 1055
Eliminating pj, we obtain
p i p w = [q / 2kh ] (ln x / rw / 2) = [q / 2kh ] lne / 2 x / rad (C.4)

Comparing this to the well model we see that
re = e / 2 x 0.21x
Therefore, since
re = geofac xy / (wfrac) = geofacx / wfrac ,

for this case, geofac .37.
Similar arguments show that geofac .5 for a radial coning problem.
The appropriate geometric factor can be derived for other geometries using a similar
procedure and the method of images. The results for some common cases are shown in
Figure C.1. For more details, see Peaceman, 1978 ("Interpretation of Well Block Pressures in
Numerical Reservoir Simulation," SPEJ, June 1978, pp. 183-194).
EXAMPLE: WELL INJECTIVITY INDEX
If mobility weighted injectors are used, then the following are required:
k
absolute permeability of the grid block calculated as k = (k1 k2)1/2 where k1
and k2 are the two absolute permeabilities in the directions normal to the
wellbore
h
grid block (formation) thickness
wfrac
well fraction (1 for wells near block centres, 1/2 for half wells on boundaries
and 1/4 for wells at block corners)
geofac
geometric factor
re
effective well radius is given as re = geofac(area/wfrac )1/2 where area denotes
the area of the block face normal to the wellbore and CC is a geometric factor
described in Appendix C
rad
wellbore radius
skin
skin factor
1056 Appendix C
User's Guide IMEX
Note that k, h, and re are calculated in IMEX once the direction normal to the well is known
(set by *GEOMETRY or the set of *PERF keywords). The remaining parameters are user set.
If unweighted injectors are used, then a single injectivity is read and the injectivity is
allocated to all layers with weights based on (kh).
Often an estimate of the formation permeability is available, but no injectivity index has been
measured. The well injectivity index can then be estimated as follows.
If a single fluid is injected, and assuming that only irreducible saturations of the other fluids
are present in the vicinity of the wellbore, the well flow rate is given by
q l = l h (p w p i )
wfrac2k
re
+ skin
ln
rad
(C.5)
Converting to surface conditions gives
q ls = l
ls
wfrac2k
l h (p w p i )
re
ln rad + skin
(C.6)
Hence the well injectivity index is,
( )
~
wfrac2k
WIl = l l Sl h
re
ls
+ skin
ln
rad
(C.7)
The terms in the above equations can be defined as follows:

ls
l
S% l
wfrac
h
re
rad
k
skin
pw
pi
User's Guide IMEX
surface density of the injected liquid

mobility of the injected liquid
saturation at steady state in the formation near the well
well fraction
formation thickness
effective radius = geofac xy / (wfrac)
wellbore radius
absolute permeability
skin factor
wellbore pressure
grid block pressure
Appendix C 1057
wfrac = 1
geofrac = 0.37
wfrac = 1
geofrac = 0.34
wfrac = 1
geofrac = 0.35
(a) Centre of
Square Grid Block
(b) Centre of Co rner

Block
(c) Centre of Edge

Block
wfrac = 1
geofrac = 0.41
(d) C entre Rectangular Grid

Block R atio x/ y=2
geofrac = 0.39
(e) Centre Rectangular Grid

Block Ratio x/ y=3/2
i
i
wfrac = 1
wfrac = 1/4
geofrac = 0.64
wfrac = 1/4
geofrac = 0.56
wfrac = segment
geofrac = 0.5
(f) W ell in Corner
(g) Edge Blocks

Half Size
(h) Radial Segm ent

wfrac = 1
2 radian
radian
wfrac = 1
1/2 radian wfrac = 1/4
wfra c = 1/2
geofrac = 0.54
wfrac = 1/2
geofrac = 0.47
(i) W ell on Edge
(j) Edge Blocks

Half Size
1058 Appendix C
User's Guide IMEX
Appendix D
Figure D-1a: Numbering of the Grid (Non-Corner Point)

Z
(24)
4, 3, 2
(23)
(20)
3, 3, 2
(22)
4, 2, 2
(19)
2, 3, 2
(21)
(16)
3, 2, 2
(18)
1, 3, 2
3, 1, 2
(14)
1, 2, 2
4, 1, 2
(15)
2, 2, 2
(17)
2, 1, 2
(13)
1, 1, 2
(12)
4, 3, 1
(11)
(8)
3, 3, 1
(10)
4, 2, 1
(7)
2, 3, 1
(9)
(4)
3, 2, 1
(6)
1, 3, 1
(2)
1, 2, 1
(1)
1, 1, 1
User's Guide IMEX
(3)
2, 2, 1
(5)
4, 1, 1
3, 1, 1
2, 1, 1
*Numbers in parenthesis
indicate the gridblock numbers
Appendix D 1059
Figure D-1b: Numbering of the Grid (Corner Point)

i
(4)
4, 1, 1
(3)
(8)
3, 1, 1
(2)
4, 2, 1
(7)
2, 1, 1
(1)
(12)
3, 2, 1
(6)
1, 1, 1
3, 3, 1
(10)
1, 2, 1
4, 3, 1
(11)
2, 2, 1
(5)
2, 3, 1
(9)
1, 3, 1
j
(16)
4, 1, 2
(15)
(20)
3, 1, 2
(14)
(19)
2, 1, 2
(13)
(18)
4, 2, 2
(24)
3, 2, 2
1, 1, 2
(17)
(22)
1, 2, 2
(21)
1, 3, 2
1060 Appendix D
(23)
2, 2, 2
4, 3, 2
3, 3, 2
2, 3, 2
*Numbers in parenthesis
indicate the gridblock numbers
User's Guide IMEX
Figure D-2: Cylindrical Reservoir
(24)
4, 2, 3
(23)
3, 2, 3
Z
q
(17)
(18)
(19)
1, 1, 3
2, 1, 3
3, 1, 3
(9)
(10)
(11)
(12)
1, 1, 2
2, 1, 2
3, 1, 2
4, 1, 2
(1)
(2)
(3)
(4)
1, 1, 1
2, 1, 1
3, 1, 1
4, 1, 1
User's Guide IMEX
(20)
4, 1, 3
Appendix D 1061
Figure D-3: Calculation of Gravity Components for a Tilted

Grid System
Z
X
Y
= dip in x direction
= dip in y direction
GX = sin
GY = sin
GZ = 1 sin 2 sin 2
1062 Appendix D
User's Guide IMEX
Figure D-4: Radial (Cylindrical) Coordinates

Z
horizontal
axis
g
=0
GR = sin
GZ = 1 sin 2
User's Guide IMEX
Appendix D 1063
Figure D-5a: Depth Measurement

Surface
Depth
1, 1, 3
3, 1, 3
2, 1, 3
1, 1, 2
2, 1, 2
1, 1, 1
2, 1, 1
3, 1, 2
3, 1, 1
4, 1, 3
5, 1, 3
6, 1, 3
4, 1, 2
5, 1, 2
6, 1, 2
4, 1, 1
5, 1, 1
6, 1, 1
7, 1, 3
8, 1, 3
7, 1, 2
8, 1, 2
7, 1, 1
8, 1, 1
1064 Appendix D
User's Guide IMEX
Figure D-5b: 3-Dimensional Variable Thickness Grid System

(8x3x2)
z
y
x
User's Guide IMEX
Appendix D 1065
Figure D-6: Fault Throw Definition

SURFACE
THROW
1066 Appendix D
User's Guide IMEX
Figure D-7: Typical Data Curves
RS
BO
CO = -tan/BO(Pb)
EG
VISO
CVISO = tan
VISG
User's Guide IMEX
Appendix D 1067
Figure D-8: Typical Relative Permeability Curves
Water-Oil Relative
Permeability Curves
Gas-Liquid Relative
Permeability Curves at
Connate Water Saturation
(Swr + So)
1068 Appendix D
User's Guide IMEX
Appendix E
Polymer Option
Conventional waterflooding operations can significantly increase the recovery of hydrocarbons
from a reservoir. Due to unfavorable mobility ratios and reservoir heterogeneities, waterflooding
can result in poor and incomplete sweeps of the reservoir volume. One method for improving
waterflood performance is including additives in the injected water so that the viscosity of the
injected water will increase and result in more favorable mobility ratios. The mobility ratio is
defined as the mobility of the displacing phase to the mobility of the displaced phase. In
waterflooding,
M w o =
( )
( )
w k rw S w / w
=
o k ro (S wc ) / o
(E.1.1)
where k rw S w is the relative permeability to water at the average water saturation behind
the waterfront and kro(Sw) is the relative permeability to oil ahead of the front at irreducible
water saturation.
Several types of additives have been developed for use in waterflood projects. Water-soluble
polymers are the most popular choice due to their lower cost compared to other types of
additives. In addition to an increase in mobility ratio, other factors and mechanisms have
been determined to influence the performance of polymer floods.
Factors and Mechanisms Affecting Polymer Floods

Figure E.1.1 shows a schematic representation of a polymer flood. A slug of polymer solution is
injected into the reservoir which has already had a prior injection of low salinity brine or fresh
water. The polymer slug is followed by an additional slug of fresh water and by continuous drive
water injection. Note that the polymer solution slug is injected between the two fresh (low
salinity) water buffers in order to reduce the direct contact with the saline reservoir water, since
high water salinity can significantly reduce the viscosity of the polymer solution.
Experimental work from previous studies have shown that polymer flooding improves the
recovery over conventional waterflooding by increasing the reservoir volume contacted. The
polymer solution injected has no effect on the residual oil saturation. Also, polymer flooding
accelerates oil production, resulting in a higher recovery at breakthrough. Polymer flooding
is more successful when applied during the early stages of a waterflood process, when the
mobile oil saturation is high. A polymer flood will not be useful when applied to uniformly
waterflooded fields with low oil viscosity or for fields with high water saturation. Reservoirs
User's Guide IMEX
Appendix E 1069
with high permeability variations or rapid water breakthrough are also good candidates for
polymer floods but generally pose a high degree of risk.
There are five major aspects of a polymer flood that need to be rigorously represented in a
numerical model. They are:
1. Mobility Control
2. Polymer Retention
3. Physical Dispersion
4. Inaccessible Pore Volume
5. Apparent Viscosity and Resistance Factor
These factors are all very important when predicting the performance of a polymer flood,
however, the most important benefits are due to the decrease in mobility ratio and the
increase in apparent viscosity due to polymer adsorption. The effectiveness of polymer
floods is reduced due to dispersion and reservoir permeability variation. The five factors
listed above are described in detail below.
Mobility Control
An important aspect of polymer flooding is the improvement of the volumetric sweep

efficiency and the increase in the microscopic displacement efficiency. The role of watersoluble polymers is to increase the water viscosity and also to reduce the permeability of the
rock to water. Directly, this results in the reduction of the water-oil mobility ratio given in
(E.1.1) close to unity or less. Then, the volumetric sweep efficiency will be improved and a
higher oil recovery will be achieved at breakthrough compared to conventional waterflooding.
The reduction in permeability and higher water viscosity will result in an increase in the
resistance to flow. This will divert the polymer solution toward areas unswept by water.
injector
oil
fresh water
polymer solution
water
producer
1
3
4
Figure E.1.1: Schematic View of Polymer Flood
1070 Appendix E
User's Guide IMEX
The polymer mass balance equation given in (E.1.12) contain a term which represents the
water-polymer transmissibility
A
Tw = w f
l
(E.1.2)
where Af is flow area between adjacent grid blocks and l is the distance between grid block
centres. The term w is a modified mobility to water in the presence of polymer defined as:
w =
k k rw
w
(E.1.3)
This modified mobility incorporates a reduced permeability k due to polymer retention/loss

as a result of rock interaction and a modified water viscosity w which accounts for the
increase in viscosity of the polymer-water solution.
The reduction in permeability is due to the mechanism of polymer retention. As the polymer
solution interacts with the reservoir rock, polymer is adsorbed or desorbed from the rock
surface. The variation in rock permeability due to this process is given by
k=
k
Rk
where k is the absolute rock permeability prior to polymer flooding and Rk is a permeability
reduction factor. Rk is a function of polymer adsorption and the residual resistance factor
RRF. Rk is given by the expression
R k = 1.0 + (RRF 1.0)
Ad
A d max
(E.1.4)
where Ad is the cumulative adsorption of polymer per unit volume of reservoir rock and
Admax is the maximum adsorptive capacity of polymer per unit volume of reservoir rock.
The maximum adsorptive capacity of the reservoir rock and the residual resistance factor are
functions of reservoir rock permeability.
The polymer solution viscosity, w , is the second parameter for mobility reduction. The
modification of the viscosity accounts for the mixing of polymer and water. Note that the
resulting viscosity is dependent on the concentration of polymer in the solution . Different
mixing rules can be employed to calculate the mixed solution viscosity. Two commonly used
mixing models are the Linear-Mixing Model which is given by
w = op + (1 ) w
(E.1.5)
and a Nonlinear Mixing Model which is given by

= o (1 )
(E.1.6)
w
User's Guide IMEX
Appendix E 1071
In the above expressions (E.1.5) and (E.1.6), op is a reference polymer solution viscosity at a
reference polymer concentration. is a parameter dependent on polymer concentration given as:
=
Cp
C op
where Cp is the current polymer concentration and C op is the reference polymer

concentration.
Polymer Retention
The process of polymer retention consists of two separate mechanisms. These mechanisms,
whose effects are difficult to separate are adsorption of the polymer on rock surfaces and
entrapment of polymer molecules in small pore spaces. Both these mechanisms have the effect
of increasing the resistance to flow by essentially a permeability reduction. It is difficult to
quantify what percentage of injected polymer is adsorbed and what percentage is trapped in
small pore spaces since only the produced polymer concentration can be measured. Both these
mechanisms result in a loss of polymer to the reservoir. Of the two commonly used polymers
for field applications, polyacrylamide polymers adsorb on the surface of most reservoir rock.
Polysacchrides are not retained on rock surfaces and consequently, do not exhibit a residual
resistance effect. Calcium carbonate has a greater affinity for polymer than does silica and will
exhibit a higher adsorption. The porous space within reservoir rock offers a variety of opening
sizes. The long chain of the polymer molecule can flow through into large pore openings but
can become trapped when the other end has a smaller opening. Entrapment can also take place
when the flow is restricted or stopped. When this happens, the polymer molecule loses its
elongated shape and coils up. Desorption of the polymer from the reservoir rock can also take
place if sufficient polymer has already been adsorbed above a residual sorption level. This
residual sorption level is basically a function of the reservoir permeability. Assuming
equilibrium sorption with the reservoir rock, the sorption phenomenon can be described as
( )
Ad = f Cp
(E.1.7)
where f (Cp) is a function of polymer concentration only. The function f (Cp) is specified in
the form of a table. It should be noted that field observations indicate the higher the polymer
concentration before flowing through the porous space, the higher will be the adsorption on
the rock surface. Consequently, Ad increases with increasing polymer concentration Cp.
Physical Dispersion
Movement of a polymer bank through the porous medium results in mixing, primarily due to
physical dispersion. This mixing is characterized by longitudinal and transverse mixing
dispersivities which are multiplied by the fluid velocities to give local dispersion coefficients.
Molecular diffusion, although not very important, can also be included as a component of the
local dispersion coefficient. The effective dispersion coefficient, which is the required input
data, can be written by a two-parameter model of the form
D
D e = ~m + v w
l
1072 Appendix E
(E.1.8)
User's Guide IMEX
where
De
Dm
~
l
= effective dispersion coefficient

= molecular diffusion coefficient
= tortuosity
vw
= local dispersion parameter tensor

