PE Essentials

Petroleum Engineering
& Economics Essentials

Tools and Techniques to Evaluate Unconventional
(and Conventional) Wells and Reservoirs
Don LeBlanc, PEng

Copyright ©2020
Petroleum Engineering & Economics
Essentials
(PE2 Essentials)
Tools and Techniques to Evaluate

Unconventional (and Conventional) Wells and Reservoirs
Don LeBlanc, PEng

Copyright ©2020
Disclaimer
Although every effort has been made to ensure that the programs that are a part of PE² Essentials
work as expected by checking them against historical models that have been built and used in
the past, the author does not warrant any of the programs for accuracy.
Use of PE² Essentials and its output results are entirely at the user’s risk. The author does not
accept any liability for use of the programs or for decisions made based on the output results.
Acknowledgements
I would like to acknowledge Dr. Viannet Okouma for his input into the eDCA, PDA and Asset
Evaluation tools. His assistance has been greatly appreciated.
Copyright Information
This book and all documentation associated with the PE² Essentials software is Copyright (©2021)
by Eastex Petroleum Consultants Inc.
Petroleum Engineering & Economics Essentials i
Mr. Don LeBlanc, PEng has an in-depth background in petroleum

engineering that includes reservoir engineering, production
engineering, hydraulic fracturing, field development, reservoir
management, formation evaluation, reservoir modelling,
production modelling and reserves estimation resulting from
extensive independent and operating company experience in
Canada and Internationally.
During his 40+ year career he has lived in Indonesia, Denmark and
the US and worked with super-majors, national oil companies, small
independents and consulting firms all over the world. He has worked on projects in the US, SE
Asia, Australia, Croatia, North Sea (Denmark/Norway/UK), Russia, Egypt, East Coast Canada,
Beaufort Sea and Alberta ranging from green (new) fields to brown (old) fields, from offshore to
onshore fields and from conventional to unconventional and shale reservoirs.
Don has extensive experience with reservoir and production engineering and simulation of
conventional and unconventional reservoirs. He has been working with unconventional and shale
reservoirs for >20 years which led to the development of PE² Essentials which includes tools to
evaluate conventional and unconventional wells and reservoirs. Don developed the LeBlanc-
Okouma Power Law (LOPL) decline curve model which was published in the March 2018 issue of
World Oil and the Artificial Neural Network based oil recovery factor model (ANN-RF) which was
published in the March 2019 issue of World Oil.
Don has extensive experience with field development and rehabilitation activities including
IOR/EOR and water and gas flooding and is very knowledgeable in the use of Eclipse, CMG,
Exodus, BOAST, IPM, IAM and numerous other reservoir and production simulators.
Don’s areas of expertise include:
• Field Development, Field Rehabilitation
• Production Surveillance and Reservoir Management
• Classical Reservoir Engineering and Reservoir Modeling (Eclipse, etc)
• Production Engineering and Production Modeling (GAP, etc)
• HPHT, conventional, unconventional, shale fields
• Coupling Reservoir and Production Models
• Artificial Lift / Hydraulic Fracture Stimulation Analysis & Modeling
• Reserves Evaluation / Auditing
• Peer Review / Peer Assist
• Expert Witness (High Court of London)
• IOR / EOR
• Mentoring / Teaching (college and industry courses)
Don received a Bachelor`s degree (Magnis Cum) in Engineering-Physics in 1979 from Dalhousie
University and is a registered Professional Engineer.
ii Table of Content
Table of Contents
Introduction ......................................................................................................................................1
PE Asset Management Essentials ........................................................................................................5
Production Database Tool ..................................................................................................................5
PDB.1 Data Import .................................................................................................................................... 7
PDB.2 Database Editing ........................................................................................................................... 10
Production Data Analysis Tool .......................................................................................................... 14
PDA.1 Reservoir Data Entry..................................................................................................................... 15
PDA.2 Production Data Import ............................................................................................................... 16
PDA.3 PVT Data Entry.............................................................................................................................. 18
PDA.4 Oil and Gas Pseudo Pressure ........................................................................................................ 19
PDA.4.1 Gas Pseudo Pressure....................................................................................................................................................................... 19
PDA.4.2 Oil Pseudo Pressure ........................................................................................................................................................................ 20
PDA.5 Data Validation / Diagnostics ....................................................................................................... 20
PDA.5.1 Average Well................................................................................................................................................................................... 22
PDA.5.2 P90, P50 and P10 Wells .................................................................................................................................................................. 23
PDA.6 Pseudo Steady State vs Boundary Dominated Flow .................................................................... 24
PDA.7 Flow Regime Identification........................................................................................................... 25
PDA.7.1 Example: PDA/eDCA ....................................................................................................................................................................... 27
PDA.8 Flowing Material Balance Analysis ............................................................................................... 28
PDA.8.1 A-G Flowing Material Balance ......................................................................................................................................................... 29
PDA.8.2 Flowing Productivity Index .............................................................................................................................................................. 31
PDA.8.3 Flowing PR/ZR and Flowing PR .......................................................................................................................................................... 32
PDA.9 Analytical Simulator ..................................................................................................................... 34
PDA.10 Numerical Simulator................................................................................................................... 35
PDA.11 INTERPRET-PDA .......................................................................................................................... 36
PDA.12 INTERPRET-PDA WI..................................................................................................................... 37
PDA.13 Exporting Data from the PE Tools Database .............................................................................. 41
PDA.14 Converting Forecast Data to Production Data ........................................................................... 41
PDA.15 PE Tools Database Management ............................................................................................... 42
PDA Appendix – Concepts of Production Data Analysis ......................................................................... 46
A.1 General Concepts ................................................................................................................................................................................... 46
A.2 Basic Diagnostics Definitions .................................................................................................................................................................. 49
A.3 Time Functions ....................................................................................................................................................................................... 52
A.4 Flowing Material Balance (FMB) ............................................................................................................................................................ 55
A.5 Specialized Analysis ................................................................................................................................................................................ 56
A.6 Production Diagnostics Plots .................................................................................................................................................................. 57
PE Basic Essentials ........................................................................................................................... 58
Oil/Gas/Water PVT and Rel Perm Tool.............................................................................................. 58
PVT.1 Gas Properties............................................................................................................................... 61
PVT.1.1 Gas Tc and Pc .................................................................................................................................................................................. 62
PVT.1.2 Gas Z Factor ..................................................................................................................................................................................... 63
PVT.1.3 Gas Viscosity.................................................................................................................................................................................... 63
PVT.1.4 Gas Isothermal Compressibility ....................................................................................................................................................... 64
PVT.1.5 Gas Formation Volume Factor ......................................................................................................................................................... 65
PVT.1.6 Condensate-Gas Ratio ..................................................................................................................................................................... 65
PVT.1.7 Hydrate Formation Temperature .................................................................................................................................................... 66
PVT.1.8 Converting SG to Recombined Reservoir Specific Gravity ............................................................................................................... 67
PVT.2 Gas Component Generation and Analysis .................................................................................... 67
PVT.2.1 Gas Pseudo Component Generation ............................................................................................................................................... 68
PVT.2.2 Gas Shrinkage Factor ....................................................................................................................................................................... 70
PVT.2.3 Gas Liquids ...................................................................................................................................................................................... 71
PVT.2.4 Gas GHV .......................................................................................................................................................................................... 71
PVT.2.5 Gas Dew Point Pressure .................................................................................................................................................................. 72
Petroleum Engineering & Economics Essentials iii
PVT.3 Oil Properties ................................................................................................................................ 73

PVT.3.1 Oil Bo and Pb Correlations ................................................................................................................................................................ 73
PVT.3.2 Oil Viscosity Correlations ................................................................................................................................................................. 75
PVT.3.3 Gas Gravity Correction for Separator Conditions ............................................................................................................................ 77
PVT.3.4 Solution Gas-Oil-Ratio ..................................................................................................................................................................... 77
PVT.3.5 Oil Isothermal Compressibility ........................................................................................................................................................ 78
PVT.3.6 Pseudo Oil Components .................................................................................................................................................................. 79
PVT.4 Water and Rock Properties ........................................................................................................... 81
PVT.4.1 Water Density ................................................................................................................................................................................. 81
PVT.4.2 Solution Gas-Water Ratio ................................................................................................................................................................ 82
PVT.4.3 Water Isothermal Compressibility (Gas Saturated) ......................................................................................................................... 83
PVT.4.4 Water Viscosity ............................................................................................................................................................................... 84
PVT.4.5 Water Formation Volume Factor ..................................................................................................................................................... 84
PVT.4.6 Formation Isothermal Compressibility ............................................................................................................................................ 85
PVT.5 Interfacial Tension ........................................................................................................................ 85
PVT.6 Relative Permeability Data Generation ........................................................................................ 87
Basic Equation of State (EOS) PVT Tool ............................................................................................. 90
EOS.1 Equations of State......................................................................................................................... 90
EOS.2 Soave-Redlich-Kwong (SRK) EOS................................................................................................... 94
EOS.3 Peng-Robinson (PR) EOS ............................................................................................................... 95
EOS.4 Mixing Rule and Binary Interaction Coefficients .......................................................................... 95
EOS.5 EOS Basic Example and EOS Validation ........................................................................................ 99
EOS.6 EOS Tuning .................................................................................................................................. 100
EOS.7 EOS Flash Calculations – CCE, DLE, Separator Test..................................................................... 101
EOS.7.1 Constant Composition Expansion .................................................................................................................................................. 102
EOS.7.2 Differential Liberation Expansion .................................................................................................................................................. 102
EOS.7.3 Separator Test ............................................................................................................................................................................... 104
EOS.7.4 Determining Optimum Separator Operating Conditions ............................................................................................................... 105
EOS.7.5 Adjusting CCE/DLE Data to Optimum Separator Conditions ......................................................................................................... 105
EOS.8 Gas-Oil Recombination for EOS .................................................................................................. 107
Monte Carlo Simulation: Volumetrics ............................................................................................. 108
PE Graph Tool ................................................................................................................................ 114
GRP.1 Data Import ................................................................................................................................ 116
GRP.2 Plotting ....................................................................................................................................... 116
GRP.3 PE Graph Import File Format ...................................................................................................... 119
PE Well Essentials .......................................................................................................................... 120
THP – BHP Tool: Gas Well ............................................................................................................... 120
THPG.1 Average Temperature/Z-Factor Technique ............................................................................. 121
THPG.2 Guo-Ghalambor Technique ...................................................................................................... 122
THPG.3 Options for THP-BHP Calculations ........................................................................................... 123
THPG.3.1 Well Options ............................................................................................................................................................................... 123
THPG.3.2 Minimum gas Rates to Lift Liquid ............................................................................................................................................... 124
THPG.3.3 ∆P Correction Factor ................................................................................................................................................................... 124
THPG.3.4 THP/CHP to BHP Conversion ....................................................................................................................................................... 125
THPG.4 IPR Well Operating Point.......................................................................................................... 126
Quick Log Analysis Tool .................................................................................................................. 128
QLA.1 Log Analysis Models ................................................................................................................... 129
QLA.1.1 Shale Models................................................................................................................................................................................. 129
QLA.1.2 Porosity Models ............................................................................................................................................................................ 130
QLA.1.3 Water Saturation Models .............................................................................................................................................................. 133
QLA.1.4 Permeability Models ..................................................................................................................................................................... 135
QLA.2 Log Analysis Parameters and Results ......................................................................................... 136
QLA.3 Pressure-Depth Analysis ............................................................................................................. 138
iv Table of Content
Hydraulic Fracture Design Tool ....................................................................................................... 141