= water velocity vector
Note that since fluid motion is normally not parallel to one of the coordinate direction, the
dispersion terms form tensor quantities. Because this introduces cross derivatives into the
governing flow equations for the polymer solution, (E.1.8) can be simplified by neglecting
the off-diagonal terms in the tensoral quantities:
Vw
D
D e, x , y, z = ~m + x , y, z
l
where De,x,y,z and x,y,z are the effective dispersion coefficient and local dispersion
parameter respectively in the x, y and z directions and V w is the average Darcy velocity.
The major purpose for incorporating dispersive mixing is to simulate the smearing of the
trailing and leading edges of the injected polymer bank. The polymer slug propagates
through the porous medium at a velocity different from that of the water. Adsorption tends to
move the front edge of a polymer slug at a slower velocity than the water bank, and the
inaccessible pore volume tends to move the polymer slug at a higher velocity than the water
bank. The combination of these two effects results in a smaller slug that is shifted forward.
Consequently, further simplification results in a constant value of the effective dispersion
coefficient varying only spatially throughout the reservoir. Thus the dispersion becomes
indirectly dependent on the rock properties, namely the porosity and absolute permeability.
Inaccessible Pore Volume
When the flow of polymer molecules through the porous medium is restricted in pores with
small openings, only the passage of water or brine is permitted. These small openings not
contacted by flowing polymer molecules form what is called inaccessible pore volume (IPV).
This phenomenon of IPV was first reported by Dawson and Lautz (1972)1. They showed that
all the pore spaces may not be accessible to polymer molecules and that this allows polymer
solutions to advance and displace oil at a faster rate than predicted on the basis of total porosity.
Up to 30% of the total pore volume may not be accessible to polymer molecules. As a result,
the effective porosity for a polymer solution is less than the actual reservoir porosity. A
reduced polymer porosity p, can be defined to represent the available pore volume to polymer
solutions as:
p = (1.0 IPV )
(E.1.9)
where is the original porosity.
User's Guide IMEX
Appendix E 1073
The IPV can have beneficial effects on field performance. The rock surface in contact with
the polymer solution will be less than the total pore volume, thus decreasing the polymer
adsorbed. More importantly, if connate water is present in the smaller pores inaccessible to
the polymer, the bank of connate water and polymer-depleted injection water that precedes
the polymer bank is reduced by the amount of inaccessible pore volume. One drawback
however, is that movable oil located in the smaller pores will not be contacted by the polymer
and therefore may not be displaced.
Apparent Viscosity and Resistance Factor
The effect of polymers in aqueous solutions is to increase the viscosity for better mobility
control as mentioned earlier. Observations of the behavior of low concentration polymer
solutions in a porous medium suggest a much higher viscosity than that measured in the lab.
In fact, the viscosity of the polymer solution flowing in the reservoir is 5 to 25 times higher
when calculated indirectly using Darcy's law, assuming the same effective permeability. This
effect is shown in Figure E.1.2. In reality however, it is the effective permeability of the
formation to the polymer solution which is lower than it is to water without polymer. It is
difficult to separate the effects of permeability reduction from that of the viscosity increase.
The total effect can be measured as a reduction in mobility. The measure of the mobility
reduction, R can be expressed as:
R=
w k rw p
=
p
w k rp
(E.1.10)
where w and p are the water mobility and water-soluble polymer mobility respectively and
the other quantities are as defined previously. A plot of the resistance factor, R, as a function
of pore volume injected is shown in Figure (E.1.3) for a 300 ppm polymer solution flowing
through a core sample. The plot shows that R rapidly increases and stabilizes to a constant
value. This behavior is important since injection pressures will not have to be increased in
order to sweep the polymer bank through the reservoir. Polymers with high resistance factors
are useful for profile improvement by plugging more permeable streaks near injectors and
reducing the variation in permeability.
The residual resistance factor Rr measures the reduction of the reservoir rock's permeability
to water after polymer flow and is defined as:
Rr =
(k rw / w )before polymer flow

(k rw / w )after polymer flow
(E.1.11)
This residual resistance effect is due to the alteration of the original rock permeability by
adsorption of the polymer and mechanical entrapment of polymer molecules. The residual
resistance effect also has economic importance. Expenditures for polymer occur only during
the injection period. Long after the polymer injection phase, the residual resistance effect
continues at no additional expense. It should be noted that polysaccharide polymers are not
retained on rock surfaces and consequently do not exhibit the residual resistance effect.
1074 Appendix E
User's Guide IMEX
Polymer Types
There are two principal types of polymers that are currently being used. The first type is a
synthetic polymer, called hydrolyzed polyacrylamide's (HPAM) are obtained by the
polymerization of the acrylamide monomer. Through hydrolysis, some of the acrylamide
monomers are converted to carboxylate groups with a negative charge. This process results
in a higher molecular weight and a linear chain molecular structure which helps ease the flow
through the tortuous porous space of the rock reservoir. Electrolyten in saline water cause the
long chain molecules to coil and obstruct the flow and reduce solution viscosity. Hydrolyzed
polyacrylamide solutions, generally are sensitive to salts and dissolved solids and must be
prepared with fresh water. To be effective, polymer solutions must remain stable for a long
time at reservoir conditions. The presence of oxygen can be a source of instability and can
result in loss of mobility reduction. Other factors for polymer instability are chemical,
mechanical, thermal and microbial degradation. Mechanical degradation is a serious problem
for polyacrylamides. The long chain molecules that make up the polymer are not resistant to
shear stresses and can be easily broken up, especially at high velocities2,3,4. Mechanical
degradation for polyacrylamides becomes more severe in the presence of high salinity, high
fluxes in low permeability rock and adverse temperatures. Both chemical and thermal
degradation are aggravated under aerobic conditions. Recent studies have shown that
biocides, such as formaldehyde need to be used to prevent microbial degradation.
100
60
40
Apparent viscosity
Viscosity, cp
(in porous medium)

20
10
6
Measured viscosity
(Ostwald)
2
1
1
2
3
4
Polymer concentration %
Figure E.1.2: Resistance Effect of Polymer Solution in Porous Media
User's Guide IMEX
Appendix E 1075
16
14
12
10
8
6
4
2
0
0
20
40
60
80
100
120
Vinj / Vpor
Figure E.1.3
Polysaccharide biopolymers are the second type of polymer commonly used. The are
obtained from sugar in a fermentation process caused by the bacterium xanthomones
campestris. The polysaccharide's molecular structure yields a chain of great stiffness which
behaves like rigid-rod molecules. Consequently, in contrast to polyacrylamide polymer, the
viscosity of a polysaccharide biopolymer solution is not affected by salinity, and shearing
effects can be tolerated. Despite these advantages, a polysaccharide biopolymer's stability
degrades at temperatures above 95C (like polyacrylamide polymer) and is more susceptible
to bio-degradation. It is also much more expensive than polyacrylamide polymer.
Mass Conservation Equations

Modelling the polymer solution requires correctly describing the increase in viscosity and
mobility control as well as the polymer adsorption/desorption5. These properties are
functions of both reservoir permeability, which varies spatially, and polymer concentration,
which varies both spatially and with time. Therefore, a mass conservation is required which
describes the flow of polymer solution through the porous media of the reservoir. The mass
conservation equation combined with Darcy's law yields:
pSw D e
C p Tw k (Pw w h ) +
Bw
p C pSw
+ A d
+ q w C p =
t B w
1076 Appendix E
C p
(E.1.12)
User's Guide IMEX
where
=
=
=
=
=
=
=
=
=
=
=
Tw
Cp
Pw
De
p
Ad
k
w
qw
Sw
Bw
Transmissibility of the polymer solution

Pressure of the water phase
Effective dispersion coefficient
Polymer accessible porosity
Polymer adsorption/desorption
Absolute permeability
Water phase specific gravity
Water production/injection rate
Water saturation
This equation can be written in difference form as
pSw D e
C np +1 + q w C p
B
w
n
+
1
n
pSw C p

Vb p S w C p
+ A d
+ A d = 0
t B w
Bw
Twx C px
Pwn +1
xw D
(E.1.13)
Note that the water phase transmissibility is modified in the manner described in equations
(E.1.4, E.1.5, E.1.6) to incorporate the effects of increase of mobility due to the presence of
the polymer.
Guidelines for Polymer Application

Polymer flooding is of practical use only in reservoirs of a certain characteristic. Due to
polymer degradation, application of polymers should be limited to reservoirs with in-situ
temperatures less than 95C. The porosity of the reservoir rock should be medium to high in
order to assure a good storage capacity. Also, the salinity of the connate water should be less
than 10,000 ppm total dissolved solids in order to assure a stable polymer solution for
polyacrylamide polymer. Reservoirs with moderate to good permeabilities are generally
good candidates. Permeabilities that are too low will result in very high injection pressures
and permeabilities that are high usually give good recoveries with conventional waterfloods.
Reservoirs with large variations in permeability are also ideally suited for polymer flooding
since the polymer solution has a tendency to divert toward unswept regions of the reservoir.
Polymer flooding is applied most effectively in the early stages of a waterflood while the
mobile oil saturation is still high. Waterflooded oil reservoirs can also be good candidates
provided that the high values of WOR are caused by either a high water-oil mobility ratio
(viscous oil reservoirs) and water coning or by a low vertical sweep efficiency (heterogeneous
reservoirs). Under these conditions, the mobile oil saturation will be high and the diverting
effect of polymer flooding will be effective in reducing the mobile oil saturation. Polymer
flooding, in general, should be avoided for reservoirs with a large gas-cap and/or extensive
aquifers. The oil viscosity of candidate reservoirs should not be higher than 200 cp.
User's Guide IMEX
Appendix E 1077
References for Polymer Option

1. Dawson, R. and Lautz, R, "Inaccessible Pore Volume in Polymer Flooding," SPE
Journal, October 1972.
2. Maerker, J.M., "Mechanical Degradation of Partially Hydrolyzed Polyacrylamide
Solutions," SPE Journal, August 1976.
3. Chang, H.L., "Polymer Flooding Technology - Yesterday, Today and Tomorrow,"
JPT, August 1978.
4. Needham, R.B. and Doe, P.H., "Polymer Flooding Review," JPT, December, 1987.
5. Vela, S., Peaceman, D.W. and Sandvik, E.I., "Evaluation of Polymer Flooding
With Crossflow, Retention and Degradation," SPE Journal, April 1976.
1078 Appendix E
User's Guide IMEX
Pseudo-Miscible Option
The objective of miscible displacement is to reduce the residual oil saturation through the
complete elimination of the interfacial tension (IFT) between oil and the displacing fluid
(solvent). This is achieved if oil and the displacing fluid are miscible, i.e. they mix together
in all proportions to form one single-phase. Miscibility can be obtained on 'first contact' or
through 'multiple contact'1.
Miscible displacement is characterized by unstable frontal advances, in the form of either
viscous fingering or gravity tonguing. These instabilities are caused by the highly adverse
viscosity ratio and large density difference that generally exist between oil and the displacing
solvent. Figure E.2.1 depicts the swept zone at solvent breakthrough at an adverse mobility
ratio of 15, for a five-spot pattern flood as observed in the Hele-Shaw (parallel plate) model.
The grid structure in Figure E.2.1 contains 121 grid blocks. In order to reproduce in detail
the viscous fingering shown in the figure, the grid block sizes should be in the order of that of
a finger width or smaller. Thus, about 10 to 50 times this number of blocks might be required
for accurate numerical simulation of this single five-spot displacement.
Todd and Longstaff2 proposed a method of simulating miscible displacement performance
without reproducing the fine structure of the flow. Their method involves modifying the
physical properties and the flowing characteristic of the miscible fluids in a three-phase
black-oil simulator. They introduced a mixing parameter , which determines the amount of
mixing between the miscible fluids within a grid block. A value of zero corresponds to the
case of a negligible dispersion rate, whereas a value of one corresponds to complete mixing.
The following sections describe the equations used to form a 'four-component' model in
CMG's IMEX simulator. The equations are written in a general form so that immiscible
displacement can also be modeled if there is a loss of miscibility.
Mass Conservation Equations

It is assumed that solvent dissolves in water but not in oil. Hence, the mass conservation
equation for solvent can be expressed as follows:
R sol S w
+ E s S s + R sol q w + q s

t B w
R sol w
k ( p w w g ) + E s s k (p s s g ) = 0
Bw
User's Guide IMEX
(E.2.1)
Appendix E 1079
Figure E.2.1: Swept Zone at Solvent Breakthrough (from Todd and Longstaff, 1972)
The chase gas, which is usually injected following a solvent slug, may dissolve in oil but not
in water. Thus, the original gas equation in the black-oil model can be used to describe the
flow of the chase gas.
Care must be exercised in the PVT data preparation if chase gas is present. Rs is now the
solubility of the chase gas in the reservoir at various pressures. Eg and g represent the
expansion factor and viscosity of the chase gas. Furthermore, the bubble point pressure for
the oil/chase gas system in the reservoir should be set at a low pressure where Rs is zero. For
accurate modelling, the block pressures should not be allowed to drop below the MMP.
If there is no change gas injection, the original gas equation is used to describe the flow of the
free gas. In this case, the block pressures can drop below the bubble point pressure and the
simulation can still be properly modeled.
Equations for both immiscible and miscible displacements are discussed in the next subsection.
Pseudo-Miscible Considerations
Consider a miscible slug displacement process in which chase gas is injected following a
solvent slug. It is assumed the solvent is miscible with the oil and the chase gas. The chase
gas, however, is not miscible directly with the oil.
1080 Appendix E
User's Guide IMEX
Figure E.2.2 depicts a typical fluid distribution within a grid block. The mixing of solvent
and oil is controlled by a pressure-dependent mixing parameter, o(P), as shown in Figure
E.2.3. When the block pressure is so much lower than the minimum miscibility pressure
(MMP) that o(P) = 0.0, solvent is displacing oil immiscibility. As the block pressure
increases, this mixing parameter reaches its maximum value omax at the MMP. The
maximum value omax, however, cannot be estimated adequately. There is only a limited
amount of published material to aid in this estimation. Several authors3,4 have attempted to
estimate a value for the mixing parameter by history-matching field pilot tests. This history
matches mixing parameter is then used judicially for the predictions of full-scale
performance. When no better data is available, the limited work to date suggests a value in
the range of 0.5 to 0.7 as a first approximation.
The mixing of solvent and chase gas (or free gas) is governed by g, which is assumed
pressure independent. g is bounded by zero and one. Since solvent/gas has a lower
mobility ratio than oil/solvent, g is usually greater than omax. Notice that in Figure E.2.2 it
is possible for the oil/solvent and solvent/gas dispersion zones to merge, thereby miscibility is
lost. This situation is assumed to have occurred when the local solvent saturation drops
below a certain value (say, Ssmin = 0.01). The model assumes that the remaining solvent is
immobile and the oil and gas are treated as immiscible phases.
The presence of water may block the solvent from the oil, preventing the complete displacement
of oil by solvent even under miscible conditions5. It has been observed that the water-blocking
effect is represented by a function relating water saturations and irreducible oil saturations.
Flow Direction
Solvent
Oil
Immiscible
Displacement
of Oil by Gas
Solvent
Oil
Gas
Solvent
Gas/Solvent
Dispersed Zone
Solvent/Oil
Dispersed Zone
Figure E.2.2: Schematic Representation of a Three-Component Miscible

Displacement in a Grid Block (from Todd and Longstaff, 1972)
User's Guide IMEX
Appendix E 1081
omax
o(P)
0
MMP
Figure E.2.3: WI Versus P
Relative Permeabilities Modifications
The relative permeabilities of the three hydrocarbon phases are divided into two parts, one for
miscible flow and a second for immiscible flow. For each phase the effective relative
permeability is calculated as
k eff
rp =
o (P) m
(P) Im
k rp
k rp + 1 o
o max
o max
(E.2.2)
The miscible portion of the effective relative permeability k m

rp is calculated as:
m
km
rp = Fp k row (S w )
p = oil, gas, solvent
(E.2.3)
where Fpm is a miscible partitioning function which acts to partition krow(Sw) according to
the saturation of each hydrocarbon phase weighted by each phase's mixing parameter p.
The immiscible portion of the effective relative permeability k Im
rp is calculated as
k Im
ro = k ro (S w , S L )
p = oil
where SL = So + Sw and kro is the immiscible three phase relative permeability.
Im
k Im
rp = Fp k rg S g + S sol
p = gas, solvent
where FpIm is an immiscible partitioning function which acts to partition krg (Sg + SsoL)
according to the saturations of gas and solvent.
1082 Appendix E
User's Guide IMEX
MISCIBLE PARTITIONING FUNCTION
The miscible partitioning function Fpm is calculated as follows:

Define Frac
=
p
p
p max
o is a function of pressure and pmax is defined as the maximum value of p.
(The values of can vary from pmax either due to pressure or due to the loss of miscibility
as Ssol approaches the minimum solvent saturation.)
We obtain
Fom =
oFrac S*o
(E.2.4)
Frac
oFrac S*o + gFrac S g + sol
Ssol
S*o = S o S orm (S w )
Fgm =
gFrac S g
(E.2.5)
Frac
oFrac S*o + gFrac S g + sol
Ssol
m
Fsol
= 1 Fom Fgm
Frac
sol
is not entered as input in the model but is calculated as a function of the gas and oil
parameters:
Frac
gFrac S g
Frac S*o
Frac
o +
sol
= Frac *o
oFrac S*o + gFrac Sg
o So + gFrac S g
Frac
g
(E.2.6)
IMMISCIBLE PARTITIONING FUNCTION
The immiscible partitioning function FpIm p = gas, solvent is calculated as:
Sg
FgIm =
Sg + Ssol
Im
Fsol
= 1.0 FgIm
CAPILLARY PRESSURE
The capillary pressure between oil and gas is determined as:

eff
Pcog
=
o
o
Pcog S g + 1
o max
o max
User's Guide IMEX
( )
Pcog S g + S sol
(E.2.7)
Appendix E 1083
Density Modification
The mixture densities are functions of pure component densities and represented by the
following expressions.
m
Im
eff
p = p p + 1 p p
S*o
om =
S* + S
sol
o
+ S soL
o
S* + S
sol
sol
(E.2.8)
Sg
gm =
S g + S sol
g + S soL
S g + S sol
sol
(E.2.9)
g Sg
o S*o
m
+
sol
=
o
S* + S + S
S* + S + S g
g s
soL sol
g g
sol sol
o o
o o
(E.2.10)
sol Ssol
+
S* + S + S sol
g g
sol sol
o o
g
o S*o
+
sol =
o
o S*o + g S g
o S*o + g S g
(E.2.11)
Im
are the pure component densities S*o = S o S orm (S w ) .
p
Viscosity Modification
The mixture viscosities are determined by a power fluid mixing rule and are represented by
the following expressions.
( )
m
eff
p = p
om =
gm =
m
sol
=
( )(
x Im
p
1 p
when p = oil, gas, solvent
o sol
S*o
S* + S
sol
o
1/ 4 SsoL
+
S* + S
sol
sol
o
1/ 4

o
1 / 4
g
g soL
Sg

S + Ssol
g
1 / 4 + Ssol
sol
S g + Ssol
(E.2.12)
(E.2.13)
g sol o
o S*o 1 / 4 1 / 4 g Sg 1 / 4 1 / 4 sol Ssol 1 / 4 1 / 4

o sol +
o g
+
S
Sn sol g
n
Sn
1084 Appendix E
(E.2.14)
User's Guide IMEX
where
g Sg
o S*o
+
sol =
o
g
*
o S*o + g S s
S
S
g g
o o
(E.2.15)
and
S n = o S*o + g S g + sol S sol , S*o = S o S orm (S w ) ,
are the pure component viscosities.

Im
p
Estimation of the Mixing Parameter for Oil and Gas Above the Minimum Miscibility
Pressure
Oil:
For the oil phase mixing parameter it is recommended that the Koval formula be used. This
would represent a realistic maximum value of o.
o = 1 4 Log 0.78 + 0.22 M 1 / 4 / Log(M )

M = oil / solvent at reservoir conditions
(E.2.16)
Gas:
We normally recommend a gas mixing parameter equal to or greater than o. Equation
(E2.16) can be used with M = solvent / gas to estimate g. This would result in gas mixing
parameters very close to 0.78. A value of 1.0 would imply that the solvent and gas are
completely miscible with no gas-solvent fingering.
LOSS OF MISCIBILITY
In the IMEX model we allow the mixture to lose miscibility when the gas saturation is above
an input threshold value (*SGTHRESH) and the solvent saturation is lower than an input
minimum solvent saturation (*MINSS).
As IMEX is an adaptive implicit simulator and can be subject to numerical problems when
abrupt changes occur, we apply a smoothed transition (*MINSS *SMOOTHEND) between
the miscible and the immiscible range of solvent and gas saturations.
Pressure Dependence of Miscibility
o is considered to be a function of pressure and is entered as such a function on the PVTS

keyword. g is assumed to be independent of pressure.
References for Pseudo-Miscible Option

1. Stalkup, F.I. Jr., "Miscible Displacement", First Printing, SPE Monograph, 1983.
2. Todd. M.R. and Longstaff, W.J., "The Development, Testing and Application of a
Numerical Simulator for Predicting Miscible Flood Performance", Trans. AIME
253 (1972) pp. 874.
3. Bilhantz, H.L. et al, "A Method for Projecting Full-Scale Performance of CO2 in
the Willard Unit", paper SPE 7051 presented at the 1978 SPE Symposium on
Improved Methods for Oil Recovery, Tulsa, April 16-19.
User's Guide IMEX
Appendix E 1085
4. Youngren, G.K. and Charlson, G.S., "History Match Analysis of the Little Creek
CO2 Pilot Test", J. Pet. Tech., November 1980, pp. 2042-52.
5. Raimondi, P. and Torcaso, M.A., "Distribution of the Oil Phase Obtained Upon
Imbibition of Water", SPEJ, March 1964, pp. 49-55.
1086 Appendix E
User's Guide IMEX
Appendix F
The Use of IMEX in Well Testing Applications

INTRODUCTION
This document describes some guidelines for the use of IMEX in simulating well tests. Such
tests typically call for a well to be operated at a sequence of different flowing conditions,
usually involving high rate withdrawals followed by shut in periods, under careful pressure
monitoring. This monitoring is done with (electronic or mechanical) down-hole gauges
placed in the wellbore of the well undergoing the test, and sometimes in the wellbores of
selected neighboring wells. The single well test is most often encountered (no neighbor
monitoring, or no neighbors), although multiwell interference testing is also done. The
resulting down-hole pressure history is analyzed to obtain an indication of the reservoir's
character. The analysis hopes to recover at least the reservoir's permeability-thickness
product as well as other information regarding well properties (such as damage and the
presence of near wellbore fracturing). Certain reservoir parameters, such as the Warren and
Root model's LAMBDA and OMEGA parameters for multiple porosity reservoirs, are also
sought. The reservoir analysis is carried out with the aid of specialized graphical
presentations. These graphical techniques have been well developed over time and are
derived from the consideration of simplified reservoir and wellbore models.
Well test modelling assumes that a single phase fluid flows according to Darcy's law, and
with known PVT properties, though a reservoir into a well. The reservoir is assumed
homogeneous (or is sometimes assumed to be made up of homogeneous layers) of known
porosity and thickness. The well may be damaged or stimulated (that is, may have positive or
negative skin) and may possess some non-trivial volume (which leads to wellbore storage
effects (after flow)). The reservoir may also manifest certain special features, such as a
massive hydraulic fracture or multiple, connected porosities. Certain types of reservoir
boundaries can be examined. The goal of well testing is to determine the numerical values of
the parameters that characterize these phenomena in the well test model.
A black-oil reservoir simulator like IMEX can also model the above types of phenomena, as
well as much more general situations. For instance, heterogeneous reservoir properties,
reservoir zonation, multiphase flow effects (perhaps in the wellbore), realistic well
perforation and operating strategies, and complete descriptions of reservoir geometry can be
addressed by a simulator. These latter effects can only be hinted at in a conventional well test
analysis. This document will describe how to prepare an IMEX data set so that IMEX can
replicate the phenomena observed during a conventional well test.
User's Guide IMEX
Appendix F 1087
Both well test software and IMEX start with the full equations describing the flow of fluids in
a reservoir, and both types of software take the PVT description of the fluid into account.
However, well test software then turns to semi-analytical techniques to solve simplified,
linearized forms of the basic equations. These techniques sometime use specialized
transformations (the pressure-squared gas formulation and the real gas pseudo-pressure, for
example) that are not required in IMEX. IMEX uses sophisticated numerical techniques to
solve the full reservoir flow equations without the need to resort to such methods. Because of
the extreme flexibility gained through the use of numerical techniques, IMEX is the model of
choice for doing forward predictions that involve the eventual appearance of multiple flowing
phases (and alternate reservoir drive mechanisms), realistic multiple well operations and
actual reservoir geometry. Well test software remains the method of choice for determining
initial estimates for parameter values from field measurements using a simplified model.
A history match of a well test should be carried out before using IMEX to do more advanced
field modelling. A well test is often the only historical information available for a new pool,
and plays an important role in understanding reservoir behavior. History matches provide the
necessary simulator calibration to give confidence in the prediction of pool depletion
strategies, particularly with respect to new well locations. With a history match in hand,
sensitivities to changes in parameters can also be estimated and confidence limits in the
analysis of the original test can be established. This allows estimation of risk factors in
further calculations. Note that matching must be carried out in an ad hoc manner as CMG
does not currently supply the software to present IMEX pressure output in the forms most
often used in well test analysis. The use of a simple well test package often facilitates the
examination of IMEX output in well test situations.
With a match in hand, the engineer can go on to estimate well deliverability, production
profiles, reserves, and, eventually, an optimal depletion strategy.
Well tests can be simulated before they are carried out to help design economic draw down
and build up durations, along with rates. Even though these latter simulations must be carried
out in IMEX using data from analogue pools or from other sources, invaluable insight is
obtained regarding the set up of the proposed test.
Some guidelines will now be given for preparing data sets for well test simulations using
IMEX. The results of simulations prepared using these guidelines have been calibrated with
well test software and have been found to generate valid tests. They also have been used to
match actual field tests. A sample data set, and a Log-Log plot of the output generated by
IMEX using that data set, follow.
RESERVOIR DATA SECTION
The Radial grid option provides the simplest approach to grid design if the shape of the
reservoir is generally unknown. This type of grid will account for the volume of the reservoir
without requiring any detailed specifications for the reservoir boundary. A Cartesian or
Variable Thickness grid can be used if seismic or other geological indications of pool geometry
are available. However, the simplicity of the Radial grid option should not be overlooked, and
its use should be carefully considered before more involved models are created. As well test
modelling often assumes a simple homogeneous reservoir of uniform thickness, a single layer
IMEX model usually suffices (NK = 1). Reservoir thickness can be derived from log analysis
and entered into IMEX. Because of the overall simplicity of the model, NI (the number of
1088 Appendix F
User's Guide IMEX
blocks in the radial direction) can be fairly large. This permits high accuracy, yet running times
will not be excessive. Values for NI on the order of 25-50 should suffice.
Radial block sizes should begin as fractions of a metre, where the first block should be
somewhat larger than twice the actual wellbore radius. Block sizes should increase so that
the radial locations of the block centres are roughly in geometric progression. The sizes of
the outer blocks should be adjusted so that the total reservoir size agrees with known
geological estimates, or the information obtained about the reservoir's boundary location from
long time behavior analysis of the well test. The IMEX adaptive implicit formulation ensures
that the small, near wellbore grid blocks can be easily tolerated by the simulator. Note that
the wellbore radius specified for the grid (RW) should be equal to the actual wellbore radius.
Sudden large increases in the radial block dimensions can have the effect of imposing an
artificial, internal boundary in the simulation, causing boundary like slope changes in
pressure profiles. If there is concern about grid block sizing, simulations can be always be rerun with smaller (halved) blocks, or with a more gradual increase in block size, and the
results compared to those generated by the coarser grid. If important discrepancies appear,
there should be cause for concern and grid block sizes should be reduced for subsequent runs.
Porosity and rock compressibility information should be set based on the best information
available, duplicating what was used in the well test analysis. Reservoir permeability should
be set based on the results of the well test analysis.
If the reservoir is naturally fractured, the double porosity reservoir modelling feature of
IMEX should be used. Porosities are required for both the Fracture and Matrix systems in
these situations. (IMEX regards these values as describing the fraction of bulk reservoir rock
that is attributed to the appropriate system.) Matrix porosity will likely be determined from
logs and will be the same value as was entered into the well test analysis. Fracture porosity
will be backed out of the OMEGA parameter determined during the well test analysis and
then entered into IMEX. OMEGA represents the ratio of storativities between the Fracture
and Matrix systems and if equal compressibilities are assumed (an assumption likely
predicated by lack of detailed information about the fractures), OMEGA is determined to a
large extent by the simple ratio of porosities.
Like porosities, permeabilities also have to be entered for the both the Matrix and Fracture
systems in IMEX. These values represent the permeabilities of their respective systems, as
determined across a cross section of bulk reservoir rock. Fracture permeability is essentially
the permeability determined from well test analysis (for instance, from Horner analysis), and
it should be entered directly into IMEX. Matrix permeability should be backed out from the
LAMBDA parameter determined from well test analysis. Note that a choice has to be made
for matrix block geometry and fracture spacing before this calculation can be carried out.
Basically, any reasonable choice will do, as it is only the composite LAMDA that determines
flow properties in both the well test model and IMEX. Nevertheless, these choices affect the
derived value of matrix permeability, and if it is desirable to correlate the derived value with
other indicators (such as core determined values), a reasonable set of parameter values should
be used. The choices made should be the same for well test analysis and for IMEX.
If a multiple layer model is being examined, perhaps as a more geologically palatable way to
explain multiple porosity behavior in a well test, it is easy to set up IMEX accordingly.
User's Guide IMEX
Appendix F 1089
PVT AND INITIAL DATA SECTION

PVT properties should be entered in the usual way for the flowing phase. Simple, and
reasonable, estimates should be made for the properties of the remaining two non-flowing
phases present in IMEX. For instance, if gas flow is being modelled, constant oil formation
volume factors and viscosities (= 1), and increasing, but small, solution gas-oil ratio values
should be specified.
Relative permeabilities should be entered as simple straight lines, with unity endpoint values.
(Thus, the oil-water table needs only two data lines.) The connate water saturation should be
set to what exists in the field. Some type of residual oil saturation will need to be input,
which means the liquid-oil table can have as few as three data lines.
The reservoir's initial conditions should be set with a (non zero) connate water saturation.
The remaining pore space should be made up of the flowing phase of interest. If the reservoir
is a gas reservoir, a tiny amount of saturated oil should be initialized. The initial reservoir
pressure should be set to the same value as entered into the well test analysis.
WELL DATA
The well should be defined in the first block for a Radial grid, or suitably located in a more
general Cartesian grid. The well radius should be set to be the actual radius. A skin value
can be set as determined from the well test. Note that if a large negative skin is required, a
more detailed model will have to be constructed incorporating an actual fracture in a
Cartesian grid. Note also that skin can be modelled as a region of lowered permeability
around the wellbore, if desired. This would involve the resetting of the permeability of a few
of the blocks nearest the well. The extent of the region and the factor by which
permeabilities should be lowered would have to be determined by matching.
The well should begin operation on the appropriate day and time at the required (high) rate.
When the time comes for the well to be shut in, a tiny (but nonzero) rate should be set. (This
rate could be a hundredth of a standard cubic metre per day for fluids, or a standard cubic
metre per day for gas.) This maintains the well as an operating well in IMEX, and permits
continued printing of its Bottom Hole Pressure (BHP) in the IMEX output. These latter
values are interpretable as the required flowing BHP when the well is operating at the high
rate, and as the shut in pressures when the well is on trickle. This approach allows consistent
use of suitable software to obtain the BHPs throughout the simulation, regardless of the state
of the well, and avoids having to activate block pressure output.
Note that IMEX does not allow for the specification of non-Darcy flow effects near the
wellbore for gas wells (turbulent flow) other than by explicitly adjusting permeabilities and/or
transmissibility multipliers near the wellbore, or by adjusting the skin. Rate dependent skins
can be modelled by calculating the skin for the new rate and re-setting the skin at the changes.
WELL DATA, WELLBORE STORAGE
No provision has been made so far for wellbore storage in the well model. Such storage effects
can be simulated in IMEX by the use of a block volume modifier (VOLMOD) to increase the
volume of the well block. (Specify two VOLMODs using both *FRACTURE and *MATRIX
if a Double Porosity model is being used. It is the Fracture volume that matters in these
situations.) The first choice for the multiplier should be the wellbore (tubular) volume (the
entire volume, from reservoir to surface) divided by the (Fracture) pore volume of the wellbore
block. This value for the volume modifier can be checked by ensuring that the value for the
1090 Appendix F
User's Guide IMEX
(Fracture) block volume that is reported for the well block in the IMEX output file corresponds
to the wellbore volume. The value can than be adjusted to match the observed behavior of the
well test at early times.
The procedure just described is based on the idea that wellbore storage arises from having a
wellbore filled with compressible fluid, where the storage coefficient can be approximated as
the wellbore volume multiplied by the fluid compressibility at the well block pressure. The
volume provided by the VOLMOD-adjusted block volume accounts for the wellbore volume
in this application and IMEX automatically handles the compressible nature of the fluid it
contains. Note that multipliers on the order of hundreds or more may be required.
NUMERICAL PARAMETER CONTROL
As regards the numerical parameters, the simple nature of well test modelling allows
specification of small time steps without great run time penalties. The initial time step can be of
the order of a minute, or even less. A time step of the order of a few seconds is not reasonable
however, due to the lack of accurate wellbore modelling at length scale of a fraction of a metre.
Time steps will then be primarily controlled by pressure NORMs. Values of up to about half an
atmosphere are reasonable here. Finally, as the simulation progresses, the maximum time step
will be limited by time step size. A maximum time step limit of a day (or a few days) should
suffice. These type of parameters will give up to thousands of quickly converging timesteps,
mimicking the quantity of data obtained from an electronic gauge.
Once a representative run is obtained with the time step levels defined by the above
conservative settings, it is easy to try a few sensitivity runs with larger limits. Such
experiments could very possibly obtain accurate results in considerably less time, and thereby
lead to suitable alterations in run time parameters for later trials.
SUMMARY
These guidelines should generate simulation results that will almost exactly overlay the well test
results. Some tuning might be required if the well test analysis was not particularly accurate, as
some of the data for IMEX requires well test derived results. Once a tuned IMEX data set is
available, modifications can be made to simulated the more realistic operating conditions
encountered in a developing pool, thereby starting the search for an optimal depletion strategy.
IMEX Buildup Well Test Results
1
0.8
0.6
0.4
Log (pD)
0.2
0
-0.2
-0.4
-0.6
-0.8
-1
0
Log (tD)
User's Guide IMEX
Appendix F 1091
*****************************************************************
* IMEX Example Data Set for Well Test Analysis.
* This is a dual porosity reservoir. With 40 radial blocks.
* The reservoir is a gas reservoir.
****************************************************************
**************************
* I/O Section
**************************
*TITLE1
'WELL TEST EXAMPLE DATA SET'
*TITLE2
'Gas Reservoir. Pressure Build-up Test.'
*INCLUDE 'caseid.dat'
*INUNIT *SI
*OUTUNIT *SI
*WPRN
*WPRN
*WPRN
*WELL 0
*GRID 0
*ITER *BRIEF
*OUTPRN
*OUTPRN
*WSRF
*WSRF
*WELL
*GRID
*WELL
*GRID
*BRIEF
*PRES
*DOUBLE
*NEVER
*OUTSRF *WELL *ALL