Artificial Lift Design Tool ................................................................................................................ 147
ALD.1 Rod Pumps .................................................................................................................................. 150
ALD.1.1 Overview ....................................................................................................................................................................................... 150
ALD.1.2 Rod Pump Design .......................................................................................................................................................................... 153
ALD.1.3 Rod Pump Example ....................................................................................................................................................................... 156
ALD.2 Plunger Lift .................................................................................................................................. 158
ALD.2.1 Overview ....................................................................................................................................................................................... 158
ALD.2.2 Plunger Lift Design ........................................................................................................................................................................ 160
ALD.2.3 Plunger Lift Example...................................................................................................................................................................... 162
ALD.3 Hydraulic Jet Pump ..................................................................................................................... 164
ALD.3.1 Overview ....................................................................................................................................................................................... 164
ALD.3.2 Hydraulic Jet Pump Design ............................................................................................................................................................ 166
ALD.3.3 Hydraulic Jet Pump Example ......................................................................................................................................................... 167
ALD.4 Electrical Submersible Pump ...................................................................................................... 169
ALD.4.1 Overview ....................................................................................................................................................................................... 169
ALD.4.2 Electrical Submersible Pump Design ............................................................................................................................................. 171
ALD.4.3 Electrical Submersible Pump Example .......................................................................................................................................... 171
Pressure Transient Analysis Tool .................................................................................................... 173
PTA.1 Pressure Buildup Analysis ........................................................................................................... 175
PTA.2 Analytical Well Test Simulator .................................................................................................... 179
PTA.2 Gas Well Example ....................................................................................................................... 181
Volumetric Analysis Tool ................................................................................................................ 184
MBA.1 General Material Balance .......................................................................................................... 184
MBA.2 Data Input.................................................................................................................................. 186
MBA.3 General Oil Material Balance .................................................................................................... 187
MBA.3.1 Undersaturated Oil / Depletion Drive Reservoir .......................................................................................................................... 188
MBA.3.2 Gas Cap Drive Oil Reservoir ......................................................................................................................................................... 190
MBA.3.3 Water Drive Oil Reservoir ............................................................................................................................................................ 191
MBA.3.4 Combination Gas Cap / Water Drive Oil Reservoir ....................................................................................................................... 192
MBA.3.5 Drive Indices for an Oil Reservoir ................................................................................................................................................. 193
MBA.3.6 Water Drive Reservoir Example ................................................................................................................................................... 195
MBA.4 General Gas Material Balance ................................................................................................... 199
MBA.4.1 Gas Expansion / Volumetric Gas Reservoir .................................................................................................................................. 200
MBA.4.2 Water Drive Gas Reservoir ........................................................................................................................................................... 201
MBA.4.3 Depletion Drive Gas Reservoir Example ....................................................................................................................................... 202
Hydrate Analysis Tool .................................................................................................................... 205
HYD.1 Hydrate Prediction – Specific Gravity Methods ......................................................................... 206
HYD.1.1 Katz (1944).................................................................................................................................................................................... 207
HYD.1.2 Salufu (2013)................................................................................................................................................................................. 208
HYD.1.3 Towler (2005) ............................................................................................................................................................................... 209
HYD.1.4 Makogan (1997)............................................................................................................................................................................ 209
HYD.1.5 Motiee (1991) ............................................................................................................................................................................... 209
HYD.1.6 Sun (2013)..................................................................................................................................................................................... 209
HYD.2 Hydrate Prediction – Gas Composition Methods....................................................................... 210
HYD.2.1 Berge (1986) ................................................................................................................................................................................. 211
HYD.2.2 Poettmann (1989) – Mann (1988) ................................................................................................................................................ 212
HYD.3 Hydrate Temperature Depression.............................................................................................. 213
HYD.3.1 Salinity Temperature Depression – McCain (1990) ....................................................................................................................... 213
HYD.3.2 Basic Inhibitor Temperature Depression ...................................................................................................................................... 213
HYD.4 Hydrate Inhibitor Volume .......................................................................................................... 216
HYD.4.1 Inhibitor Required for Water Phase .............................................................................................................................................. 216
HYD.4.2 Inhibitor Lost to the Vapor Phase ................................................................................................................................................. 219
HYD.4.3 Inhibitor Lost to Condensate Phase .............................................................................................................................................. 221
HYD.4.4 Total Inhibitor Required ................................................................................................................................................................ 222
PE Forecast Essentials .................................................................................................................... 224
Unconventional Forecast Simulator Tool ........................................................................................ 224
Petroleum Engineering & Economics Essentials v
UNC.1 Analytical Model ........................................................................................................................ 226

UNC.2 Numerical Model ....................................................................................................................... 230
UNC.3 PVT Model .................................................................................................................................. 232
UNC.4 Oil/Gas Reservoir Models .......................................................................................................... 235
UNC.5 Fracture Model .......................................................................................................................... 236
UNC.6 Wellbore Model ......................................................................................................................... 239
UNC.7 Forecast Plotting ........................................................................................................................ 240
UNC.8 Unconventional Forecast Economics Model ............................................................................. 241
UNC.8.1 Total Well Costs ............................................................................................................................................................................ 243
UNC.8.2 Escalation Factors ......................................................................................................................................................................... 244
UNC.8.3 Discount Rate ............................................................................................................................................................................... 244
UNC.8.4 Oil/Gas Prices ............................................................................................................................................................................... 245
UNC.8.5 CapEx, OpEx, Revenue, Net Present Value and PIR ...................................................................................................................... 245
UNC.9 History Match Tool .................................................................................................................... 247
UNC.10 Unconventional Forecast Example .......................................................................................... 251
Basic Reservoir Simulation Tool ...................................................................................................... 259
SIM.1 Grid Construction ........................................................................................................................ 261
SIM.2 PVT Properties ............................................................................................................................ 266
SIM.3 Relative Permeability .................................................................................................................. 267
SIM.4 Reservoir Parameters ................................................................................................................. 270
SIM.5 Wellbore Parameters .................................................................................................................. 271
SIM.6 Well Schedule ............................................................................................................................. 272
SIM.7 Simulator Options ....................................................................................................................... 274
SIM.8 Quick Plot .................................................................................................................................... 277
SIM.8.1 Production Forecast Plots.............................................................................................................................................................. 277
SIM.8.2 Static Grid Plots ............................................................................................................................................................................. 278
SIM.8.2 Dynamic Grid Plots ........................................................................................................................................................................ 281
SIM.9 Example – History Matching a Horizontal Oil Well Test ............................................................. 282
Streamtube Waterflood Simulation Tool ........................................................................................ 286
SWF.1 Reservoir Model ......................................................................................................................... 288
SWF.2 Pattern Waterflood Model ........................................................................................................ 288
SWF.3 Example - Choosing a Waterflood Pattern ................................................................................ 289
Misc/Immis CO2 WAG Simulation Tool ............................................................................................ 292
CO2.1 Reservoir Model ......................................................................................................................... 294
CO2.1.1 Minimum Miscibility Pressure ...................................................................................................................................................... 295
CO2.1.2 Degree of Mixing .......................................................................................................................................................................... 295
CO2.1.3 Miscible Phase Relative Permeability Option................................................................................................................................ 295
CO2.1.4 Dykstra-Parsons Coefficient .......................................................................................................................................................... 296
CO2.2 Relative Permeability ................................................................................................................. 296
CO2.3 Pattern Waterflood Model ......................................................................................................... 297
CO2.4 Example: Waterflood Model Comparison .................................................................................. 297
Gas Material Balance Analysis Tool................................................................................................. 299
GMB.1 Multi-Tank Gas MB Model ........................................................................................................ 300
GMB.2 Water Drive Gas MB Model ...................................................................................................... 301
GMB.3 MB Tank Parameters ................................................................................................................. 301
GMB.4 Material Balance Tank Forecast ................................................................................................ 304
GMB.5 Gas Material Balance Example.................................................................................................. 305
Oil Material Balance Analysis Tool .................................................................................................. 309
OMB.1 General Material Balance ......................................................................................................... 309
OMB.2 Aquifer Models ......................................................................................................................... 311
vi Table of Content
OMB.2.1 Schilthuis Steady State Aquifer Model ........................................................................................................................................ 312

OMB.2.2 Fetkovich Aquifer Model ............................................................................................................................................................. 313
OMB.2.3 Van Everdingen and Hurst Unsteady State Infinite Aquifer Model .............................................................................................. 315
OMB.3 Depletion Drive Option ............................................................................................................. 318
OMB.4 Height versus Pore Volume....................................................................................................... 319
OMB.5 Aquifer Model Comparison....................................................................................................... 320
OMB.6 Oil Material Balance Example – History Match ........................................................................ 321
Decline Curve Analysis ................................................................................................................... 323
DCA.1 Introduction ............................................................................................................................... 323
DCA.2 Decline Curve Analysis Tool ....................................................................................................... 323
DCA.3 Production Data Import ............................................................................................................. 328
DCA.4 Enhanced DCA (eDCA) ................................................................................................................ 331
DCA.4.1 Model Fitting in eDCA ................................................................................................................................................................... 331
DCA.4.2 Equivalent Arps Parameters ......................................................................................................................................................... 333
DCA.5 DCA Parameters ......................................................................................................................... 334
DCA.6 DCA Production Forecasting ...................................................................................................... 337
DCA.7 Normalized DCA ......................................................................................................................... 339
DCA.8 DCA Example – Marcellus........................................................................................................... 344
DCA Appendix 1– DCA Theory & Practice ............................................................................................. 347
A1.1 Introduction ....................................................................................................................................................................................... 347
A1.2 Decline Curve Analysis Theory ........................................................................................................................................................... 348
A1.3 Decline Curve Analysis Practice .......................................................................................................................................................... 357
DCA Appendix 2 – Model Formulations ................................................................................................ 367
A2.1 Arps Hyperbolic Model ....................................................................................................................................................................... 367
A2.2 Modified Hyperbolic Model ............................................................................................................................................................... 368
A2.3 Stretched Exponential Model (SEPD) ................................................................................................................................................. 368
A2.4 Power Law Exponential Model (PLE) .................................................................................................................................................. 370
A2.5 Duong Model...................................................................................................................................................................................... 371
A2.6 Logistic Growth Model (LGM) ............................................................................................................................................................ 372
A2.7 LeBlanc-Okouma Power Law Model (LOPM) ...................................................................................................................................... 373
A2.8 Normalized DCA ................................................................................................................................................................................. 374
DCA Appendix 3 – Miscellaneous Definitions ....................................................................................... 377
Monte Carlo DC Production Forecast .............................................................................................. 378
MCD.1 Monte Carlo DCA....................................................................................................................... 379
MCD.2 Monte Carlo DC Forecast Example – Marcellus ........................................................................ 382
PE IOR/EOR/Heavy Oil Essentials .................................................................................................... 384
IOR / EOR Screening Tool ............................................................................................................... 384
IOR.1 Screening Parameters ................................................................................................................. 385
IOR.2 IOR / EOR Screening .................................................................................................................... 386
EOR / Heavy Oil Tool ...................................................................................................................... 388
EOR.1 Minimum Miscibility Pressure .................................................................................................... 389
EOR.2 Reservoir Thermal Properties..................................................................................................... 394
EOR.3 Hot/Cold Water Injection ........................................................................................................... 395
EOR.4 SteamFlood Residual Oil ............................................................................................................. 398
EOR.5 PCP and Sucker Rod Pump Rates ............................................................................................... 399
PE Field Development Essentials .................................................................................................... 400
Field Development Planning Tool ................................................................................................... 400
FDP.1 Well Type Curves ........................................................................................................................ 401
FDP.2 Well Drilling and Completion Planning ....................................................................................... 403
FDP.3 Rig Schedule Planning and Drilling Learning Rate ...................................................................... 404
FDP.4 Facility Planning and Uptime Modelling ..................................................................................... 405
Petroleum Engineering & Economics Essentials vii
FDP.5 Production Modeling .................................................................................................................. 406

FDP.5.1 Well Pressure Calculation .............................................................................................................................................................. 408
FDP.6 Statistical Modeling Using Gaussian Distributions ..................................................................... 409
FDP.6.1 Statistical Well Productivity ........................................................................................................................................................... 410
FDP.6.2 Statistical Rig Availability ............................................................................................................................................................... 411
FDP.6.3 Statistical Facility Uptime .............................................................................................................................................................. 412
FDP.7 Model Output ............................................................................................................................. 413
System Nodal Analysis Tool ............................................................................................................ 415
NOD.1 Node-Based Analysis ................................................................................................................. 417
NOD.2 System Parameters.................................................................................................................... 421
NOD.3 PVT Parameters ......................................................................................................................... 422
NOD.4 Reservoir Parameters ................................................................................................................ 425
NOD.5 Wellbore Parameters ................................................................................................................ 428
NOD.6 Artificial Lift Parameters ............................................................................................................ 429
NOD.7 Surface Parameters ................................................................................................................... 431
NOD.8 Wellhead Results ....................................................................................................................... 432
NOD.9 Surface Results .......................................................................................................................... 433
NOD.10 Lift Tables ................................................................................................................................ 434
NOD.11 THP/BHP Gradient Matching ................................................................................................... 436
NOD.12 THP/BHP Pressure Conversion ................................................................................................ 437
Appendix Nodal Analysis ....................................................................................................................... 439
A1 Hagedorn and Brown Correlation ......................................................................................................................................................... 439
A2 Beggs and Brill Correlation .................................................................................................................................................................... 441
A3 Orkiszewski Correlation......................................................................................................................................................................... 441
A4 Mukherjee and Brill Correlation ............................................................................................................................................................ 442
A5 Gray Correlation .................................................................................................................................................................................... 442
A5 Mechanistic Model Correlation ............................................................................................................................................................. 442
PE Recovery Factor Essentials ......................................................................................................... 444
Recovery Factor Analysis Tool ........................................................................................................ 444
RFA.1 Deterministic/Probabilistic Oil RF ............................................................................................... 445
RFA.1.1 Deterministic Oil Recovery Factor ................................................................................................................................................. 446
RFA.1.2 Probabilistic Oil Recovery Factor ................................................................................................................................................... 447
RFA.2 Deterministic Gas RF ................................................................................................................... 447
RFA.3 Unconventional Oil/Gas RF ......................................................................................................... 448
RFA.4 Reservoir Complexity Index Oil RF .............................................................................................. 450
RFA.5 Alternative North Sea Oil RF ....................................................................................................... 453
RFA.6 Artificial Neural Network (ANN) Oil RF Model............................................................................ 455
RFA.6.1 Introduction to Artificial Neural Networks (ANN).......................................................................................................................... 456
RFA.6.2 Building the ANN Oil RF Model ...................................................................................................................................................... 457
RFA.6.3 ANN Oil RF Model – Sandstone Model .......................................................................................................................................... 462
RFA.6.4 ANN Oil RF Model – Carbonate Model .......................................................................................................................................... 464
PE Asset Valuation Essentials ......................................................................................................... 465
Project Economics Analysis Tool ..................................................................................................... 465
ECO.1 Oil and Gas Prices ....................................................................................................................... 466
ECO.2 Escalation Factors ....................................................................................................................... 467
ECO.3 Discount Rate, Internal Rate of Return and Discount Year ........................................................ 468
ECO.4 Operating, Capital and Sunk Costs ............................................................................................. 469
ECO.5 Well Count and Well Profile ....................................................................................................... 471
ECO.6 Economic Results........................................................................................................................ 474
ECO.7 Corporate Economics ................................................................................................................. 475
ECO.8 Project Economic Example – Marcellus ..................................................................................... 476
viii Table of Content
Asset Economic Evaluation Tool ..................................................................................................... 480