**************************
* Reservoir Description
**************************
*GRID *RADIAL 40 1 1 *RW
0.10 ** Note that the well radius must be

** same as that defined in the
** well data section ( after *RUN).
*DI IVAR
0.2
0.1
0.1
0.1
0.1
0.2
0.2
0.5
1
1
2
2
5
5
7.5
7.5
10
10
15
15
20
20
25
25
30
40
50
70
90
100
100
100
100
100
100
100
100
0.75
100
100
*DJ *CON 360

*DK *CON 100
*DEPTH 1 1 1 1000
*DUALPOR
*SHAPE *GK
*DIFRAC *CON 1
*DJFRAC *CON 1
*DKFRAC *CON 1
*POR
*POR
*PRPOR
*PRPOR
*MATRIX
*FRACTURE
*MATRIX
*FRACTURE
*CON 0.050
*CON 0.005
25000
25000
*VOLMOD *MATRIX *IJK

1 1 1 150
** Model the wellbore storage by
*VOLMOD *FRACTURE *IJK
1 1 1 150 ** block volume multipliers.
*CPOR
*CPOR
*MATRIX
*FRACTURE
1092 Appendix F
0.1E-6
0.1E-6
User's Guide IMEX
*PERMI
*PERMI
*PERMJ
*PERMJ
*PERMK
*PERMK
*MATRIX
*CON 0.00025
*FRACTURE *CON 2.5
*MATRIX
*CON 0.00025
*FRACTURE *CON 2.5
*MATRIX
*CON 0.00025
*FRACTURE *CON 2.5
*****************************
* Fluid properties.
*****************************
*MODEL *BLACKOIL
*PVT 1
** P
RS BO EG VISO VISG
11200
1.01 1 112 1
0.015
13000
1.03 1 128 1
0.016
13800
1.05 1 135 1
0.016
14700
1.07 1 143 1
0.017
15600
1.09 1 152 1
0.017
16400
1.11 1 159 1
0.017
17300
1.13 1 167 1
0.018
18100
1.15 1 174 1
0.018
19000
1.17 1 182 1
0.019
19900
1.19 1 190 1
0.019
20700
1.21 1 197 1
0.020
21600
1.23 1 206 1
0.020
22500
1.25 1 214 1
0.020
23300
1.27 1 221 1
0.021
24200
1.29 1 229 1
0.021
25100
1.31 1 237 1
0.022
*DENSITY
*DENSITY
*DENSITY
*OIL
*GAS
*WATER
800
1
1100
*CO 0.1E-6
*CVO 0
*BWI
*CW
*REFPW
*VWI
*CVW
1.0
0.1E-6
25000
0.5
0
*******************************
* Relative permeability data.
*******************************
*ROCKFLUID
*RPT 1 *STONE2
*SWT
** SW KRW KRO PCOW
0.1 0
1
0
1
1
0
0
*SLT *QUAD *ON
** SL KRG KROG
0.1 1
0
0.3 0.7
0
1
0
1
User's Guide IMEX
PCOG
0
0
0
Appendix F 1093
*******************************
* Initialize reservoir.
*******************************
*INITIAL
*VERTICAL *OFF
*PRES
*MATRIX
*CON 25000
*PRES
*FRACTURE
*CON 25000
*PB
*PB
*MATRIX
*FRACTURE
*CON 25000
*CON 25000
*SO
*SO
*MATRIX
*FRACTURE
*CON 0.00001
*CON 0.00001
*SW
*SW
*MATRIX
*FRACTURE
*CON 0.1
*CON 0.1
**********************************
* Numerical controls.
**********************************
*NUMERICAL
*MAXSTEPS 9999
*NORM *PRESS 25
*DTMIN 0.00001
**********************************
* Well data section.
**********************************
*RUN
*DATE 1992
6
0.01
*DTWELL 0.001
*DTMAX 1.00
*WELL 1 'WELL'
*PRODUCER 1
*OPERATE *GAS 1000000
** Produce at a large rate.
*OPERATE *MIN *BHP 100 *STOP
*GEOMETRY *K 0.1 0.5 1.0 0
*PERF *GEO 1
1
1
1
1.0
*DATE 1992
6
*DTWELL 0.00005
*ALTER
1
1
*DATE
3.00
** Shutin the well. (Produce at a very-very small rate.)
1992
0.01
*STOP
1094 Appendix F
User's Guide IMEX
Appendix G
Analytical Aquifers
The role of aquifers in studying reservoir performance is usually quite large. The aquifers
supply additional energy to a connected reservoir in the form of water influx. In reservoir
simulation studies, aquifers can be represented as 1) Numerical Aquifers (additional grid
blocks added) or 2) Analytical Aquifers.
Aquifers represented by Numerical Methods have the disadvantage of increasing the number
of blocks, which increases both the required computer processing time (CPU) and storage.
The Analytical Methods can often provide the same results as the Numerical Aquifer, while
using less computer resources. Analytical Aquifers will fail to properly model reservoir fluids
flowing back to the aquifer. In this case the use of Numerical Aquifers is preferred.
There are two well-known analytical methods, which are widely used in the simulation
industry. The first method is known as the Carter-Tracy method, which is a modified version
of the original Van-Everdingen and Hurst method. The second and more recent, is the
Fetkovich method. The first method is a rigorous mathematical solution based on the solution
of the radial flow equation, while the second is a simple material balance equation. In most
cases, both methods predict the water influx volumes within acceptable engineering accuracy.
User's Guide IMEX
Appendix G 1095
Mathematical Background
Van-Everdingen and Hurst Method
This method is based upon solving the radial equation using the Laplace transform as applied
for different boundary conditions. The radial flow equation can be written as:
2 P(t d )
r
1 P(t d ) P(t d )
=
(1)
r r
t d
Where
P(td) = Pressure at distance r and dimensionless time td
'td' = Dimensionless time, defined as given in Equation (2)
td =
k
fC e R o2
(time)
( 2)
Time = Real simulation time

= Viscosity of fluids
f = Fraction of aquifer/reservoir connection

Ce = Effective fluid and rock compressibility
k = Absolute permeability
Ro = Effective reservoir/aquifer radius
In the reservoir simulation study the Constant Rate solution is used for finite or infinite
aquifer extent.
Infinite Aquifer with Constant Rate Solution

The dimensionless pressure drop P(td) at any dimensionless time (td) for the infinite aquifer
and under a constant rate solution, is expressed by Equation (6) below:
P(t d ) =
4
2
1 e u 2 t d du
J12
(u ) +
Y12
(u )]
(3)
Where J and Y are Bessel functions and u is an operator. The numerical solution of the above
integral provides the general solution of an infinite aquifer with constant Rate. The solution is
tabulated in Table (1) for (td) values between td = 0.01 to td = 1000. The value of P(td) at for
(td) below 0.01 is computed from Equation (4), while for (td) greater than 1000.0, P(td) is
calculated from Equation (5).
P(t d ) =
2 1/ 2
t d , t d < 0.01 (4)
P(t d ) =
1
[log t d + 0.80907] , t d > 1000 (5)
2
1096 Appendix G
User's Guide IMEX
Table(1) is saved in the CMG aquifer model, and the extrapolation beyond td of of 1000,
equation (5) is used.
Limited Aquifers with Constant Rate Solution

When Equation (3) above is applied to the limited aquifers with constant rate, a simpler
solution is derived as given by Equation (6) below,
P(t d ) =
4
4
2
1
(3R d 4R d log R d 2R d 1)
+
t
(R d2 1) 4
4(R d2 1) 2
td 2
J 1 ( n R d )
2 2
2
1 , 2 n [ J 1 ( n R d ) J 1 ( n )]
+2
( 6)
Where the B1, B2 are the roots of the following equation
[J1 ( n R d )Yo ( n R d ) Y1 ( n R d )J o ( n R d )] = 0
(7)
The above equation is solved and tabulated for several values of Rd, where Rd is defined as
Re/Ro, where (Re = external aquifer radius, Ro = external reservoir radius). Table (III) of the
original work in Reference 1 gives values of P(td) versus td for Rd values of 1.5, 2.0, 2.5, 3.0,
3.5, 4.0, 4.5, 5.0, 6.0, 7.0, 8.0, 9.0, and 10.0. The same table is attached in this report as Table
(2) for reference. Figure (1) displays these results.
The last term in the above equation becomes insignificant at larger values of td. Thus, beyond
the tabulated td values and for any Rd, Equation (8) is used for extrapolating td.
P (t d ) =
4
4
2
1
3R d 4R d log R d 2R d 1
+
t
(8)
d
R d2 1 4
4 R d2 1
In CMG models, equation (8) is not used directly, the last two entry points in the user input
table are extrapolated for larger dimension time. The user has to be sure Equation (8) has
approximate linearity for his last two values of td. The user should enter the entire table from
the supplied tables (Table 2) to ensure this is the case.
Carter-Tracy Method
This is a modified Van-Everdingen and Hurst method, which doesn't require the
superposition theorem. It changes the assumption from a field producing at a constant
hydrocarbon rate to an aquifer that influx water at a constant rate. This assumption
simplifies the solution to that shown in Equation (9), which is implemented in the simulator.
UU * P( t dj ) We ( t dj1 ) * P ( t dj )
We ( t dj ) = We ( t dj1 ) +
( t dj t dj1 ) (9)
P( t dj ) t dj1P ( t dj )

UU = Aquifer constant = 1.119 * * C e * R o2 * h *
(10)
360
P( t dj ) = Change in boundary pressure = Pres t j Pres t j1
( )
( )
P( t dj ) = Dimensionless pressure drop

Ce = Effective compressibility = Cw + Cf
User's Guide IMEX
Appendix G 1097
Cw = Water compressibility
Cf = Rock compressibility
P ( t dj ) =
P( t dj )
t dj
We(tdj) = Cumulative water in flux at time (tdj)

H = Aquifer thickness
Ro = Reservoir radius
= Contact angle between reservoir/aquifer
Fetkovich Method
The Fetkovich method is based on a simple material balance concept, and it is considered an
approximation method compared to the Van-Everdingen and Hurst method. This method
doesn't need tables, or dimensionless groups. It assumes the flow of aquifer water into a
hydrocarbon reservoir is modeled in precisely the same way as the flow of oil from a
reservoir into a well. Thus an inflow flow equation combined with an average reservoir
pressure change constitute the essential ingredients to arrive to the following equation:
W ei
W en =
Pan 1 Pn * 1 . 0 e ( J w Pi t n / W ei )
(11 )
Pi
Where
j= n 1
Wej
j=1
= Average aquifer pressure (12)
Pan 1 = Pi 1.0
Wei
Pn 1 + Pn
Pn =
= Interface pressure between aquifer/reservoir (13)
2
7.08 * 10 3 kh
= Aquifer productivity index (14)
Jw =
Re
3
ln(
)
4
Ro
[ ]
Wei = C e * f * * (R e2 R o2 ) * h * Pi = Initial amount of encroachable water (15)

Wen is the incremental water influx in the interval tn.
1098 Appendix G
User's Guide IMEX
Practical Aspects of Using Analytical Aquifers

There are several points of concern regarding how to represent an aquifer in a simulation
study. Those are the Aquifer TYPE, SOLUTION METHOD, AQUIFER
PARAMETERS and the LEAK SWITCH. The following shows some user guidelines for
each point.
a) Selecting Type of Aquifer

In IMEX, a user is able to select the type of aquifers to be connected to a specific reservoir.
Those types are BOUNDARY or REGION or BOTTOM
The TYPE should be selected according to the actual aquifer location, and on how the
aquifer and reservoir are in contact. This could be determined from the geological description
of the problem. If the geology did not provide enough information about the aquifer type,
matching the reservoir performance using different types of aquifer is possible. An aquifer
type and/or aquifer parameters could serve as a matching parameter when a field history
indicates more energy is needed to maintain the reservoir pressure, or additional water has to
be produced. More than one aquifer can be connected to a simulation model. In addition
different types can be specified for each aquifer.
b) Selecting Method of Solution

An aquifer can be represented numerically or by either of the analytical methods described
above. The three methods should provide similar results when using the same input
parameters, provided the reservoir does not efflux fluids to the aquifer. Selection of one
method over the others can be aided by using the following criteria:
1. Using Numerical Aquifers: This method is the most accurate method, since it
provides an exact geometrical and probable geological description of the problem.
This is helpful if the aquifer is allowed to contain either slight volumes of oil or
some dissolved gases. The drawback of this method stems from the additional use
of computer storage and CPU time. Thus, this is best used when the computer
memory storage and computing time are not of concern. For large aquifers extent
(Rd >10), huge number of large blocks sizes would be required to correctly present
the aquifer. This yields to significant decrease in the simulation speed.
2. Using Carter-Tracy Aquifers: This method is based on a rigorous numerical
solution and it provides an accurate solution, however, it requires a different table
entry to simulate a given aquifer extent (value of Rd). Thus changing the aquifer
extent, require a change of the entry table. There are limited Rd values (1.5, 2, 2.5,
3.0, 3.5,4.0,4.5, 5,5.5,6,7,8,9,10, and infinite). Other values need the solution of
equation (3) above.
3. Using Fetkovich Aquifer: This method is the simplest method of analytical aquifer
presentation, and still provides reasonable answers. It requires the definition of the
aquifer parameters, while it can handle any aquifer extent, except the infinite. For
infinite aquifer extent, a value of Rd > 100 should be adequate.
User's Guide IMEX
Appendix G 1099
Aquifer Parameters
Same aquifer parameters are used for both of Carter-Tracy and Fetkovich solutions.
For the current version (IMEX9800+) the following parameters are needed:
Aquifer Thickness (H_AQ).
Aquifer Porosity (POR_AQ).
Aquifer Permeability (PERM_AQ).
Effective Reservoir External Radius (Ro).
Reservoir/Aquifer contact angle (radians) (F_AQ).
Ratio between the external Aquifer radius to the Reservoir external radius (Rd).
The model is able to calculate default values for all of the above parameters for either CarterTracy or Fetkovich aquifers. However, it is highly recommended that a user enter all of the
above parameters, since these parameters would play an important role in the reservoir
performance. The model uses the reservoir data to calculate defaults of parameters not
explicitly defined, the models choice of defaults is consistent, but may not be intuitive. A list
of model defaults is provided below. Accordingly, care must be exercised when the user
allows the model defaults to simulate an aquifer.
If the CARTER-TRACY METHOD was selected, the user has to consider the following:
a) For an infinite aquifer: The Model has this function as the default. The internal
table terminates at td=1000. Beyond this value, a proper extrapolation is done
automatically. Thus, the user need not enter a table.
b) For a finite aquifer: The user has to enter the correct table for a pre-determined
value of Rd (Table 2).
c) The finite aquifer table the user inputs from Table (2), has to have the same infinite
table value up to its starting time, while for times larger than the table last entry,
the model linearly extrapolates using the last two entry points. Thus, the user has
to enter the table correctly and in full.
1100 Appendix G
User's Guide IMEX
Model Defaults
Assume the following:
H
=
H_AQ
=
POR_AQ
=
PERM_AQ
=
AREA_Contact
=
F_AQ
=
Ro
=
Re
=
Rd
=
Effective Reservoir Thickness- Computed from Reservoir Geometry