CPE.1 CAPE Introduction ....................................................................................................................... 481
CPE.2 Fiscal Terms for E&P Projects ..................................................................................................... 482
CPE.3 Tax and Royalty Contracts (TRC) ................................................................................................. 483
CPE.4 Production Sharing Contracts (PSC) ............................................................................................ 484
CPE.5 Service Contracts (SC) ................................................................................................................. 486
CPE.6 PSC/SC Advantage versus TRC .................................................................................................... 486
CPE.7 Ring Fencing ................................................................................................................................ 486
CPE.8 Typical Project Life Cycle ............................................................................................................. 487
CPE.9 CAPE – Data Input ....................................................................................................................... 488
CPE.10 CAPE – Model Parameters ........................................................................................................ 491
CPE.11 CAPE – Production Data ............................................................................................................ 492
CPE.12 CAPE – Capital Cost ................................................................................................................... 494
CPE.13 CAPE – Operating Cost .............................................................................................................. 496
CPE.14 CAPE – Fiscal Parameters .......................................................................................................... 496
CPE.15 CAPE – Sensitivities and Analysis Results.................................................................................. 498
CPE.16 CAPE – Purchase Options .......................................................................................................... 500
General Appendix .......................................................................................................................... 502
Appendix 1 – Conversion Factors .................................................................................................... 503
Appendix 2 – Workflow Examples .................................................................................................. 504
W.1 Granite Wash Well......................................................................................................................... 504
W.1.1 GW-01 Well ...................................................................................................................................................................................... 506
W.1.2 Gas Analysis ..................................................................................................................................................................................... 508
W.1.3 Production History Match ................................................................................................................................................................ 511
W.1.4 Monte Carlo Volumetrics: Gas Volume ............................................................................................................................................ 515
W.1.5 Decline Curve Analysis ..................................................................................................................................................................... 515
W.1.6 Scoping Economics ........................................................................................................................................................................... 520
W.1.7 Material Balance Type Curve Generation ........................................................................................................................................ 524
W.2 North Sea Well Recompletion ....................................................................................................... 526
W.2.1 Decline Curve Analysis ..................................................................................................................................................................... 527
W.2.2 Scoping Economics ........................................................................................................................................................................... 531
W.3 Hydraulic Fracture Uplift ............................................................................................................... 533
W.4 FDP: Sequential Wells vs Pad Drilling ............................................................................................ 539
W.4.1 Production Type Curve ..................................................................................................................................................................... 539
W.4.2 Project Forecast ............................................................................................................................................................................... 541
W.4.3 Project Economic Comparison ......................................................................................................................................................... 544
W.5 Multi-Frac/NoFrac Gas Well .......................................................................................................... 546
W.6 Single-Well Simulation Models ...................................................................................................... 555
W.6.1 Vertical Oil Well – Drainage Area Sensitivity .................................................................................................................................... 555
W.6.2 Vertical Oil Well – Water Coning/Completion Interval Comparison................................................................................................. 558
W.6.3 Horizontal Oil Well – Lateral Length Comparison ............................................................................................................................ 562
W.6.4 Horizontal Oil Well – Fracture Spacing/Stage Size ........................................................................................................................... 571
W.7 PDA – Eagle Ford Example ............................................................................................................. 574
W.7.1 Flowing Material Balance – Eagle Ford Example .............................................................................................................................. 578
W.7.2 Analytical Simulation – Eagle Ford Example ..................................................................................................................................... 580
W.7.3 Numerical Simulation – Eagle Ford Example .................................................................................................................................... 581
W.7.4 Decline Curve Analysis – Eagle Ford Example .................................................................................................................................. 582
W.8 Asset Evaluation – 6-Well Pad ....................................................................................................... 585
W.8.1 Marcellus Historical Wells - DCA ...................................................................................................................................................... 585
W.8.2 Marcellus Pad Production – Asset Evaluation Import Parameters ................................................................................................... 587
W.8.3 Marcellus Pad Comparison – Asset Evaluation Results .................................................................................................................... 592
W.9 Hydrate Protection Requirement .................................................................................................. 593
W.9.1 Hydrate Formation Curves ............................................................................................................................................................... 593
W.9.2 Graphical Analysis ............................................................................................................................................................................ 595
Petroleum Engineering & Economics Essentials 1
Introduction
“Petroleum Engineering Essentials, Tools and Techniques to Evaluate Unconventional (and
Conventional) Wells and Reservoirs” is a how-to manual for evaluation of oil and gas wells and
reservoirs. It includes tools, theories and techniques covering Reservoir Engineering and
Production Engineering. It was originally conceived to be a handbook for evaluation of
unconventional reservoirs but quickly evolved to include conventional and some IOR/EOR
evaluation techniques as well. This book is a companion to the PE² Essentials suite of tools.
PE² Essentials is a comprehensive suite of software tools comprised of Reservoir/Production

Engineering routines and programs that have been built and used during the 30+ year history of
Eastex Petroleum Consultants Inc. The PE² Essentials suite of tools can be used for all types of
wells/reservoirs: unconventional as well as conventional, onshore as well as offshore.
Figure 1: PE² Essentials
The tools included in the current version of the program are:

2 Introduction
Basic Essentials
• Gas/Oil/Water PVT and Relative Permeability Curve Generation
• Basic Equation of State Model for Oil
• Monte Carlo Simulation for Oil and Gas In-Place and Recoverable Volumes
• Chart (plots output from all PE² Essentials tools)
Well Essentials
• THP-BHP Tubing Pressure Drop Calculations for a Gas Well
• Quick Log Analysis – Includes GIIP/OOIP Estimate
• Hydraulic Fracture Design for Horizontal Wells
• Artificial Lift Design
• Pressure Transient Analysis including Analytical Test Simulator
• Volumetric (Material Balance) Surveillance and Analysis
• Hydrate Analysis
Forecast Essentials
• Horizontal Hydraulically Fractured Well Forecasting (includes a History Matching tool)
• Basic Reservoir Simulator – General Purpose Reservoir Simulator
• Streamline WaterFlood Simulator (Leighton and Higgins streamline simulator)
• Miscible/Immiscible CO2 WAG and Waterflood Streamline Simulator
• Multi-Tank Gas Material Balance Forecast – Multi Tank and Aquifer Options
• Oil Material Balance Forecast – Aquifer Options
• Decline Curve Analysis and Production Forecast – Normalized DCA
• Monte Carlo Decline Curve Production Forecasting (Probabilistic Forecast)
IOR/EOR/Heavy Oil Essentials

• IOR/EOR Screening Tool (screens 18 different processes)
• IOR/EOR/Heavy Oil Tools (MMP, Thermal Properties, Hot/Cold Injection, Pump Rates)
Asset Management Essentials

• Production Database
• Production Data Analysis
Field Development Essentials

• Field Development Planning (Multi-well Project Planning and Forecasting)
• Nodal Analysis (Includes Wellbore. Pipeline, Artificial Lift and Compressors)
Recovery Essentials
• Recovery Factor Analysis – Monte Carlo, Unconventional, Reservoir Complexity, ANN
Asset Valuation Essentials

• Basic Project Economics including Corporate Economics
• Asset Economic Evaluation including Generic PSC
Overall, there are 29 Petroleum Engineering modules and each module can include multiple
tools. A Graphing tool and a book are also included with PE² Essentials. In total, there are more
than 50 unique tools included in the PE² Essentials suite of tools.
The book can be used as a stand-alone resource for some of the techniques but for others, the
software is required to generate the results.
The “Petroleum Engineering & Economics Essentials, Tools and Techniques to Evaluate
Unconventional (and Conventional) Wells and Reservoirs” book describes the equations and
theories incorporated into the tools and, where appropriate, the file formats for input data files.
Examples for the use of each tool are included with the software and an example PE Tools
database containing input files for all tools is included in the “PE Essentials\Example Input Files”
directory. The input/output files for the solved examples in the book are included in the “PE
Essentials\Book Examples” directory.
WorkFlow examples are included throughout the book and comprehensive, integrated examples
are included in the ‘WorkFlow Examples’ document.
All PE² Essential tools are capable of handling metric or oilfield units. It should be noted that
unless otherwise specified, the metric routines use kPa rather than bars for pressure (1 bar = 100
kPa). For simplicity, this book presents oilfield units for all equations. An example of a metric
problem is included in the workflow examples. All tools include an example metric input file in
the example PE Tools database.
The software requires 64bit Microsoft Windows and can include a free trial period, by request.
An authorization code is required to unlock the program (click ‘Program Info’ on the main screen).
The PVT tool is always available.
Extract and run ‘PE Essentials Setup.exe’ and store the files to a hard drive and directory of your
choice then run the program ‘PE Essentials.exe’. It should be noted that the programs are not
stand-alone routines and require the main PE Essentials.exe program to execute them. A pre-
installed setup is available on request.
In terms of importing production data, data is imported and stored in the Production Database
from an Excel spreadsheet. All production data should be imported into the ‘PE Essentials
Database’ tool. The ‘PE Essentials Production Data Analysis’ tool is then used to extract the data
for the specific well or wells and is used to generate the data input files for the other tools.
The ‘Gas/Oil/Water PVT / Rel Perm’ tool is always available for use. To use the other tools, an
authorization code can be requested by sending the program s/n (‘Program Info’ > ‘Serial
Number’) and a request for a demo to PEEssentials@eastexpetroleum.com (Figure 2). After the
free trial period, only the ‘Gas/Oil/Water PVT / Rel Perm’ tool will be available for use.
4 Introduction
Figure 2: PE² Essentials – Authorizing the Program
It should be noted that although there is a rudimentary numerical reservoir simulator included
with PE² Essentials, the tool can export files that can be used with industry standard simulators.
The author tends to use an open-source simulator called OPM-Flow (https://opm-project.org/)
along with an open-source visualization program called ResInsight (https://resinsight.org/).
OPM Flow is a reservoir simulator for three-phase black-oil problems using a fully-implicit
formulation. There are also specialized variants for solvent and polymer problems. Flow can read
and write standard industry formats.
ResInsight is the professional quality, open source 3D visualization, curve plotting and post-
processing tool for reservoir models and simulations. The program can read industry standard
simulator input and output files.
PE Asset Management Essentials

The Asset Management Essentials section contains the following:
• Production Database
• Production Data Analysis
Production Database Tool

The PE² Essentials ‘Database’ tool (Figure PDB-1) enables input of production, injection and other
data into an SQL database for use in the PE² Essentials Production Data Analysis tool (PDA). The
data in the database is imported into the PDA tool for editing and storing in the PE Tools database
for use in the other PE² Essentials tools. The Production Database tool can also be used to
perform general comparisons (Figure PDA-2) of multi-well performance characteristics.
Figure PDB-1: PE² Essentials Database Tool
Note that prior versions of the production database will be automatically updated to the current
database format. Following the database upgrade, the database cannot be read by previous
versions of the PE Essentials Database Tool.
6 Production Database Tool
Make sure a copy of the old database is saved prior to loading it into the new database tool.
Figure PDB-2: Plotting of Database Data – Multi-Well Plot
The summation of wells can also be performed in the Database tool (Figure PDB-3). In order to
sum wells/assets together, the times must be comparable: start time and increment. If this is not
true, then a message will be displayed. The ‘Import Data’ tab can be used to confirm that the
time scales are compatible.
After summing the data, the new data set can be saved to the database by selecting/unselecting
the appropriate wells and clicking “Save Selected Wells to Database”.
All plots can be saved to a png file by clicking “Save Graph” on the appropriate tab.
Figure PDB-3: Summing Wells in Production Database
PDB.1 Data Import