Aquifer Thickness- Computed Below
Aquifer Porosity
Aquifer Permeability
Connected Area Between Reservoir and Aquifer (obtained from model)
Contact Angle between Reservoir and Aquifer
Effective Reservoir radius
Effective Aquifer External Diameter
Re/Ro
Parameter
Boundary
POR_AQ
H_AQ
Volumetrically Average
Porosity for connected blocks Porosity for connected
blocks
Permeability for connected Permeability for connected
blocks
blocks
H
H
Ro
AREA_Contact/(H_AQ*2)
F_AQ
1.0
PERM_AQ
Region
(AREA _ Contact / )
1.0
Bottom
Porosity for connected
blocks
Permeability for connected
blocks
(AREA _ Contact )
(H _ AQ / H )2
2*Atan(H_AQ/H)/2
Example:
Nx=Ny=Number of grids in x,y directions=11
Nz= Number of grids in Z direction = 12
Porosity = Constant = 0.3
Kx=Ky=Kz = x,y,z absolute permeability = 100 md
Dx=Dy= Grid thickness for x, y directions = 100 ft
Dz = Grid thickness in Z direction = 20 ft
Model calculate defaults:
Parameter
Boundary
POR_AQ
0.3
PERM_AQ
AREA
100
1056000 ft2
H_AQ
Nz*Dz=240
Ro
1056000/(240*2)=700.28
F_AQ
1.0
Region
0.3
Bottom
0.3
100
100
AREA-R (depends 1210000 ft2
on defined region)
Nz*Dz=240
(1210000 ) = 1100 '
(AREA / )
1.0
(1100 * 240 / ) = 289.88 '

2*Atan(1100/240)/2=0.4316
In some applications, when using the Fetkovich method, one could use different input
parameters such as aquifer injectivity index, volume of water initially available for injection,
and the initial reservoir pressure, in this case a user has to manually converts this data to the
required parameters above. This could be furnished as follows:
User's Guide IMEX
Appendix G 1101
Assume
Wi
Jw
Type
Contact Area
=
=
=
=
Volume of water available for Injection. (STB)

Aquifer Injectivity Index (STB/D/PSI)
Boundary Aquifer
Known from simulation
1. Dermine aquifer thickness H_AQ (ft), and from the contact area determine the
effective reservoir radius (ft) as follows:
Ro =
AREA Contact
H _ AQ * 2
2. Use the Wi (STB) in the equation below to determine the external reservoir radius
(Re - ft) use contact angle as 1.0, and use any reasonable value for the aquifer
porosity.
Wi = *
R e2 R 2o
* H _ AQ * POR _ AQ
5.615
Wi (ft3) is reported in the simulation output, which can be used either for checking
or for computing Re.
3. Use the Aquifer Injectivity Index (STB/D/psi) to determine the effective aquifer
permeability (md), as follows:
Jw =
7.08 * 10 3 * * PERM _ AQ * H _ AQ
W * ln [R e / R o ]
The w is the water or (aquifer fluid viscosity) in cp.

For further description of the above subject, the reader could consult references (1) to (6)
below:
1. Van Everdingen, A.F., Hurst, W., The Application of the Laplace Transformation
to Flow Problems in Reservoirs, Petroleum Transactions, AIME, December 1949.
2. Carter, R.D., Tracy, G.W., An Improved Method for Calculating Water Influx,
Petroleum Transactions, AIME, vol. 219, 1960.
3. Van Everdingen, A.F., Timmerman, E.H., McMahon, J.J., Application of the
Material Balance Equation to Partial Water Drive Reservoir, Petroleum
Transactions, AIME, vol. 198, 1953.
4. Fetkovich, M.J., A Simplified Approach to Water Influx Calculations Finite
Aquifer Systems, J. Pet. Tech., July 1971, pp. 814-828.
5. Dake, L.P., Fundamentals of Reservoir Engineering, Elesvier Scientific
Publishing Company, 1978.
6. Craft, B.C., Hawkins, M.F., Applied Petroleum Reservoir Engineering, PrenticeHall Inc., 1959.
1102 Appendix G
User's Guide IMEX
Table 1 - Radial Flow, Constant Terminal Pressure

and Constant Terminal Rate Cases for Infinite Reservoirs
t
1.0 (10)-2
5.0 "
1.0 (10)-1
1.5 "
2.0 "
2.5 "
3.0 "
4.0 "
5.0 "
6.0 "
7.0 "
8.0 "
9.0 "
1.0 "
1.5 "
2.0 "
2.5 "
3.0 "
4.0 "
5.0 "
6.0 "
7.0 "
8.0 "
9.0 "
1.0 (10)1
1.5 "
2.0 "
2.5 "
3.0 "
4.0 "
5.0 "
6.0 "
7.0 "
8.0 "
9.0 "
1.0 (10)2
1.5 "
2.0 "
2.5 "
3.0 "
4.0 "
5.0 "
6.0 "
7.0 "
8.0 "
9.0 "
1.0 (10)3
User's Guide IMEX
P(t)
0.112
0.229
0.315
0.376
0.424
0.469
0.503
0.564
0.616
0.659
0.702
0.735
0.772
0.802
0.927
1.020
1.101
1.169
1.275
1.362
1.436
1.500
1.556
1.604
1.651
1.829
1.960
2.067
2.147
2.282
2.388
2.476
2.550
2.615
2.672
2.723
2.921
3.064
3.173
3.263
3.406
3.516
3.608
3.684
3.750
3.809
3.860
Appendix G 1103
Table 2 Constant Terminal Rate Case Radial Flow Limited Reservoirs

R = 3.0
1 = 1.6358
2 = 3.1787
R = 2.5
1 = 2.1564
2 = 4.2230
R = 2.0
1 = 3.1965
2 = 6.3118
R1 = 1.5
1 = 6.3225
2 = 11.924
t
P(t)
P(t)
P(t)
P(t)
6.0 (10)-2
0.251
2.2 (10)-1
0.443
4.0 (10)-1
0.565
5.2(10)-1
0.627
8.0 "
0.288
2.4 "
0.459
4.2 "
0.576
5.4 "
0.636
1.0 (10)
0.322
2.6 "
0.476
4.4 "
0.587
5.6 "
0.645
1.2 "
0.355
2.8 "
0.492
4.6 "
0.598
6.0 "
0.662
1.4 "
0.387
3.0 "
0.507
4.8 "
0.608
6.5 "
0.683
1.6 "
0.420
3.2 "
0.522
5.0 "
0.618
7.0 "
0.703
1.8 "
0.452
3.4 "
0.536
5.2 "
0.628
7.5 "
0.721
2.0 "
0.484
3.6 "
0.551
5.4 "
0.638
8.0 "
0.740
2.2 "
0.516
3.8 "
0.565
5.6 "
0.647
8.5 "
0.758
2.4 "
0.548
4.0 "
0.579
5.8 "
0.657
9.0 "
0.776
2.6 "
0.580
4.2 "
0.593
6.0 "
0.666
9.5 "
0.791
2.8 "
0.612
4.4 "
0.607
6.5 "
0.688
1.0
0.806
3.0 "
0.644
4.6 "
0.621
7.0 "
0.710
1.2
0.865
3.5 "
0.724
4.8 "
0.634
7.5 "
0.731
1.4
0.920
4.0 "
0.804
5.0 "
0.648
8.0 "
0.752
1.6
0.973
4.5 "
0.884
6.0 "
0.715
8.5 "
0.772
2.0
1.076
5.0 "
0.964
7.0 "
0.782
9.0 "
0.792
3.0
1.228
5.5 "
1.044
8.0 "
0.849
9.5 "
0.812
4.0
1.578
6.0 "
1.124
9.0 "
0.915
1.0
0.832
5.0
1.828
1.0
0.982
2.0
1.215
2.0
1.649
3.0
1.596
3.0
2.316
4.0
1.977
5.0
3.649
5.0
2.358
-1
1104 Appendix G
User's Guide IMEX

(Continued)
R = 3.5
1 = 1.3218
2 = 2.5526
R = 4.0
1 = 1.1120
2 = 2.1342
R = 4.5
1 = 0.9609
2 = 1.8356
P(t)
P(t)
P(t)
1.0
0.802
1.5
0.927
2.0
1.023
1.1
0.830
1.6
0.948
2.1
1.040
1.2
0.857
1.7
0.968
2.2
1.056
1.3
0.882
1.8
0.988
2.3
1.072
1.4
0.906
1.9
1.007
2.4
1.087
1.5
0.929
2.0
1.025
2.5
1.102
1.6
0.951
2.2
1.059
2.6
1.116
1.7
0.973
2.4
1.092
2.7
1.130
1.8
0.994
2.6
1.123
2.8
1.144
1.9
1.014
2.8
1.154
2.9
1.158
2.0
1.034
3.0
1.184
3.0
1.171
2.25
1.083
3.5
1.255
3.2
1.197
2.50
1.130
4.0
1.324
3.4
1.222
2.75
1.176
4.5
1.392
3.6
1.246
3.0
1.221
5.0
1.460
3.8
1.269
4.0
1.401
5.5
1.527
4.0
1.292
5.0
1.579
6.0
1.594
4.5
1.349
6.0
1.757
6.5
1.660
5.0
1.403
7.0
1.727
5.5
1.457
8.0
1.861
6.0
1.510
User's Guide IMEX
9.0
1.994
7.0
1.615
10.0
2.127
8.0
1.719
9.0
1.823
10.0
1.927
11.0
2.031
12.0
2.135
13.0
2.239
14.0
2.343
15.0
2.447
Appendix G 1105

(Continued)
R = 5.0
1 = 0.8472
2 = 1.6112
R = 6.0
1 = 0.6864
2 = 1.2963
R = 7.0
1 = 0.5782
2 = 1.0860
P(t)
P(t)
P(t)
3.0
1.167
4.0
1.275
6.0
1.436
3.1
1.180
4.5
1.322
6.5
1.470
3.2
1.192
5.0
1.364
7.0
1.501
3.3
1.204
5.5
1.404
7.5
1.531
3.4
1.215
6.0
1.441
8.0
1.559
3.5
1.227
6.5
1.477
8.5
1.586
3.6
1.238
7.0
1.511
9.0
1.613
3.7
1.249
7.5
1.544
9.5
1.638
3.8
1.259
8.0
1.576
10.0
1.663
3.9
1.270
8.5
1.607
11.0
1.711
4.0
1.281
9.0
1.638
12.0
1.757
4.2
1.301
9.5
1.668
13.0
1.801
4.4
1.321
10.0
1.698
14.0
1.845
4.6
1.340
11.0
1.767
15.0
1.888
4.8
1.360
12.0
1.815
16.0
1.931
5.0
1.378
13.0
1.873
17.0
1.974
5.5
1.424
14.0
1.931
18.0
2.016
6.0
1.469
15.0
1.988
19.0
2.058
6.5
1.513
16.0
2.045
20.0
2.100
7.0
1.556
17.0
2.103
22.0
2.181
7.5
1.598
18.0
2.160
24.0
2.267
8.0
1.641
19.0
2.217
26.0
2.351
9.0
1.725
20.0
2.274
28.0
2.434
10.0
1.808
25.0
2.560
30.0
2.517
11.0
1.892
30.0
2.846
12.0
1.975
13.0
2.059
14.0
2.142
15.0
2.225
1106 Appendix G
User's Guide IMEX

(Continued)
R = 8.0
1 = 0.4999
2 = 0.9352
R = 9.0
1 = 0.4406
2 = 0.8216
R = 10
1 = 0.3940
2 = 0.7333
P(t)
P(t)
P(t)
8.0
1.556
10.0
1.651
12.0
1.732
8.5
1.582
10.5
1.673
12.5
1.750
9.0
1.607
11.0
1.693
13.0
1.768
9.5
1.631
11.5
1.713
13.5
1.784
10.0
1.653
12.0
1.732
14.0
1.801
10.5
1.675
12.5
1.750
14.5
1.817
11.0
1.697
13.0
1.768
15.0
1.832
11.5
1.717
13.5
1.786
15.5
1.847
12.0
1.737
14.0
1.803
16.0
1.862
12.5
1.757
14.5
1.819
17.0
1.890
13.0
1.776
15.0
1.835
18.0
1.917
13.5
1.795
15.5
1.851
19.0
1.943
14.0
1.813
16.0
1.867
20.0
1.968
14.5
1.831
17.0
1.897
22.0
2.017
15.0
1.849
18.0
1.926
24.0
2.063
17.0
1.919
19.0
1.955
26.0
2.108
19.0
1.986
20.0
1.983
28.0
2.151
21.0
2.051
22.0
2.037
30.0
2.194
23.0
2.116
24.0
2.090
32.0
2.236
25.0
2.180
26.0
2.142
34.0
2.278
30.0
2.340
28.0
2.193
36.0
2.319
35.0
2.499
30.0
2.244
38.0
2.360
40.0
2.658
34.0
2.345
40.0
2.401
45.0
2.817
38.0
2.446
50.0
2.604
40.0
2.496
60.0
2.806
45.0
2.621
70.0
3.008
50.0
2.746
User's Guide IMEX
Appendix G 1107
1.80
PRESSURE DROP IN ATMOSPHERES - P(t)
1.90
2.00
R=6
2.10
R=8
2.20
R=00
R=10
2.30
12 16 20 24 28 32 36 40
TIME (t)
Figure G-1
Radial Flow, Constant Terminal Rate Case,
Pressure Drop Versus Time, P(t) Versus 1
1108 Appendix G
User's Guide IMEX
Appendix H
Running IMEX Coupled to the GAP Surface Network Model

IMEX (2005.10 and later) can be linked to Petroleum Experts GAP surface network model
through the RESOLVE framework. RESOLVE is a tool written by P.E. to allow connectivity
between arbitrary Petroleum Engineering packages; it is a high-level framework that allows
engineering data to be passed between applications.
When a case of a client application such as IMEX is loaded into RESOLVE, RESOLVE will
create a slave instance of that application.
RESOLVE will then query the case for its sources and sinks these are the inputs and outputs
from the system. In the case of IMEX, for example, wells can be sources or sinks depending on
whether they are production of injection wells.
Before running the coupled model, the individual models have to be modeled satisfactorily in
their separate application. Thus an IMEX data set which is the case to be loaded into
RESOLVE must be able to run without RESOLVE.
Once the case is loaded into RESOLVE, equivalent sources and sinks can be connected
together graphically. For example, wells in an IMEX reservoir can be connected to their
equivalent wells in GAP (P.E.s surface network analysis package).
When a simulation is run, RESOLVE collects and passes data between all the coupled
applications. The data passed in all cases is an IPR table (Inflow Performance Relationship a table of node pressures against phase rates).
When a simulation is run, RESOLVE will perform the following steps:
1. Reload and initialize all cases
2. Broadcast to all clients (IMEX) to calculate inflows at all sources
3. Pass the inflow tables to the connected sinks
4. Broadcast a solve command to all connected modules
5. Pass the optimized results back to the clients (IMEX) that generated the inflow
relation
6. Broadcast to all clients (IMEX) to perform a timestep of the requested length
7. Return to step 2
At each of the above steps, the command (to perform timesteps etc) is broadcast to all the
client applications simultaneously; this allows the surface network to couple to multiple
reservoirs models.
User's Guide IMEX
Appendix H 1109
IMEX Timesteps and Dates / RESOLVE Timesteps and Dates

During a coupled run RESOLVE allows GAP to run using a timestep, whose size is
determined by RESOLVE, to solve the Surface network (RESOLVE timestep).
After completing the timestep, RESOLVE passes well constraints for each coupled well back
to IMEX along with the date (RESOLVE date) at the end of the RESOLVE timestep.
IMEX is now expected to solve the reservoir problem until the same RESOLVE date.
GAP waits until IMEX completes this solution. IMEX can take as many internal (IMEX
timesteps) as required to reach the RESOLVE date.
When IMEX reaches the RESOLVE date, it passes IPR tables and other information back to
GAP through RESOLVE.
Considerations and Guidelines
Connections between the reservoir and the surface network are made at well bottom-hole
conditions. Thus GAP must handle flow in the wellbore.
Some or all IMEX wells can be linked to the GAP model through RESOLVE. IMEX
generates IPR tables for those wells that are linked to the GAP model and in turn, gets
instructions from RESOLVE on how to control those wells.
In general IMEX wells linked to GAP have their operating constraints and schedules
controlled by GAP. IMEX wells not linked to GAP have their operating constraints and
schedules controlled by IMEX.
IMEX (after 2006.12) can control some aspects of a wells operating condition, specifically:
1. Any well monitor defined in the *MONITOR keyword is allowed
2. Keywords which control well and/or layer conditions in a *TRIGGER definition
are allowed.
Details are given in the Recurrent data section of Preparation of IMEX data set section below.
IMEX is connected to GAP through RESOLVE on a per well basis, using IPR curves. VFP
tables must be defined in GAP using a program like Prosper. In order to use the linked model,
the GAP method for determining optimum rates must be set to VFP/IPR curve intersection.
Normally GAP (through RESOLVE) controls the timestep the coupled model takes, however
IMEX can be made to provide input to RESOLVE about timestep selection. RESOLVE will
then select the minimum of the GAP and IMEX timesteps.
If the RESOLVE timestep to the next RESOLVE date is too large, it is possible that an IMEX
wells BHP may vary considerably while running to this date. Because IMEX no longer
controls its own well constraints for coupled wells, a failsafe was implemented to limit a
wells minimum pressure to 14.7 psi. The failsafe can be avoided by reducing the RESOLVE
maximum timestep size and/or by letting RESOLVE use its own adaptive timestep selector.
The IMEX start date defined in the recurrent data set should be equal to or later than the
GAP/RESOLVE start date. It may not be earlier than the GAP/RESOLVE start date. If the
IMEX start date is later than the GAP/RESOLVE start date, wells in the linked model will be
masked out until IMEX start date is reached.
1110 Appendix H
User's Guide IMEX
A linked model can restart from a non-linked run, e.g. a GAP linked model can run a
prediction starting from the end of an IMEX only reservoir history match simulation.
Preparation of the IMEX Data Set
There are no GAP link specific keywords required for IMEX to couple with GAP. Although
the following must be done:
1. Reservoir temperature (*TRES) must be defined in the IMEX Component
Properties section.
2. The PVT of the reservoir and surface network should be identical. In other words,
at reservoir temperature the surface network model PVT should match the PVT in
IMEX.
But as listed below, some restrictions and differences in how well/recurrent data is interpreted
are imposed. The following well data restrictions only apply to IMEX-GAP linked wells.
Recurrent Data
In recurrent data, for linked wells:
Any operation that changes the value of a well rate constraint is ignored. Rates are
exclusively determined by GAP.
ON-TIME Fraction
For linked wells, at each RESOLVE date, the GAP down-time percentage is passed to IMEX
as an on-time fraction. IMEX uses the received on-time fraction as if the same value was
input through the IMEX recurrent data. Essentially the definition/handling of on-time
fraction for coupled wells is moved to the GAP network.
For unlinked wells, on-time fraction should be given in IMEX recurrent data as usual.
IMEX passes back on-time fraction for all wells to RESOLVE, so that it is accessible via a
RESOLVE script and readable in the RESOLVE reporting window.
Opening/Closing Wells
IMEX passes open / shut-in status for linked wells to RESOLVE at each RESOLVE date.
After receiving this well status information from IMEX, GAP honors those wells IMEX shuts
in and may add other shut-in wells based on its schedule or optimization procedure.
RESOLVE then runs a new GAP timestep and passes the updated well status back to IMEX.
IMEX keeps the received well status for the IMEX timesteps required to reach the new
RESOLVE date.
IMEX can attempt to open a well (using a monitor or well control in a trigger), but if the
GAP schedule or the GAP network optimization overrides this, the well remains shut-in until
the next RESOLVE date, where it will be checked again by GAP, and if the network schedule
or optimization allows the well to open, it will open. This checking will occur at every
RESOLVE date.
So either IMEX or GAP can shut-in a well, but both IMEX and GAP must agree that a well
should be open before it actually opens.
User's Guide IMEX
Appendix H 1111
From release 2010.10, IMEX is able to receive a shutin with cross-flow command from
RESOLVE, which closes the well above the formation. If a well in an IMEX dataset is defined
with a FULLY-MIXED cross-flow model (the default), the well will cross-flow below the
formation when it receives the shutin with cross-flow command from RESOLVE. This
improves well pressure continuity. Please check the RESOLVE manual for availability.
Monitors and Triggers
IMEX well monitors, group monitors and triggers are allowed to operate on wells or well
layers only. Details are described below. The allowed operations are restricted to those which
open or shut-in wells or well layers, or that modifies productivity / injectivity indices of the
wells or well layers.
IMEX and GAP ONLY exchange information at each RESOLVE date, it is critical that the
well and well layer status is not changed between RESOLVE dates.
There are two options that have been modified for IMEX-GAP coupled runs which ensure
information exchange only at RESOLVE dates. This is done by only checking Monitor
(*MONITOR) violations (for linked wells) and all trigger actions (*TRIGGER) at RESOLVE
dates.
a) Monitor
For linked wells, monitors (*MONITOR) are checked at RESOLVE dates ONLY.
For unlinked wells, monitors are checked at each IMEX timestep.
For well groups defined in IMEX, if a group includes linked wells, the applicable
group monitor option is limited to the group monitor (*GCONM) and the group
monitor will only be checked at RESOLVE dates. On the other hand, if a group
does not include linked wells, group monitors can be applied as usual.
b) Trigger
Triggers (*TRIGGER) are the means by which the user can ensure that well
controls are only applied when IMEX and GAP are synchronized. The user must
place all well control keywords within triggers as in linked runs triggers are only
checked at RESOLVE dates.
Keywords that directly change the well status or that modify productivity for
linked wells (i.e. not in triggers) must be avoided. For example, well 2 is a linked
well and the date 2000 1 1 is not a RESOLVE date. The inputs,
*DATE 2000 01 01
*SHUTIN 2
will shut-in well 2 between two RESOLVE dates, and therefore cause a rate
mismatch between IMEX and GAP. By rate mismatch we refer to the situation
where the network model (GAP) and reservoir model (IMEX) are assuming
different rates or well indices for the same well. This may include having a well
open in one model and shut-in in another.
This inconsistency must be avoided by using triggers (as shown in the example
below).
1112 Appendix H
User's Guide IMEX
In this example, we redefine the layers for well no. 2, then multiply its original
productivity index by five times at the first RESOLVE DATE after 2000 1 1
*DATE 2000 1 1
*TRIGGER
'trg_well2'
*ON_ELAPSED
*GEOMETRY
*K
0.07
0.37
*PERFV
*GEO
2
** KF
FF
2:3
1.0
*SETPI *MULTO 2
5.0
*END_TRIGGER
'TIME'
1.0
0.0
treltd
>
0.0
The use of this elapsed time trigger ensures that the keywords within the trigger
will be read at the first possible RESOLVE date following 2000 1 1
The use of the *ON_ELAPSED TIME treltd > 0.0 trigger would normally
force the trigger to fire immediately, but as this is a linked model, the trigger is
checked at the next RESOLVE date.
Keywords *OPEN and *SHUTIN for wells should be included in triggers. For
example,
*DATE 2000 1 1
*TRIGGER
'shut 2'
*ON_ELAPSED
*SHUTIN
'PRODUCER-2'
*END_TRIGGER
'TIME'
treltd
>
0.0
The use of this trigger above will shut-in 'PRODUCER-2 at the first RESOLVE
date following 2000 1 1.
Of course triggers can still be used to control when a well is shut-in based on a real
criteria, such as in the example below, where a well is shut-in when its well
bottom-hole pressure is less than 1000.00 psia.
*TRIGGER
'trg_bhp2'
*ON_WELL
*SHUTIN
'PRODUCER-2'
*END_TRIGGER
'PRODUCER-2'
BHP
<
1000.0
The trigger above will be fired at the RESOLVE date when the bottom-hole
pressure of PRODUCER-2 is lower than 1000 psia.
IMEX does not exchange well layer status information with GAP. However, similar to well
status, changes for well layer status should be made to happen at RESOLVE dates using
triggers or if possible monitors.
Well Constraint Definitions/Use/Restrictions
For coupled producers, a non-zero surface rate constraint should be given just to define the
target stream phase for IMEX to use with the coupled well. For example,
*OPERATE *MAX *STO 500
indicates the target stream of the producer is oil, the rate itself is ignored.
The target stream phase is used to determine the IMEX constraint from the three phase rates
received from GAP. Within the recurrent data, the keyword *TARGET with non zero rate
can be used to shift the target stream phase from one phase to another. *TARGET for linked
wells should also be placed within a trigger.
User's Guide IMEX
Appendix H 1113
Similarly, non-zero STG and STW can be used to indicate a gas and a water producer (if
applicable) respectively. If more than one of the STO, STG and STW constraints are defined,
the first will be used.
For injectors, the injection stream to be used is given by keyword *INCOMP.
Bottom-hole pressure constraints, if applicable, should be defined in GAP. A BHP constraint
given in IMEX is ignored for rate calculation purposes. However, for injectors, if the keyword:
OPERATE *MAX *BHP 1000
is given, the value of *BHP (1000) will be used to define as the upper limit for pressure in that
wells IPR table which IMEX passes to GAP through RESOLVE. It will not be used as an
The violation action *SHUTIN within the *OPERATE keyword should be avoided. For
example,
*OPERATE *MIN *BHP 1000 *SHUTIN
for a producer may act at any IMEX timestep and cause a rate mismatch between IMEX and
GAP. Such conditional operations may be performed instead by the monitor keyword or
within a trigger.
Injectors must be of the default mobility weighted type.
DATE and TIME
DATE and TIME keywords can normally be used within the IMEX data set. These can be
used to obtain IMEX output when the WPRN/WSRF DATE output options are used.
The start date of IMEX cannot be earlier than the start date of the RESOLVE date
schedule. The IMEX start date can be an IMEX Restart start date.
The stop date in the IMEX data set should be later than or equal to the stop date specified in
the RESOLVE schedule.
Scripting
Using RESOLVE Scripts
A RESOLVE script is able to access a portion IMEXs simulation results, and can be
written to do calculations using information from both IMEX and GAP. This allows
RESOLVE to access IMEX variables which can be used by RESOLVE, as part of a
scripted strategy, to control the network in GAP.
All of the wells, groups and sectors defined in IMEX pass results back to RESOLVE,
where the well and group rates are average values between two synchronized dates. For
accessible IMEX results and their scripting format, please see the IMEX Results
Accessible from RESOLVE Script in this appendix.
Accessible IMEX results are also reported in the RESOLVE Reporting window.
A script is normally designed to execute pre or post the solution of a specified instance
(for example, before the network is solved or after the network is solved).
1114 Appendix H
User's Guide IMEX
Voidage Replacement
RESOLVE is equipped with a wizard which generates scripts to do voidage
replacement in a reservoir model.
IMEX provides the necessary simulation results to let the script work out the amount to
inject from the GAP surface network. Wells can be grouped as a voidage replacement
unit so as to support pattern replacement.
Groups in a voidage replacement script do not have to be defined in IMEX (as well
groups) or in GAP.
Please refer the RESOLVE manual for instructions in how to use the wizard.
Editing Scripts
Generated scripts can be modified for various purposes and scenarios. For example, it
may become necessary to carry out voidage replacement with a reservoir model which
does not have all the wells linked in GAP. In this case, the automatically generated script
will only include linked wells. However, the script can be modified (hand edited) to
access the results for all wells in the reservoir model.
Script Naming Conventions
Care should be taken when naming instances in scripts due to RESOLVEs naming
conventions. Scripting is designed to execute pre or post the solution of a specified
named instance. If two IMEX instances are given the names IMEX and IMEX-1, a
script for instance IMEX-1 will also be executed for IMEX as the name IMEX is
contained within IMEX-1.
This can be avoided by giving the names IMEX-1 and IMEX-2 to the two IMEX
instances.
Resolve Reporting
The Resolve reporting window contains IMEX result data at each synchronized date.
Different from GAP reporting, the IMEX reporting value at a corresponding
RESOLVE date represents the result for the passed timestep, that is, from the previous
synchronized date to current synchronized date.
The user may notice that under the simulator instance name, for example IMEX,
there are well aggregate values. The same reporting entries can also be found under
IMEX group names. The top group name, FIELD for example, includes all IMEX
wells. The difference between well values reported under FIELD and IMEX is that
the instance name IMEX includes the aggregate for only linked wells while FIELD
includes all IMEX wells.
Other Considerations
Multiple Reservoirs
Other instances of IMEX can be set up in a RESOLVE model, allowing a surface
network model to connect to multiple IMEX reservoir models. These IMEX models
may have different start dates. One of the models should have a start time equal to the
GAP/RESOLVE start date.
User's Guide IMEX
Appendix H 1115
When multiple instances of IMEX are set up on the RESOLVE interface, each instance
should be given a different name. Each instance may point to the same executable.
IMEX and RESOLVE communicate through ASCII files. Four ASCII files per IMEX
instance are created and updated during the linked run. The files have prefix of the IMEX
data set and suffixes related to their use within the linked model (LSResolve, LDResolve,
LSImex, and LDImex). These files must not be altered during a linked model run.
Getting Started
a) Register the imexgaplink.dll as a driver for RESOLVE
Start RESOLVE and then choose the Register Driver from Drivers. In the
window similar to window below, click the Register button to assign the location
of imexgaplink.dll. Step a) only needs to be done initially or when you need to
update drivers as RESOLVE remembers the location of registered Drivers.
Click OK to finish.
b) Open a new RESOLVE Graphic View
Click the File -> New to start a new graphic view in RESOLVE.
c) Add the GAP model into RESOLVE
From Edit System -> Add Client Program, choose GAP. Then add the GAP
icon onto the RESOLVE view. Double click the GAP icon. In the edit window,
give the location of .GAP file.
1116 Appendix H
User's Guide IMEX
d) Add the IMEX model into RESOLVE

Similar to the steps used to add a GAP model, from Edit System -> Add Client
Program, choose IMEX and add it onto the RESOLVE view. Edit the CMG
Case setting by double clicking the IMEX icon.
e) IMEX Case Settings (see graphic below)
Select a computer
Local or remote job submission is possible using the Computer combo box. By
default, <Local Computer> is selected for running a simulation on a local
machine. The user can also select a network computer as the target machine. Using
the Computer combo box, the user can type the computer name, choose a listed
computer or select <More > to browse the network. The targeted remote
computer should be able to receive a SSH command to invoke the simulator. Please
see Remote Simulation Job Submission via SSH in this appendix for explanations.
User's Guide IMEX
Appendix H 1117
Assigning the simulator and dataset

In all cases, the user needs to give the location of both the IMEX executable and the
IMEX data set. Please note the given paths should be locatable by the target machine.
When the target machine is a remote Windows computer, paths using the
Universal Naming Convention (UNC path, e.g. \\SIM-SERVER1\dataset) are
expected. The user can use the browser button to find them.
Entries for remote job submission
The Entries for remote job submission text boxes (SSH command update and
Local path to work folder) are only active when a remote computer is selected.
The SSH command button is used to display/update the ssh command. The
displayed SSH command is also editable. For example, it is necessary to edit this
command if the user has a different login name on the server computer than that on
the local computer. In this case it is necessary to alter ssh l username_on_client
to ssh l username_on_server.
Local path to work folder is the local path that points to the folder/directory
where the IMEX dataset exists. Please see Location of the work folder in
Remote Simulation Job Submission via SSH below for a more detailed
description of the entry.
Other options
The Time step selection for network solution radio buttons provides the option
for IMEX feedback in determining the next timestep size. If feedback is enabled,
the timestep size used by RESOLVE will not be greater than the timestep selected
by IMEX for the next step.
Entry of Additional command switches for the executable can be used to input
additional simulation options.
Restart case
It is possible to give the restart file (.irf file) via command option. Such as,
r case1.irf
However there is a little complication in the case of remote job submission. The
ssh command needs the file name, case1.irf, be quoted and have back slash (\) be
escaped. Hence it is recommended to give the .irf file within the dataset using
INDEX-IN keyword.
If the -r option has to be used anyway, the next two examples can be used as a
guide.
For Unix server with restart file /cmg/datasets/case1.irf, the option should be:
-r \"/cmg/datasets/case1.irf\"
For Windows server with restart file \\cmg\datasets\case1.irf, the option should be:
-r \"\\\\cmg\\datasets\\case1.irf\"
1118 Appendix H
User's Guide IMEX
Use of Start and OK button