To import data (Figure PDB-4), a new database has to be created (“New Database”) or an existing
database must be opened (“Open Database”).
After linking an Excel file to the tool (“Link Excel”), data is imported into the database from the
spreadsheet. All data is imported in basic volumetric units - bbls or m3 for oil and water and mscf
or 103sm3 for gas. The data interval along with the specified input time parameter (Daily Data /
Monthly Data / Yearly Data) is used to calculate rates. By specifying the input time parameter, it
is possible to have missing data but still calculate the proper rate.
Prior to importing data, ensure that the “Units” and “Fluid Type” and “Well Type” are specified.
Although it is possible to include oil wells, gas wells, injection wells, oilfield and metric units in
the same database, it is better to load wells with consistent parameters at one time.
If only a limited subset of the data is to be imported, the “End Row” box should be used to specify
the end of the data import.
Figure PDB-4: PE² Essentials Database - Importing Data
If the Excel file contains multiple wells sequentially, it is possible to enter all well data into the
database at once by ensuring there is a “Well Name” column included in the input stream. The
Database tool will import data by examining the well name and the time to determine when a
new well is to be imported. If only one well is being imported, leave the “Well Name” column box
blank. Import of a well is complete either when the well name changes or when the next time
step is blank or reduces.
Up to 5 user specified parameters as well as time-based comments can be imported for each
well. The 5 user parameters can be plotted in the Database tool for each well (Figure PDA.2) but
they cannot be imported to the other PE² Essentials tools.
Once the data is imported into the database, or a database is opened, there are a couple options
available to view the data. Plots of the data can be generated (Figure PDA-2) or the data itself
can be presented in table format (Figure PDB-5). Note: ‘Primary Data - 2’ lists the injection data.
Figure PDB-5: PE² Essentials Database – Data Table List
To view a specific well, the drop-down menu is used to choose a well from the database. It is
possible to view the imported primary data, the imported user-defined date, or the calculated
variable generated with the imported data by choosing the appropriate radio button. It is also
possible to list the cumulative volumes rather than the production volumes. Note – that rates
can be viewed through the plotting options or by importing the well data into the PDA tool.
The units displayed in the data listing is selected by choosing the appropriate radio button in the
“Reporting Units” group box.
It should be noted that the data is automatically stored in the SQL as soon as it has been
imported. It is possible to modify or delete wells by editing the database.
PDB.2 Database Editing

The data contained in the database can be edited as shown in Figure PDB-6.
Figure PDB-6: PE² Essentials Database - Editing Data
The options available for editing are to delete a well, change the well name, append data to a
specified well and reloading multi-well data. Reloading well data can also be used to append new
data to the well – in this case the original production data is reloaded as well. The advantage of
using ‘Reload Well Data’ to append data is that multiple wells can be updated at the same time.
It is possible to edit a specific data entry in a specified well by selecting the well and data row and
clicking ‘Edit Table Data’.
Prior to deleting a well, the user will have to confirm that the specified well is to be deleted
(Figure PDB-7).
Figure PDB-7: PE² Essentials Database – Delete Well

To change a well name, a screen will pop-up for entering the new name (Figure PDB-8).
Figure PDB-8: PE² Essentials Database – Change Well Name
After changing a well name, it is possible to confirm the change by clicking “List Database Wells”
(Figure PDB-9). The list will also include other parameters for the wells.
Figure PDB-9: PE² Essentials Database – List Wells

To append data to a well in the database, enter the Excel parameters and click “Append Data”.
Confirmation is required before the import is started (Figure PDB-10). You can only one well at a
time with this option.
Figure PDB-10: PE² Essentials Database – Append Data
To reload multiple well’s production data and/or append data to multiple wells at the same time,
use the ‘Reload Well Data’ option. This is useful if data was loaded incorrectly, for instance, if
fluid type was specified as oil when gas should have been specified. This option can be used to
reload the well data with the proper options.
It should be noted that only wells that currently exist in the database will be reloaded. To import
a new well, the ‘Import Data / Save to Database’ option should be used.
To edit data for a specific date, highlight the target row and click “Edit Data”. An edit window will
pop-up to allow modification of the data for the indicated date
Figure PDB-11 shows a production well and Figure PDB-12 shows an injection well.
Figure PDB-11: PE² Essentials Database – Production Well Data Edit

Figure PDB-12: PE² Essentials Database – Injection Well Data Edit
Once the appropriate data is entered, clicking “Save” will update the data in the database.
Note – it is strongly recommended that the PE Production Database be backed up on a regular

basis. Click ‘Backup dB’ on the main menu and a dated copy of the Production Database will be
placed in the “PEE Production Database Backup” directory.
14 Production Data Analysis Tool
Production Data Analysis Tool

The PE² Essentials Production Data Analysis (PDA) tool is a multi-analysis tool (Figure PDA-1).
Figure PDA-1: PE² Essentials Production Data Analysis Tool
The tool is comprised of a number of different modules:

• Production Data: Data import and editing
• PVT: Input of PVT parameters
• Data Validation/Diagnostics: Multi-well data comparison
• Flow Regime Identification: includes superposition
• Flowing Material Balance Analysis: includes flowing P/Z and PI
• Analytical Simulator: Includes numerous analytical models
• Numerical Simulator: Build models for the PE² Essentials Simulators
• PE Tools Database management tools
• INTERPRET-PDA and INTERPRET-PDA WI tools
The PDA tool is a very versatile multi-well analysis tool that builds and maintains the PE Tools
database which is used to integrate all the PE2 Essentials tools. Refer to Section PDA.13 for
information on managing the PE Tools database.
PDA.1 Reservoir Data Entry

The reservoir data is entered on the main PDA screen (Figure PDA-2).
Figure PDA-2: PDA – Reservoir Data
It will be necessary to enter the reservoir parameters before any analysis is performed. This is
especially true for a gas well since all analysis is performed using pseudo pressures.
An estimate of the initial in place volume needs to be input in order to allow iterations to initialize
properly. If a value is not entered, it will be requested when it is needed.
It should be noted that for single well analysis, the well to be analysed is chosen from the drop-
down menu (Figure PDA-3).
Figure PDA-3: PDA – Choosing a Well
When THP, CHP or BHP is available, the pressure to use for analysis is chosen under “ΔP/AOF
Parameters” (Figure PDA-4). The initial pressure must be entered if THP or CHP is chosen.
Figure PDA-4: PDA – Choosing Pressure Source for analysis
All wells can have unique reservoir parameters. If all wells are similar, reservoir parameters can
be copied to all wells in the database by clicking the “Copy Res Parameters to all Wells” button.
PDA.2 Production Data Import

The Production Data tab is used to import data into the PDA tool (Figure PDA-5).
Figure PDA-5: PDA – Production Data Import
Production data is imported into the PDA tool by linking to a PE2 Production Database, ‘Link to
PE Production dBase’, selecting the wells to be imported, then clicking ‘Import Wells From PE
Production dBase’. Additional wells can be loaded at anytime.
To add wells to an existing PE Tools database, first load the database (‘Load PE Tools dBase’) then
‘Link to PE Production dBase’ and import the additional wells. Note that there is a limit of 175
wells that can be imported into the PDA tool.
When well data is imported into PDA, it is not stored in the PE Tools database until the ‘Add Data
to PE Tools dBase’ is clicked.
It should also be noted that if the well being imported does not contain production data (i.e
injection well) then that well will not be imported into PDA.
To generate BHP, push the ‘Generate BHP’ button. This button is visible only if THP or CHP has
been imported. To convert surface pressures to bottom hole pressures it is necessary to load a
THP-BHP well model from the database by clicking the ‘Load Well Model’ button.
The BHP calculation sheet will allow use of either THP or CHP if both data sets are available and
will default to CHP when both are available. If one set of surface pressure is not available, that
column will show “n/a” instead of a check box.
Note – when using CHP for conversion to BHP for an oil well, it is assumed that that the annulus
is filled with gas.
After BHP is calculated, the option to calculate BHP will be disabled for that well (Figures PDA-6a
and PDA-6b). This allows a visual determination of which wells require a BHP calculation.
Figure PDA-6a: PDA – BHP Generation Figure PDA-6b: PDA – BHP Calculation Disabled
To re-calculate BHP, click the reset BHP button on the Production Data page (Figure PDA-5). BHP
data will be cleared and the BHP calculation for the specified well or wells will be re-enabled.
The well data in the database can be updated by clicking ‘Update Database’. A screen will open
so the specific wells can be selected (Figure PDA-6c). TO discard changes, do not update the well
in the database.
Figure PDA-6c: PDA – Update Well in PE Tools Database
When entering production data, two sets of data are stored in the database: the original data set
and an analyzable data set that has all zero production rates removed from the dataset. By
checking the “Check=Raw Import Data / UnCheck=Analyzable Data” box, the table will cycle
between the raw data and the edited data.
To view the cum data rather than the rate data, choose “Show Cum”.
PDA.3 PVT Data Entry

The PVT sheet is used to import fluid and rock data for each well (Figure PDA-7).
Figure PDA-7: PDA – PVT
PVT data input is straightforward and is required for use in the analysis routines. There are no
defaults for the parameters so all parameters should be entered.
If more than one well has been entered into the database, the entered PVT data can be copied
to all wells by clicking the “Copy PVT Properties to all Wells” button.
It should be noted that, for a gas well, the oil properties represent the condensate properties. If
Condensate/Gas Ratio is 0, then input of the condensate properties is not required.
The PVT data can be imported from the PE Tools database (Figure PDA-8).
Figure PDA-8: PDA – PVT Import
PVT data from another PDA well or from a PVT Model stored in the database can be imported.
To import the data from a PVT model, select the button and a list of the available PVT Models
will be listed.
PDA.4 Oil and Gas Pseudo Pressure

The PE² Essentials Production Data Analysis tool performs gas analysis using gas pseudo pressure.
For oil analysis, there is an option to use pressure or oil pseudo pressure.
PDA.4.1 Gas Pseudo Pressure

For gas, most of the PVT properties can vary significantly with pressure. To compensate for
changing gas properties, the concept of pseudo-pressure (ψ) was developed by Al-Hussainy et al.
(1966) and is defined as follows:
p
p
g ( p ) = 2  2 dp
p0
Z
This transformation of pressure to pseudo-pressure is an exact transformation, which accounts

for variation of gas density and viscosity. Assuming that changes of total compressibility (ct),
porosity, and fluid saturations are insignificant, the gas flow equation can be written in a form
that is similar to the liquid equation. Therefore, the liquid-flow solution can be used for gas well
analysis provided pressure is replaced by pseudo-pressure.
PDA.4.2 Oil Pseudo Pressure

The Flowing Material Balance (FMB) equation presented in Section PDA.8 is valid under the
assumption that the variation of oil and rock properties (ct, Bo, μo, and k) with pressure is
negligible. This assumption may be valid for most cases, but sometimes (e.g. for oil with high gas
content, or for cases with pressure-dependent rock properties), it is important to accurately
account for variations in oil and rock properties (Refer to Stalgorova, E. and Mattar, L. “Analytical
Methods for Single-phase Oil Flow: Accounting for Changing Liquid and Rock Properties.” SPE
180139).
Oil pseudo pressure is defined as follows:
Boi  oi
p
k ( p)
o ( p) =
ki 
p0
Bo ( p)  o ( p)
dp
For use in PE² Essentials PDA, it is assumed that permeability is constant, k(p)/ki = 1, so only fluid
properties are used in the oil pseudo pressure formulation.
PDA.5 Data Validation / Diagnostics

The Data Validation / Diagnostic sheet is used to review and edit the data (Figure PDA-9).
Figure PDA-9: PDA – Data Validation /Diagnostics

Any spikes in the data can be removed using options on this tab. Note that all zero rates are
removed when the well data is loaded. Spikes can be removed using the “Data Editing” option
(Figure PDA-10).
Figure PDA-10: PDA – Data Editing
Both low and high rate spikes can be removed from the analyzable data set – the original data
set is unchanged. For instance, to remove the two low rates in Figure PDA-9, enter 1100 in the
“Less Than” box and click “Remove Rates”. Figure PDA-11 shows the result. To undo the edits,
enter 0 for the high/low rates and click “Remove Rates”. The impact on the well’s cum as a result
of editing data is shown on the Production Data sheet for that well in the “Edited data, Cum
Reduction ==>” box.
Figure PDA-11: PDA – Data Editing: Removing Spikes
The Data Validation / Diagnostics tab has multi-well capability (Figure PDA-12).
To select well data to plot, check the well and click “Update Plot”. To remove a well from the
plot, uncheck the well and click “Update Plot”.
When there are less than 25 wells in the database, all wells can be selected and plotted by clicking
“Select All Wells”. When there are more than 25 wells in the database, 25 wells at a time can be
selected and viewed by clicking “Select 25 Wells”. Each subsequent click of the button will add
25 wells to the plot until all wells are selected.
Figure PDA-12: PDA – Multi-Well Plotting
PDA.5.1 Average Well

When more than one well is selected, it is possible to generate an average for all the rates by
clicking “Generate Average Well”.
The average well is generated as a simple calculation of the monthly average rate for the given
monthly time period. The average well will be plotted and is available for use as a well in the
analysis routines of PDA (Figure PDA-13).
Figure PDA-13: PDA – Average Well
To keep the average well, the database should be saved.

One note of caution, if another average well is generated using the same number of wells, then
there will be two wells in the database with the same name. To delete one of the wells, return to
the Production Data sheet, select the well to be deleted from the Base Well drop-down menu
and click “Delete Well”. To generate new names for average wells, generate the average well
using a different well count.
PDA.5.2 P90, P50 and P10 Wells

When more than one well is selected, it is possible to generate a P90, P50 and P10 well by clicking
“Generate P90/P50/P10 Wells”. These wells are not true probabilistic profiles but are based on
calculating percentiles for the well data included in the calculation (Figure PDA-14).
Figure PDA-14: PDA – P90/P50/P10 Wells
The meaning of “percentile” can be state as: the Pth percentile of a data distribution is a number
such that approximately P percent (P%) of the values in the distribution are equal to or less than
that number. So, if for example 50 is the 90th percentile of a larger distribution of numbers, 90%
of those numbers are less than or equal to 50, so it can be termed P90.
Mathematically, a percentile can be calculated directly for values that exist in the distribution or
interpolated for values that don’t exist. To calculate percentiles, the data is sorted so that x1 is
the smallest value, and xn is the largest, with n = total number of data points. Then xi will be the
Pi percentile of the data set where:
Pi = 100 * (i - 0.5) / n
Conversely, to calculate the value at a specific percentile (P90, P50, P10), the above equation is
solved for i as follows:
i = n * Pi / 100 + 0.5
If the calculated i is an integer, then the Pi value will be xi. If i is not an integer, then a linear
interpolation is used to determine the xi value for that Pi.
PDA.6 Pseudo Steady State vs Boundary Dominated Flow

The difference between pseudo steady state flow and boundary dominated flow is presented in
Figure PDA-15 (ref: Figures in Appendix B SPE Well testing book by John Lee).
Figure PDA-15: PDA – Pseudo Steady State vs Boundary Dominated Flow
Pseudo steady state is normally associated with well testing or wells operating under a rate
constraint. The well flow rate is maintained constant and the flowing pressure is allowed to fall.
Boundary dominated flow is normally associated with long term production where a well is
constrained by surface pressures (separators, pipelines, etc). In this case, the bottom hole
pressure is approximately constant, and the rate is allowed to decline. Boundary dominated flow
is the flow of interest for FMB.
In reality, although well head pressure is constant, the bottom hole pressure may not be constant
since tubing pressure drop changes as the flow rate declines. To account for this, the rate is
“normalized” in terms of ΔP and used instead of q. For oil, the normalized rate is defined as
qN=q/ΔP and for gas it is defined as qN=q/ΔΨ.
In addition, the concept of “Material Balance Time” (tmb=cum/rate) was developed to provide
the normalization necessary to make constant pressure and constant rate solutions equivalent.
Plotting production data using tmb also allows solutions with both declining rates and pressures
to look like the equivalent constant rate solution, similar to the superposition time function in
PTA but applied to boundary dominated flow in PDA.
When normalized rate (qN) is plotted against tmb on log-log scale, boundary dominant flow is
evident as negative unit slope.
PDA.7 Flow Regime Identification

The Flow Regime Identification tab is used to identify flow regimes in the data (Figure PDA-16) to
confirm, for example, that boundary dominated flow has been achieved.
Figure PDA-16: PDA – Flow Regime Identification
To assist in flow regime identification, lines with ¼ slope, ½ slope and unit slope can be placed on
the graph and moved around to the appropriate area of the plot.
As shown in Figure PDA-15, bilinear flow (¼ slope line) was evident until approximately 110 days.
Linear flow (½ slope line) occurred until approximately 320 days.
Boundary dominated flow (unit slope) was evident after 320 days of production. For this well, an
exponential Arps analysis would be valid for data after 300 days, although the LOPL eDCA model
could be used to model the entire flow period.
The DCA “base governing equation” presented in Section DCA.8.7 of the DCA Tool
documentation, can be plotted on this sheet. Figure PDA-17 shows a plot of Equation PDA.1
assuming no derivative smoothing.
1 dq(t )
D(t )  − (PDA.1)
q (t ) dt
Figure PDA-17: PDA – Plot of D-Parameter, No Smoothing
There is a lot of scatter in the data so trends in the data are not obvious. A “Mid” smoothing was
applied to the data (Figure PDA-18) using rate and cum as the derivative basis.
Figure PDA-18: PDA – Plot of D-Parameter, Mid Smoothing (Left-Rate Based, Right-Cum Based)
The plot on the left (rate based) of Figure PDA-18 now appears to exhibit a trend in the data
where the right plot (cum based) shows the same trend but it appears to be more obvious.
Note that caution should be used when smoothing derivative data since the amount of
smoothing applied can result in an overprint on the derivative resulting in false trends. The
minimum amount of smoothing required for analysis should be used.
The “Export Plot Data” button will save all the plot data to a csv file and the “Save Graph” button
will save the graph to graphics file.
PDA.7.1 Example: PDA/eDCA

Figure PDA-19 shows the production data for a gas well with 2100 producing days and the flow
regime identification plot showing that boundary dominated flow was achieved after
approximately 600 days of production.
Figure PDA-19: PDA – Example Well: Data and Flow Regime Identification
The data was imported into the DCA tool and analysed with eDCA. The data was matched to the
LOPL model and equivalent Arps parameters were generated. A forecast was then generated to
a minimum rate of 100 mscf/d (Figure PDA-20).
Figure PDA-20: PDA – Example Well: eDCA Results
The final LOPL-Arps matched was achieved with a Dlim of 0.13.

PDA.8 Flowing Material Balance Analysis

The Flowing Material Balance Analysis tab is used to perform analysis of the data to estimate
initial volumes in place (Figure PDA-21). It is only possible to perform this analysis if pressure is
included in the imported well data.
Figure PDA-21: PDA – Flowing Material Balance Analysis
Flowing material balance (FMB) analysis is a practical method for determining original
hydrocarbon volumes in-place. It has become popular because it enables performing material
balance analysis without having to shut-in wells to obtain estimates of reservoir pressure. Its
application is valid for single-phase oil and/or gas during the stabilized, or boundary dominated
flow period. For FMB of a gas well, pseudo-pressure is used to account for pressure-dependant
gas properties. For an oil well, use of pseud pressure s optional.
The basis of flowing material balance analysis is the Agarwal and Gardner FMB technique
augmented by flowing P/Z and flowing PI (for gas wells). To perform FMB, a material balance
model is required to estimate declining reservoir pressure. it is very important to load the proper
material balance model to estimate reservoir pressure. For a gas reservoir, the default straight-
line material balance model should work since the gas recovery process is normally a straight-
line P/Z process which extrapolates to initial gas in place.
For oil reservoirs, the material balance is a complex process of depletion, gas and water drives
and as a result, the default straight line will not work, and a suitable material balance model
should be imported from the PE Tools database (click ‘Load MBal Model’)
PDA.8.1 A-G Flowing Material Balance