From release IMEX-2007-11, a new Start button is added to perform the functions
of the original OK button. Start starts the model initialization. After the simulator
finishes its initialization, the Start button converts to Stop, so that user has a
chance to stop the simulation at its initialized stage. The remaining OK button is
used to only save the current case settings, which does not execute the simulator.
However, if the file name of the dataset is changed, the OK button will become
inactive. This ensures that, in order to accept the replaced data set, the user needs to
start a simulation to update the well information.
User's Guide IMEX
Appendix H 1119
f) Link the wells between IMEX and GAP

Use the link function to connect the corresponding wells between IMEX and GAP.
1120 Appendix H
User's Guide IMEX
g) Set RESOLVEs Schedule

RESOLVEs schedule must be set before starting a run. The timestep mode can be
Fixed or Adaptive. Note that if the fixed timestep is chosen and the option
allowing IMEX timestep selection feedback is on, IMEX will still be able to
reduce the timestep to values less than the fixed timestep size, if its internal
timestep selector determines this is necessary.
h) Save the RESOLVE file and Run

At this step, the user can start the linked run by clicking the start button on
RESOLVEs tools bar. The user can also save the current settings to a RESOLVE
file with the name suffix of .rsl. Loading the .rsl file provides a shortcut to starting
a case without the duplication of the steps from b to f.
From release IMEX-2007-11, the saved .rsl file includes IMEX well information,
so that the link settings can be established by only loading the .rsl file. This is
convenient if the user wants to create a new case by only loading and then
modifying an existing case without initializing the existing reservoir model.
About the IMEX log
There may be a date mismatch between the log of IMEX and RESOLVE /GAP of less than
one day at most. This is due to differences in how IMEX and RESOLVE/GAP calculates date
from time. The mismatch only affects the date displayed and does not affect the actual time
used in the simulation. The mismatch will be addressed shortly.
User's Guide IMEX
Appendix H 1121
Remote Simulation Job Submission via SSH

The IMEX-GAP link DLL (IMEX-2007-11 and later) is able to submit simulation jobs to
network computers via Secure SHELL (SSH). Here the network refers to the intranet/LAN
and simulator refers to IMEX-2007-11 or later. The required environments for remote
submission are:
a) The targeted remote computer should be a SSH server (running SSH service) and
the local computer should have a SSH client program installed. The SSH client
issues a ssh command to execute the simulation on a SSH server.
b) Both local and remote computers should be able to create/read/write files in the
folder/directory where the simulation dataset exists. IMEX and the link DLL
exchange data in that work folder.
SSH Server
Linux and IBM AIX computers are equipped with OpenSSH which provides SSH Daemon
(SSHD) as the SSH service.
Windows based computers need to install third party software to run SSH service. For all our
tests, WinSSHD (http://www.bitvise.com/ ) was used to provide the SSH service for
Windows. Tested Windows platform were Windows Server 2000, Server 2003 and XP 64 bit.
SSH Client
The SSH client must be on Windows, because the local machine needs to run RESOLVE, a
Windows based program. There are quite a few software vendors provide SSH clients on
Windows at a low cost. We developed/tested using CopSSH
(http://www.itefix.no/phpws/index.php ) which is an OpenSSH derived Windows version.
The client program has been tested on Windows 2000, XP, XP 64 bit and Vista.
Note 1: After the installation of CopSSH, the user may need to edit the path of the
Windows environment variable to add the SSH bin directory, for example: C:\Program
Files\copSSH\bin.
Note 2: The installation of CopSSH, will install both the server and client software on the
users PC. The user may want to disable the CopSSH server service as only the client
software is required. This service is called Openssh SSHD Service.
SSH Settings
There are various approaches that can be used to enable SSH communication. The following
descriptions are intended to give an example of one approach which was used in our testing
process. For technical details, user needs to refer to the SSH software manual.
The ultimate object of the SSH settings is to allow user to invoke remote simulation by
issuing an one line ssh command at a local command console. For example, the following
command line should directly start the IMEX with the dataset, dataset1.dat,
ssh l username_on server simsvr d:\cmgexe\imex.exe f d:\cmgdata\dataset1.dat
where the simsvr is a remote Windows computer.
1122 Appendix H
User's Guide IMEX
In order to do so, the remote computer (SSH server) should be set so as to accept the user
login without password. Instead of password authentication, SSH client and server, by
default, use key authentication to establish connection.
Generating Key Pair
The user can use ssh-keygen.exe available with CopSSH to generate a private/public key pair.
Using ssh-keygen.exe, two key files, for example id_rsa and id_rsapub, should be generated into
the folder .ssh, which is created under the users home directory of the local Windows. An
empty passphrase is suggested when doing key pair generation. If the key pair is made with
a passphrase, the passphrase will be asked for each time when SSH attempts to login.
Registering Public Key in the Sever
For the UNIX/OpenSSH system, the user needs to copy the public key (e.g. from file
id_rsapub) and append it into the file: $HOME/.ssh/authorized_keys.
For the Windows/WinSSHD system, the system administrator needs to start the WinSSHD
Control Panel/ Settings / Edit/View settings window. Under the Settings tree view, click
Access Control/Windows accounts and add the user and then import the users public key.
When initially setting up the newly installed WinSSHD, the system administrator also needs
to register the network domain name. That is, under the same Settings tree view, choose
Domain order and click the Add to add the domain name.
Test SSH
After the above steps are done, the user can test SSH by trying a ssh command from the local
command console:
ssh simsvr set
If the above command gives out the user environment variables in simsvr, SSH has been
successfully set up. Here, it is assumed that in the above command users login name is the
same on both the server and the local computer.
If the above test is the first time that the user tries to link to the SSH server, SSH will indicate
that it cannot establish authenticity and ask if the user wants to continue. Here, the user
only needs to reply yes so as to allow SSH to import the host key (servers public key), to
the client side.
Settings in Linux/AIX for the CMG Simulator Environment
In order to make sure the SSH login reads the UNIX shell startup file (.cshrc, .kshrc, etc.)
which includes the values for CMG_HOME, LSFORCEHOST and LD_LIBRARY_PATH, the
following should be done in the OpenSSH server:
a) In users $HOME/.ssh directory, create a file with name environment. Add one
line, ENV=/$HOME/.kshrc, into the file (assuming the Kshell is used).
b) Make sure /etc/ssh/sshd_config file contains the next line:
PermitUserEnvironment yes
SSHD needs to be restarted to ensure the updated sshd_config file is read.
Above settings make the $HOME/.kshrc be read when SSH does login.
User's Guide IMEX
Appendix H 1123
Alternatively, user can also directly edit the ssh command line to make the shell startup file
be read. Suppose the ssh command is the next line:
ssh simsvr "\"/cmg/
The command can be modified as:

ssh simsvr ". $HOME/.kshrc && \"/cmg/
The above command forces the $HOME/.kshrc to be read before the simulation is executed.
This also provides a workaround if the setting of the ENV=/$HOME/.kshrc does not work.
Another alternative method to let ssh read parameters CMG_HOME, LSFORCEHOST and
LD_LIBRARY_PATH is to provide these parameters directly into the
$HOME/.ssh/environment file.
If the $HOME is a shared directory among UNIX machines and those hosts need different
CMG variable settings, the shell startup file can include a case command to account for
different hosts. For example,
case $(hostname) in
simsvr | simsvr1 ) LD_LIBRARY_PATH=/usr/cmg/imex/Linux_x64/lib:
/opt/intel/fce/9.1.036/lib; LSFORCEHOST=lserv; export
LD_LIBRARY_PATH LSFORCEHOST;;
aixsvr | aixsvr1) LSFORCEHOST=lserv; export LSFORCEHOST;;
esac
Location of the work folder

It is recommended to locate the work folder on the local disk space of the machine on which
the simulation runs. This method simplifies file access between IMEX and Resolve. The
work folder can also be made on a disk controlled by a third party file server (by this we
mean a disk not on either the computer running IMEX or the computer running RESOLVE).
It has been confirmed by tests that the file server can be either a Windows or Linux
computer.
A Local path to work folder is required in the CMG case setting window, which tells
Resolve how to access to the work folder. In the entry, the path should be addressed simply
and directly.
Known Problems and Workaround
Samba Server
If IMEX is running in a UNIX computer and the work folder/directory is located on a drive
on that computer or a third party UNIX file server, the UNIX of the folder should run Samba
server letting Resolve be able to see the folder, it is recommended to stop the
opportunistic file locking for those communication files. That is, in the Samba
configuration file, e.g. /etc/samba/smb.conf, add a line for the share where the work folder is
located:
veto oplock files =/*.LSImex/*.LSResolve/*.LSGem/*.LDImex/*.LDResolve/*.LDGem/
(E., Robert et al: Using Samba, Chapter 5). The user needs to restart the Samba demon to
make sure the altered smb.conf is used.
1124 Appendix H
User's Guide IMEX
Linux Simulation Run with Work Folder on a Windows Computer - Using Samba Client
(smbfs)
Linux machine accesses the work folder via the smbfs file system. That is the share with the
work folder is mounted in Linux with -t smbfs option. For example,
mount -t smbfs -o username=cmguser //win64/d /cmg/win64
Due to Windows file locking, IMEX and Resolve may lose communication. If this occurs, the
user needs to check how the Local path to work folder is defined. The path should start
with the server-name of the Linux platform. In the above example, suppose the Linux
computer is named Linux_CPU, the path should be \\Linux_CPU\cmg\win64\.
IMEX Results Accessible from RESOLVE Script
IMEX variables that are accessible from RESOLVE scripts via DoGet() are described below.
All specifications are case sensitive.
Use:
DoGet("specification")
Total Numbers
Well count
instance_name.nWell
Well layer count
instance_name.Well[{well_name}].nLayer
or
instance_name.Well[{#well_number}].nLayer
Group count
instance_name.nGroup
Sector count
instance_name.nSector
Example:
DoGet("IMEX.nWell")
DoGet("IMEX.Well[{OIL1}].nLayer)
DoGet("IMEX.nGroup")
Well Variables
Format of specification:
instance_name.Well[{well_name}].well_variable_name
or
instance_name.Well[{#well_number}]. well_variable_name
User's Guide IMEX
Appendix H 1125
Note: well_number is the well number used in IMEX. Please refer the IMEX user manual.
The character # is placed before well_number to signal a number not a name is being entered.
Example:
DoGet("IMEX.Well[{OIL1}].OilRatSC")
DoGet("IMEX.Well[{#1}].OilRatRC")
Well Variable List

Variable name
Descriptions
Units/Remarks
WelTyp
Well type; Indicates for producer or injector.
value =
WelLnk
Well link state. Indicates the well is linked to

the GAP surface network or not.
Please note at the moment of initial, all IMEX
wells are assumed to be linked.
Well bottom-hole temperature
Well bottom-hole pressure
Well reference grid block pressure
Well grid block mobility weighted pressure
Oil rate at stock-tank condition
Water rate at stock-tank condition
Gas rate at stock-tank condition
Oil rate at bottom-hole condition
Water rate at bottom-hole condition
Gas rate at bottom-hole condition
Bottom-hole fluid rate
Cumulative bottom-hole fluid
Well on-time fraction
value =
TmpBHF
PrsBHF
PrsBlkRef
PrsBlkMow
OilRatSC
WatRatSC
GasRatSC
OilRatRC
WatRatRC
GasRatRC
BHFRatRC
BHFCumRC
OnTimeFrc
1: producer
-1: injector
1: linked
0: unlinked
deg.F
psig
psig
psig
STB/day
STB/day
MMscf/day
RB/day
RB/day
MMrcf/day
RB/day
RB
Fraction (0 ~ 1)
Note: Above well rates represent average rates.

Layer Variables
instance_name.Layer[{well_name}][{layer_number}].layer_variable_name
or
instance_name.Layer[{#well_number}][{layer_number}].layer_variable_name
Note: well_number is the well number used in IMEX. Please refer the IMEX user
manual. The character # is placed before well_number to signal a number not a name is
being entered.
layer_number is the input sequence number of the layers given by the IMEX
perforation definition. layer_number can also be a range of layers, expressed as
layer_num1:layer_num2, where layer_num1 is less or equals to layer_num2.
IMEX perforations are defined via PERF or PERFV keywords within the IMEX data set.
1126 Appendix H
User's Guide IMEX
Example:
DoGet("IMEX.Layer[{OIL1}][{5}].WatRatSC")
DoGet("IMEX.Layer[{#1}][{5}].GasRatRC")
Layer Variable List

Variable name
Descriptions
Units/Remarks
LaySta
Layer status; Open, shutin or closed (not

reopenable)
value =
1: Open
0: Shutin
-1: Closed
Depth
Block center depth of the layer
ft
PrsWlb
Wellbore pressure at the depth of the layer
psig
PrsLay
Layer block pressure
psig
OilRatSC
Oil rate at stock-tank condition
STB/day
WatRatSC
Water rate at stock-tank condition
STB/day
GasRatSC
Gas rate at stock-tank condition
MMscf/day
OilRatRC
Oil rate at bottom-hole condition
RB/day
WatRatRC
Water rate at bottom-hole condition
RB/day
GasRatRC
Gas rate at bottom-hole condition
MMrcf/day
BHFRatRC
Bottom-hole fluid rate
RB/day
Note: Above layer rates represent average rates. A negative layer rate indicates the layer is
backflowing with respect to its well type: for a producer this would be an injecting layer or
for an injector this would be a producing layer.
Group Variables
instance_name.Grup[{group_name}].group_variable_name
Note: If there is no user group defined in the IMEX data set, group name FIELD is
used to represent all the wells defined in the IMEX data set. Otherwise, a user given top
node name can be used.
Example:
DoGet("IMEX.Grup[{Gather-1}].OilRatSCPrd")
DoGet("IMEX.Grup[{FIELD}].OilRatRCPrd")
User's Guide IMEX
Appendix H 1127
Group Variable List

Variable name
Descriptions
Units/Remarks
OilRatSCPrd
WatRatSCPrd
GasRatSCPrd
OilRatRCPrd
WatRatRCPrd
GasRatRCPrd
BHFRatRCPrd
BHFCumRCPrd
WatRatSCInj
GasRatSCInj
WatRatRCInj
GasRatRCInj
BHFRatRCInj
BHFCumRCInj
Oil production rate at stock-tank condition

Water production rate at stock-tank condition
Gas production rate at stock-tank condition
Oil production rate at bottom-hole condition
Water production rate at bottom-hole condition
Gas production rate at bottom-hole condition
Bottom-hole fluid production rate
Cumulative bottom-hole fluid production
Water injection rate at stock-tank condition
Gas injection rate at stock-tank condition
Water injection rate at bottom-hole condition
Gas injection rate at bottom-hole condition
Bottom-hole fluid injection rate
Cumulative bottom-hole fluid injection
STB/day
STB/day
MMscf/day
RB/day
RB/day
MMrcf/day
RB/day
RB
STB/day
MMscf/day
RB/day
MMrcf/day
RB/day
RB
Sector Variables
instance_name.Sect[{sector_name}].sector_variable_name
or
instance_name.Sect[{#sector_number}].sector_variable_name
Note: Sector name with Entire Field or sector number 0 is used to represent all
active blocks of the reservoir model. Please refer the IMEX user manual.
Note: sector_number is the sector number used in IMEX. Please refer to the IMEX user
manual. The character # is placed before sector_number to signal a number not a name
is being entered.
Example:
DoGet("IMEX.Sect[{#0}].OilInPSC")
DoGet("IMEX.Sect[{Entire Field}].OilInPSC")
1128 Appendix H
User's Guide IMEX
Sector Variable List

Variable name
Descriptions
Units/Remarks
PrsTPV
PrsHPV
VolTPV
VolHPV
OilInPSC
MbOInPSC
WatInPSC
GasInPSC
FrGInPSC
OilInPRC
WatInPRC
GasInPRC
Average pressure over the total pore volume

Average pressure over the hydrocarbon pore volume
Total pore volume
Hydrocarbon pore volume
Stock-tank oil in-place
Stock-tank mobile oil in-place
Stock-tank water in-place
Stock-tank gas in-place
Stock-tank free gas in-place
Oil in-place at reservoir condition
Water in-place at reservoir condition
Gas in-place at reservoir condition
psig
psig
RB
RB
STB
STB
STB
MMscf
MMscf
RB
RB
MMrcf
Example 1
In the script for voidage replacement, replace the summation of well rates by using group rates:
If (InStr(ModuleList, "GAP") > 0) Then
' fallback values for FVFProd and FVFInj
if (Resolve.Timestep = 0) then
FVFProd0 = 1.3
FVFInj0 = 1
end if
' Estimate FVF values for producers and injectors
Qres = Abs(CDbl(DoGet("IMEX.Grup[{FIELD}].BHFRatRCPrd")))
QWsurf = Abs(CDbl(DoGet("IMEX.Grup[{FIELD}].WatRatSCPrd")))
QOsurf = Abs(CDbl(DoGet("IMEX.Grup[{FIELD}].OilRatSCPrd")))
if (Qres > 0 And QWsurf + QOsurf > 0) then
FVFProd0 = Qres / (QOsurf + QWsurf)
End If
Qres = Abs(CDbl(DoGet("IMEX.Grup[{FIELD}].BHFRatRCInj")))
QWsurf = Abs(CDbl(DoGet("IMEX.Grup[{FIELD}].WatRatSCInj")))
if (Qres > 0 And QWsurf > 0) then
FVFInj0 = Qres / QWsurf
End If
FVFRatio0 = FVFProd0 / FVFInj0
' Get the cumulative voidages so far
VoidProd0 = Abs(CDbl(DoGet("IMEX.Grup[{FIELD}].BHFCumRCPrd")))
VoidInj0 = Abs(CDbl(DoGet("IMEX.Grup[{FIELD}].BHFCumRCInj")))
End If
User's Guide IMEX
Appendix H 1129
Example 2
Set the field oil production constraint to 1/1000 of current field mobile oil in-place and also
make the constraint be limited between 7.5 MBBL and 15 MBBL:
'Get IMEX estimated field mobile oil in-place'
Qprod = Abs(CDbl(DoGet("IMEX.Sect[{#0}].MbOInPSC"))) * 0.001
If (Qprod > 15000) Then
Qprod = 15000
elseif (Qprod < 7500) Then
Qprod = 7500
End If
'Set the constraint in the production system
Call DoSet("GAP.MOD[{PROD}].SEP[{Sep1}].MaxQoil", Qprod)
Call LogMsg("max. Oil set to: " + cstr(Qprod))
1130 Appendix H
User's Guide IMEX
Operation on Well Layers via RESOLVE Script

Since the IMEX release 2008.11, well layers are available to shutin, reopen or do workovers
(change the well index of layer) via Resolve scripting (assuming your IMEXGAPLINK.DLL
supports this feature). Layer operations are only available for those wells that are linked to
the GAP model via Resolve.
Operations are specified via DoCmd() or DoSet(). In following descriptions, specifications
are case sensitive.
Layer Open/Shut Operation
Use:
DoCmd("specification")
instance_name.Layer[{well_name}][{layer_number}].layer_operation
or
instance_name.Layer[{#well_number}][{layer_number}].layer_operation
being entered.
layer_number: same as described in accessing layer variables.
layer_operation can be Shut or Open.
Example:
DoCmd("IMEX.Layer[{OIL1}][{5}].Shut")
DoCmd("IMEX.Layer[{#1}][{3:5}].Shut")
Remarks:
Shutting the final opened layer of a well will shut the well at same time.
Note that using a script to open a layer of a well which was shut because all layers were shut
previously in a Resolve script opens that well.
Note that using a script to open a layer of a well which was shut using AUTO layer status (in
the IMEX *PERF definition) opens that well.
A well cannot be opened using a Resolve script if it was explicitly defined as *SHUTIN in
the IMEX recurrent data.
Layer Workover via DoSet
Use:
DoSet("specification", value)
User's Guide IMEX
Appendix H 1131
instance_name.Layer[{well_name}][{layer_number}].PIMult
or
instance_name.Layer[{#well_number}][{layer_number }].PIMult
being entered.
layer_number: same as described in accessing layer variables.
value is the multiplier to the current PI of the specified layer(s).
Example:
DoSet("IMEX.Layer[{OIL1}][{1:5}].PIMult", 2.0)
Layer operation commands should be placed in the PreSolve section of the script for the
IMEX module in order to have an immediate effect (synchronized with the results that IMEX
reports for that time).
For example, the following script (placed in the PreSolve section of the instance IMEX-1 )
will shut layer 5, 6 and 7 of well no. 3 of the reservoir "IMEX-1", at the same moment when
water cut becomes higher than 60%:
Sub Resolve_PreSolve(ModuleList)
If (InStr(ModuleList, "IMEX-1") > 0) Then
QW = Abs(CDbl(DoGet("IMEX-1.Well[{#3}].WatRatSCPrd")))
QO = Abs(CDbl(DoGet("IMEX-1.Well[{#3}].OilRatSCPrd")))
If (QW/(QO+QW+1.0E-12) > 0.6) Then
DoCmd("IMEX-1.Layer[{#3}][{5:7}].Shut")
Endif
End If
End Sub
Layer operations placed in any other portions of the script will not have an immediate effect.
So if the above script is not placed within the PreSolve section or in the scripting section of a
module other than IMEX-1, the scripted layer operation will not take effect immediately
but will take effect at the next GAP specified timestep after instance IMEX-1 reports the
water cut is higher than 60%.
1132 Appendix H
User's Guide IMEX
Keyword
Index
A
AIM 630
AIMSET 707
AIMWELL 709
ALL 215
ALTER 768
ALTERCP 787
ALTERFS 789
API 599
APIGRAD 447
APIT 600
APPOR-METHOD 941
AQFUNC 375
AQLEAK 368
AQMETHOD 368
AQPROP 368
AQUIFER 368
AUTODRILL 737
B
BGUST 476
BHPDEPTH 780
BHPGRAD 782
BHPHEADINIT 729
BINARY_DATA 224
BOT 470
BWI 494
C
CASEID 243
CCPOR 391
CHECKONLY 197, 244
CHECKRB 647
CIRREVERS 399
CO 492
COMMENT 201
CON 209
CONVERGE 632
COORD 312
User's Guide IMEX
CORNERS 314
CORNER-TOL 365
COT 467
CPOR 346
CPRPOR 391
CROCKTAB 392
CROCKTABC 407
CROCKTABD 402
CROCKTABE 400
CROCKTABH 394
CROCKTABR 409
CROCKTABU 404
CROCKTYPE 389
CTYPE 417
CVO 493
CVW 496
CW 494
CYCPRT_END 923
CYCPRT_START 923
D
DAMP-PCOW-TROIL 547
DATE 702
DATUMDEPTH 607
DEBUG 280
DENSITY 489
DEPLETION 397
DEPTH 298
DEPTH-TOP 304
DGOC 604
DI 292
DIFLIB 437
DIFRAC 342
DIM 233
DIP 306
DISPI 423
DISPJ 423
DISPK 423
DJ 294
DJFRAC 342
DK 296
DKFRAC 342
DPBDT 499
DPCONNECT 341
DPLANES 655
DRILLQ 937
DRSDT 499
DTMAX 624, 705
DTMIN 625, 706
DTOP 300
DTWELL 704
DTYPE 657
Keyword Index 1133
DUALPERM 330
DUALPOR 329
DWGC 604
DWOC 604
E
EGUST 476
EPSPC 542
EPSPCG 542
EQUALSI 208
EQUILIBRATE 735
F
FAULT 366
FAULTARRAY 387
FILENAMES 227
FLUX_ISECTOR 381
FLUX_SECTORNAMES 381
FRACTURE 206
FRACVOL 333
FRWIDTHCOR 567
G
GAPPOR 939
GASZONE 613
GCIOFF 952
GCONCYCLE 921
GCONCYCLE_END 908
GCONCYCLE_START 908
GCONCYCR_END 919
GCONCYCR_START 919
GCONI 862
GCONIMULT 982
GCONM 904
GCONP 856
GCONPMULT 978
GCPOFF 952
GEOMETRY 812
GFUELFRAC 880
GFUELMAXR 882
GLCONTROL 806
GLIFT 803
GLOPT 809
GMKMAXR 874
GMKUPTO 872
GOC_PC 610
GOC_SW 614
GORINT 501
GPHYDDEP 931
GPRODGROUP 876
1134 Keyword Index
GPTABLE 929
GRAVITY 489
GRECYMAXR 878
GRID 283
GROUP 714
GROUPALQ 933
GSALESFRAC 884
GSALESMAXR 886
GUIDEI 949
GUIDEP 949
H
HEADITER 727
HEAD-METHOD 724
HYSKRG 542
HYSKRO 542
I
IJK 210
INCOMP 749
INITIAL 582
INJECTOR 733
INT 219
INTERRUPT 245
INUNIT 250
ISECTOR 380
ITERMAX 644
ITUBE1 993
IVAR 212
IWELLBORE 741
J
JACDOMS 658
JACPAR 660
JFGMAX 559
JFUNC 570
JFW_SHIFT 562
JFWMAX 559
JVAR 213
K
KDIR 290
KRGCL 559
KROCW 559
KROGCG 559
KROIL 548
KRPERF 851
KRWIRO 559
KVAR 214
User's Guide IMEX
L
LAYERGRAD 784
LAYERHFP 846
LAYERIJK 843
LAYERXYZ 839
LEASE 383
LIST 199
M
MANIFOLD 925
MATRIX 205
MAXCHANGE 626
MAXCPU 623
MAXERROR 255
MAXSTEPS 622
MINC 336
MINCYC 636
MINSS 503
MOD 216
MODEL 427
MODELSHUT 735
MONITOR 761
MRC-RESET 801
MXCNRPT 760
N
NCUTS 637
NDARCYCOR 566
NETGROSS 354
NETPAY 352
NEWTONCYC 635
NOLIST 199
NOLISTLIM 200, 249
NONDARCY 563
NORM 626
NORTH 640
NULL 343
NULL-PERF 731
NUMERICAL 621
NWHYFS 1001
O
OMEGASG 502
ON-TIME 989
OPEN 737
OPERATE 751
OUTDIARY 277
User's Guide IMEX
OUTPRN 262
OUTSRF 271
OUTUNIT 253
P
PADSORP 505
PAYDEPTH 302
PB 588
PBS 594
PBT 589
PCGMAX 559
PCON-MANIFOLD 927
PCOW_SHIFT 562
PCWMAX 559
PDEGAA 648
PDEGAB 649
PDEW 591
PDEWT 592
PERF 815
PERFV 824
PERFVHY 831
PERMI 350
PERMJ 350
PERMK 350
PINCHOUTARRAY 360
PINCHOUT-TOL 363
PIVOT 643
PMIX 507
PNTHRDS 656
POLYCONC 597
POR 345
PPATTERN 650
PPERM 512
PRECC 639
PREFCONC 511
PRES 587
PRIOR-FORM 945
PRODUCER 733
PRPOR 346
PSPLIT 279
PTHRESHI 568
PTHRESHJ 568
PTHRESHK 568
PTUBE1 993
PTYPE 498
PVCUTOFF 362
PVISC 510
PVT 432
PVTAPI 444
PVTCOND 455
Keyword Index 1135
PVTG 452
PVTS 485
PVTVO 461
PWELLBORE 744
R
RANGE 327
RANGECHECK 254
REFDEPTH 602
REFINE 317
REFPRES 602
REFPST 484
REFPW 494
RELAX 628
RELTOL 638
REPORTING-GROUP 987
RESTART 256
REWIND 257
RG 207
ROCKFLUID 517
RPT 518
RTYPE 551
RUN 701
S
SCLDMG 576
SCLDPS 574
SCLRMV-LAYER 797
SCLRMV-WELL 795
SCLTBL-LAYER 793
SCLTBL-WELL 791
SCONNECT 385
SD_REINF 577
SDEGREE 642
SEAWATFRC 598
SECTOR 377
SECTORARRAY 379
SECTORNAMES 380
SETPI 775
SGCON 552
SGCRIT 552
SGT 531
SHAPE 337
SHUTIN 737
SLCON 552
SLT 531
SMKMAXR 898
SMKUPTO 896
SO 595
SOIRW 552
SOLVER 645
1136 Keyword Index
SONEQ 618
SORDER 641
SORG 552
SORMAX 552
SORW 552
SPRODGROUP 900
SR2ASCII 247
SR2PREC 248
SRECYMAXR 902
SRFTNG 573
SRFTNW 573
STOP 1003
SUBDOMAIN 331
SVISC 513
SW 596
SWCON 552
SWCRIT 552
SWINIT 616
SWNEQ 618
SWT 520
T
TARGET 771
TIME 703
TITLE1 240
TITLE2 241
TITLE3 242
TRANLI 358
TRANLJ 358
TRANLK 358
TRANSD 335
TRANSF 425
TRANSFER 339
TRANSI 355
TRANSJ 355
TRANSK 355
TRANSLATE 202
TRES 431
TRIGGER 954
U
USER_INPUT 583
V
VERTICAL 583
VGUST 480
VOLMOD 348
VOT 473
VWI 496
User's Guide IMEX
W
WELL 719
WELLALQ 935
WELLINIT 711
WLISTOPEN 739
WLISTSHUT 739
WMKMAXR 890
WMKUPTO 888
WOC_PC 610
WOC_SW 614
WORKOVER-DATA 798
WPRN 259
WPRODGROUP 892
WRECYMAXR 894
WRST 257
WSRF 269
WTMULT 974
X
XCORN 310
XDR 271
XFLOW-MODEL 722
Y
YCORN 310
Z
ZCORN 308
ZGUST 476
User's Guide IMEX
Keyword Index 1137

MX 201010 en

Uploaded by

Copyright:

Available Formats

MX 201010 en

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

MX 201010 en

Uploaded by

Copyright:

Available Formats

User's Guide

By Computer Modelling Group Ltd.

Copyright 1987-2010 Computer Modelling Group Ltd.

The license management portion of this program is based on:

Tel: (403) 531-1300

Fax: (403) 289-8502

Confidentiality: All components of CMG technology including software and related

Important Changes between IMEX 2010.10 and IMEX 2009.10.................................1

Using the Rock Compaction/Dilation Model Option ................................................. 89

Keyword Data Entry System

Introduction to Keyword System.............................................................................. 189

User's Guide IMEX

Controlling Data File Listing (Optional) ..................................................................199

Command Line Input/Output Control.......................................................................221

Items in Output Print File (Optional) ....................................................................... 262

Fundamental Grid Definition (Required) ................................................................. 283

User's Guide IMEX

Pore Volume Cut-Off Threshold (Optional).............................................................362

Fluid Model (Required) ............................................................................................427

Reference Pressure for Gas-Oil Capillary Pressure (Optional) ................................ 484

Notes on Rock-Fluid Data........................................................................................ 515

User's Guide IMEX

Notes on Initial Conditions .......................................................................................579

Numerical Methods Control

Numerical Methods Control Identifier (Optional) ....................................................621

Solver Factorization Degree (Optional) ................................................................... 642

Well and Recurrent Data

Notes on Well and Recurrent Data........................................................................... 663

User's Guide IMEX

Maximum Time-Step Size (Optional) ......................................................................705

Geometric Data for Deviated Well Completions (Conditional)............................... 839

User's Guide IMEX

Alter Well Constraint Value via a Multiplier (Optional) ..........................................974

Documentation, Templates and Execution

Sample Data Sets

Notes on Sample Data Sets .....................................................................................1013

Theoretical Outline .................................................................................................1041

Well Model .............................................................................................................1051

User's Guide IMEX

Calculation of Well Parameters.............................................................................. 1055

Figure D-1a: Numbering of the Grid (Non-Corner Point) .................................... 1059

Polymer Option ...................................................................................................... 1069

The Use of IMEX in Well Testing Applications.................................................... 1087

Analytical Aquifers ................................................................................................ 1095

User's Guide IMEX

Practical Aspects of Using Analytical Aquifers .....................................................1099

Running IMEX Coupled to the GAP Surface Network Model...............................1109

User's Guide IMEX

Important Changes between IMEX 2010.10 and IMEX 2009.10

User's Guide IMEX

Hysteresis Option for Pcow and Krow (Oil-Water System).

User's Guide IMEX

IMEX-GAP Interface Improvements IMEX-GAP Interface Improvements:

User's Guide IMEX

Important Changes between IMEX 2009.10 and IMEX 2008.10

User's Guide IMEX

User's Guide IMEX

Important Changes Between IMEX 2008.10 and IMEX 2007.10

User's Guide IMEX

WOC_SW available with DEPTH_AVE Initialization.

NDBETCR, WSRFTN, GSRFTN, CRSETN, SWCON, SWCRIT, *SOIRW,

PCOW, PCOG, RFG, SWCON, SWCRIT, SOIRW, SORW, SGCON, *SGCRIT,