Figure PDA-22 shows the shows the A-G flowing material balance plot which is a plot of
normalized rate versus normalized cum.
Figure PDA-22: PDA – A-G Flowing Material Balance
Agarwal et al discussed using normalized rate versus cumulative production to determine

hydrocarbons in place in “Analyzing Well Production Data Using Combined-Type-Curve and
Decline-Curve Analysis Concepts”, SPE 57916.
Flowing material balance analysis is based on common DCA plot of rate vs. cumulative production
analysis techniques. The advancement over traditional decline analysis is the pressure
normalization of both rate and cumulative production to account for variations in flowing
bottom-hole pressure.
The normalized rate versus normalized cum approach applies to both oil and gas reservoirs and
works for constant or variable rate systems. The normalized rate for both oil and gas is defined
as Equation PDA.2 and the normalized cum is defined as Equation PDA.3.
qo qg
Oil : q N = Gas : q N = (PDA.2)
pi − p wf i − wf
Bo S o N P GIIP(i − R )
Oil : CumN = Gas : CumN = (PDA.3)
ct Boi ( pi − p wf ) i − wf
Where: qN is the normalized rate, CumN is the normalized cum, Pi is the initial pressure, Pwf is the
flowing pressure, ct is the total compressibility, Ψi is the initial pseudo pressure, Ψwf is the flowing
pseudo pressure, ΨR is the average reservoir pressure at time t.
If using oil pseudo pressure, the oil CumN equation is the same as the gas CumN equation except
for OOIP. Since the OOIP/GIIP term is part of the normalized cum, it is necessary to initialize the
procedure with an initial guess of in-place volume in the reservoir section.
A straight-line section on the A-G FMB cartesian plot indicates boundary dominated flow and can
be extrapolated to the initial volume in place. To modify the straight line for analysis, move the
red dots until the analysis is completed. The Gas/Oil in Place value will be updated as the analysis
progresses.
Since ΨR is include in the normalized gas cum calculation, there needs to be a way to calculate
the average reservoir pressure over time. This is done through the use of a normalized material
balance model. The normalized MB model is used along with the extrapolated A-G FMB GIIP to
estimate PR as the reservoir is depleted.
The default gas MB model is a straight-line P/Z plot where, since the MB model is normalized,
the initial P/Z point is 1 and the final cum gas point is 1 (Figure PDA.23).
Figure PDA-23: PDA - Default Normalized Material Balance Model
If the default MB model is not acceptable, a material balance moel can be imported from the PE
Tools database. The MB model is imported by clicking the “Load MBal Model” button. The MB
model is normalized based on the pressure and in-place volumes in the MB model. The
normalized curve is then used to calculate ΨR at each time step. A normalized MB model for each
well can be entered or the same model can be copied to all wells by clicking the “Save MB Model
to all Wells” button. The MB model is also used in the Analytical Simulator (Section PDA.9).
PDA.8.2 Flowing Productivity Index

To assist with the A-G FMB analysis, a flowing PI curve can be included on the plot. Figure PDA-
24 shows the flowing material balance plot with A-G FMB and flowing PI curves.
Figure PDA-24: PDA – A-G FMB and Flowing PI Curves
The flowing PI plot is used as a qualitative indicator for the A-G FMB analysis. The flowing PI is
calculated with Equation PDA.4 and is dependant on the reservoir pressure at each timestep.
qo qg
Oil : PI o = Gas : PI g = (PDA.4)
p R − p wf R − wf
During boundary dominated flow, both rate and P R or ΨR will decline proportionally and the
resulting PI will be a constant, or steady-state PI. When changing the slope and location of the
straight line on the A-G FMB curve, the PI will be updated with the new P R or ΨR forecast based
on the extrapolated in-place volume. When the PI plot stabilizes, the A-G FMB results can be
considered to be valid.
PDA.8.3 Flowing PR/ZR and Flowing PR

One additional plot is available, the flowing PR/ZR (for gas) or PR (for oil) plot (Figure PDA-25).
Figure PDA-25: PDA – A-G FMB, Flowing PI and Flowing PR/ZR Curves
Flowing PR/ZR was introduced by Mattar et al in 1998 (Mattar-McNeil “The flowing Gas MB”, JCPT,
Vol37 no.2 1998). The flowing PR/ZR assumes that boundary dominated flow has been achieved
so that the declining Pwf or Ψwf is directly proportional to declining PR or ΨR (Figure PDA-26).
Equation PDA.5 and PDA.6 present the stabilized flow equation for a gas well with a circular
boundary.
1422T   re  3
R = wf + ln  −  q (PDA.5)
kh   rwa  4
R = wf + b q (PDA.6)
Where b is a constant for the specific well.
The oil equations are similar where PR = Pwf + bq.

Figure PDA-26: PDA – Flowing P/Z
When the flow rate is constant, the bq term in Equation PDA.6 is constant and based on PR/ZR
material balance:
P R Pi  G p  Pwf
= 1 −  = + bq
Z R Z i 
(PDA.7)
G  Z wf
When rate is constant the difference between static PR/ZR and flowing Pwf/Zwf is bq. For variable
rates, the flowing PI is substituted for the b term as shown in Equation PDA.8.
PR Pi  G p  Pwf q
= 1 −  = +
Z R Z i 
(PDA.8)
G  Z wf PI
The PI in Equation PDA.8 is obtained from the y-intercept of the A-G FMB plot and is used to
update the flowing PR/ZR curve.
During flowing material balance analysis, the reservoir pressure can be modified to adjust the
PR/ZR line to finalize the analysis.
The flowing material balance is complete when the flowing PI is approximately constant and the
flowing P/Z points fall along the P/Z straight line as shown in Figure PDA-25.
It should be noted that only the A-G FMB line can be directly modified.
PDA.9 Analytical Simulator

The PE² Essentials Production Data Analysis tool includes an analytical simulator that can be used
to confirm the FMB results and estimate reservoir properties by history matching the production
data (Figure PDA-27).
Figure PDA-27: PDA – Analytical Simulator
The analytical simulator included in the PE² Essentials Production Data Analysis tool simulates
well pressures based on the “Method of Images”. Imaginary wells, referred to as image wells, are
used to generate the pressure effect of the reservoir discontinuity.
There are nine different reservoir models included in the PDA analytical simulator (Figure PDA-
28). After choosing a model, the required input will be requested. All wells have unique reservoir
models.
Figure PDA-28: PDA – Analytical Simulator Models
Up to four skin factors can be entered to for matching the production history (Figure PDA-29).
Figure PDA-29: PDA – Analytical Simulator Skin Factors
Checking the “Plot FMB Reservoir Pressure” will plot the PR calculated from the flowing P/Z
analysis. To modify these values, the flowing material balance analysis would have to be
performed again with the new Pi or GIIP values.
If historical pressure is not available in the data set, the analytical simulator can be used to predict
the pressure.
Following simulation, the results can be saved to a csv file by clicking on the “Save Simulation
Results” button.
PDA.10 Numerical Simulator

The PE² Essentials Production Data Analysis tool includes the option to build a data file for the
PE² Essentials Unconventional Forecast tool and the PE² Essentials Basic Reservoir Simulator tool
(Figure PDA-30).
The simulation models are built based on the reservoir parameters entered into the PDA tool. As
a quality check, the initial volumes in place will be reported on the Numerical Analysis page so
the dimensions of the simulation model can be confirmed to yield the in-place volumes calculated
with PDA. The area of the simulation model will also be reported.
The option is included to build a horizontal fractured well or a vertical well simulator model.
If building a PE² Essentials Unconventional Forecast model, there is an option to include the
production history data so that this tool can be used to refine the history match generated with
the PDA – analytical simulator.
It is possible to build a vertical, horizontal, or a single frac stage single-well model for the PE²
Essentials Basic Reservoir Simulator. The single frac stage model can be used as the basis to build
a forecast for a well with multiple hydraulic fractures.
Figure PDA-30: PDA – Numerical Simulator
Building models for the “Industry Simulator’ will build a file that can be read by Eclipse-
compatible simulators like OPM Flow.
PDA.11 INTERPRET-PDA
INTERPRET-PDA enables straight-line analysis of any plot to be performed (Figure PDA-31).
It is possible to include up to three straight lines on any plot. The straight-line data can be saved
to the PE Tools database. Individual lines can be removed by selecting the line and clicking the
‘Clear’ button. It should be noted That the straight-line info is stored on a plot basis not a well
basis. This is so multi-wells can be evaluated with a straight line based on the plotted parameter.
The straight-line data for all plots can be exported to a CSV file (Table PDA-1).
PE Essentials INTERPRET-PDA Data Export
Oil Oilfield
Database: PEE Tools Examples Database.PEEdb
Info: INTERPRET-PDA Eagle Ford
# Wells analyzed: 1
First analyzed well: Eagle Ford Example
X-Parameter Y-Parameter Line 1 Line 1 Line 2 Line 2 Line 3 Line 3
Slope Intercept Slope Intercept Slope Intercept
Time (yrs) q-Oil/Î”P -0.90209 0.05906 0 0 0 0
Time (yrs) 1/D (yr) 1.028719 0.055823 0 0 0 0
Table PDA-1: INTERPRET-PDA – Exporting Straight Line Data

Figure PDA-31: INTERPRET-PDA – Straight Line Analysis
It should be noted that no analysis is performed with this tool. It can be used to observe and plot
trends between wells.
PDA.12 INTERPRET-PDA WI
The INTERPRET-PDA WI tool (Figure PDA-32) enables a Hall Plot analysis of water injection well
data. A Hall Plot evaluated steady-state flow at an injection well. In general, the slope of a Hall
plot is an indicator of the average well injectivity. Under normal conditions, the plot is a straight
line.
Hall (Ref: Hall, H.N.: How to Analyze Waterflood Injection Well Performance, World Oil, Oct.
1963) presented a technique to interpret injection well rate and pressure data to estimate near
wellbore skin effects and average injectivity performance. The data required for Hall Plot analysis
includes the following:
• Bottom-hole injection pressures
• Average reservoir pressure
• Water injection rate/volumes
Figure PDA-32: INTERPRET-PDA – Straight Line Analysis
The Hall method assumes steady-state injection such that the injection rate can be expressed as:
(PDA.9)
Where: k = permeability, h = reservoir thickness, pwi = flowing pressure, pavg = average reservoir
pressure, μ = fluid viscosity, re = reservoir effective radius, rw = wellbore radius and S = skin
Equation PDA.8 is based on the following assumptions:

• The fluid is homogenous and incompressible
• The reservoir is uniform, both in terms of permeability and thickness
• The reservoir is horizontal, and the flow is radial
• Flow is under steady-state flow conditions
• Mobility ratio is equal to 1
• The pressure at a distance equal to re is constant
If it is assumed that k, h, μ, re, rw and S are constant, Equation PDA.8 can be rewritten as PDA.10
(PDA.10)
(PDA.11)
Rearranging Equation PDA.10 yields the following:
(PDA.12)
Integrating both sides of Equation PDA.12 with respect to time yields Equation PDA.13.
(PDA.13)
The integral on the right-hand side of Equation PDA.13 is cumulative water injected so Equation
PDA.13 can be rewritten as Equation PDA.14
(PDA.14)
where: Wi is the cumulative volume of water injected at time t.
A plot Equation PDA.14 should yield a straight line with a slope of 1/C. This type of plot is called
the Hall Plot. If the parameters h, μ, re, rw, and S are constant, then from Equation PDA.11, the
value of C is a constant and the slope is a constant. If the parameters change, C and the slope of
the Hall Plot will change, which is the diagnostic value of the plot.
Changes in injection conditions can be observed from the Hall Plot. For example, if wellbore
plugging or other restrictions are gradually occurring, the net effect is a gradual increase in the
skin factor, S. As S increases, C decreases and the slope of the Hall Plot increases. Conversely, if
S decreases (for example, if injection pressure exceeds fracture pressure), then C increases and
the slope of the Hall Plot decreases. Figure PDA.33 presents generic Hall Plot signatures for
different injection conditions.
Figure PDA-33: Hall Plot Example
For practical purposes, the integral in Equation PDA.14 can be represented by Equation PDA.15
simplifying the generation of the Hall Plot.
(PDA.15)
Where: Δp = pwi - pavg , pwi is injection pressure, pavg is the average reservoir pressure (assumed to
be constant) and Δt = number of injection days.
Changes in the slope of a Hall Plot tend to occur gradually, so several months of injection history
may be needed to reach reliable conclusions about injection behaviour. It is important to note
that changes in the slope of the Hall Plot can be the result of other factors.
Typically, during the early life of an injection well, the radius of the water injection zone increases
with cumulative injection (for example from gas fill-up) and causes the value of C to increase,
resulting in a concave upward trend in the Hall Plot. Also, the Hall Plot technique assumes a
mobility ratio of 1.0. If the mobility ratio is greater than 1, then the Hall Plot gradually trends
concave downwards (as shown in curve D in Figure PDA.33); if mobility ratio is less than 1.0, it
will gradually trend concave upwards (see curve C in Figure PDA.33). Also, as the average water
saturation in the reservoir increases with time, kw may increase, which can also affect the slope
of the plot.
The overall purpose of the Hall Plot is to detect changes in the injection well skin factor. If the
skin factor is known or can be assumed, then the permeability can be estimated using Equation
PDA.11, after the slope is determined.
PDA.13 Exporting Data from the PE Tools Database

The PE² Essentials Production Data Analysis tool has an option to export the well data to a CSV
file (Figure PDA-34) by clicking ‘Export Data to CSV File’.
Figure PDA-34: PDA – General Export Options
For the data export, any number of wells can be chosen and the type of data to export is checked.
The data is exported to a csv file that can be plotted using the PE Chart tool or imported into
Excel.
A summary file will be exported along with the data for the specified well. It is possible to export
just the summary file. The summary file is saved in a csv file format so the results can be imported
into Excel. Table PDA-2 is an example of the summary output.
Well Name OOIP Pi Tr Poro k h Rw Sg So GasG H2S CO2 N2 API Pb NaCl Depth Length Width TubID #fracs xf
Eagle Ford Example 0.578376 7000 160 0.057 1.8 50 4 0 0.73 0.8 0 0 0 30 3000 35000 8700 5037 400 1.992 15 150
Table PDA-2: PDA – Well Summary Table
PDA.14 Converting Forecast Data to Production Data

It is possible to convert a production forecast stored in the PE Tools database into an equivalent
production data by clicking ‘Convert Forecast’ on the main menu (Figure PDA-35). This will copy
the chosen forecast into a production well and store it in the PE Tools database. Note the original
forecast is not impacted.
Figure PDA-35: PDA – Convert Forecast to Production
Once converted, the forecast can be analyzed as if it were a production well.
PDA.15 PE Tools Database Management

The PDA tool creates the PE Tools database which is the repository of all the input/output files
for the PE2 Essentials tools. It is separate from the PE2 Essentials Production Database to ensure
that the integrity of the raw production data is maintained.
The PDA tool is used to modify the contents of the PE Tools database. This is done by clicking the
‘Database Management’ button (Figure PDA-36).
Figure PDA-36: PE Tools Database

The ‘List Forecast’ button lists all the forecasts stored in the PE Tools database (Figure PDA-37a)
and the ‘List Models’ button lists the tool models (Figure PDA-37b).
Figure PDA-37a: PE Tools db Forecasts Figure PDA-37b: PE Tools db Models
Clicking ‘Delete Data’ enables data to be deleted from the database (Figure PDA-38).
Figure PDA-38: PE Tools Delete

Figure PDA-38 shows that ‘Delete PE Tools dB PDA Wells’ option has been selected. To delete a
well, check the well, or wells, and click ‘Delete Data’. This will immediately delete the well from
the database. This cannot be undone so caution should be exercised when deleting data.
Similar to deleting wells, the tool models and forecasts can be deleted. This allows the database
to be ‘cleaned up of old or erroneous data.
Note – it is strongly recommended that the PE Tools be backed up before performing deletes.
Click ‘Backup dB’ on the main menu and a dated copy of the PE Tools database will be placed in
the “PEE Tools Database Backup” directory.
The fourth deletion option, ‘Delete PE Tools dB Tables’, allows the user to delete individual tables
in the database (Figure PDA-39). Caution should be exercised with this option since the integrity
of the database can be compromised if the wrong table is deleted.
This option is useful if a well with the same name is accidently added to the database. The
redundant data can be found and deleted with this option.
The database table naming convention is that the first three letters indicate the tool from which
the table came from. The term ‘Model’ indicates that it is a tool model, for instance, EORModel
indicates that this table is a model input for the EOR tool.
Figure PDA-39: PE Tools Database – General File Delete

If the name includes ‘_Forecast_’ then the table contains a production forecast. For instance,
FDPDailyOil_Forecast indicates that this is a daily oil forecast from the FDP tool.
If the name contains ‘Well’, then this is the actual well data tables. There are a number of ‘Well’
tables that comprise a well so it should only be deleted with the ‘Delete PE Tools dB PDA Wells’
option. If manual deletion is required, delete all tables associated with the particular well
number.
The main menu item, ‘Save to New PE Tools dBase’, will save the current well data to a new PE
Tools database. All the wells will be saved to the new database but none of the tool models or
forecasts.
The main menu item, ‘Create New PE Tools dBase’, will create a blank PE Tools database. Once
created, the wells to be stored in the database need to be selected. Clicking ‘Add Data to PE Tools
dBase will open a screen showing the available wells. Choose the wells to be stored in the
database and save the data. This is useful when a subset of the wells is to be separated from the
main database.
‘Update Wells to Database’ will replace the data in the database with the current PDA
information. This button will not add a well to the database – this is performed by loading the
data form the Production database and then clicking the ‘Add Data to PE Tools dBase’ button.
‘Copy Database’ will make a copy of the current database and is a quick way to save the database
before making significant changes to the database.
PDA Appendix – Concepts of Production Data Analysis

This appendix provides basic introductory concepts and an inventory of the time-pressure-rate
relations used in the book and available in PE² Essentials software, as well as formulations for the
various diagnostic functions. The Appendix begins with basic definitions and finishes off with the
various diagnostic plots that are necessary for Production Data Analysis.
A.1 General Concepts

A.2 Basic Diagnostics Definitions
A.2.1 q(t)
A.2.2 qavg(t)
A.2.3 D(t) and b(t)

A.2.4 βavg (t)
A.2.5 AOF (Absolute Open Flow Potential)

A.2.6 Material Balance Time
A.3 Time Functions

Superposition time is a time function which will create a common straight line when data from
different rates are plotted on the same plot. The formulation of superposition time depends on
the flow regime being analyzed. For example, superposition radial time is found by performing
the superposition in time of the radial flow equation for each rate specified. The following
formulae are the generalized forms of superposition time for each flow regime. They can handle
any number of step changes in rate.
A.3.1 Pseudo-Time/Pseudo-Pressure
Pseudo-time is a mathematical time function that accounts for the variable compressibility (cg)
and viscosity (µg) of gas as well as the variable total porosity (ф) with respect to time and pressure.
It is often confused with pseudo-pressure (Ψ(p)). To deal with the changing gas and formation
properties the concept of pseudo-time (ta) was developed by Agarwal (1980). It should be noted
that the concept of pseudo-time is not amenable to a completely rigorous solution, as is the case
for pseudo-pressure that will make the flow equation solvable.
The mathematical definition of pseudo-time is reminiscent of the definition of pseudo-pressure.

Agarwal was focussed on pressures in the wellbore, and he defined pseudo-time in terms of the
viscosity (µg) and compressibility (cg) at the wellbore. This definition accounted for the large
change in gas compressibility (cg) that occurs at low pressures. It had little effect on late time
data, and was generally used for pressure buildups only.
In the 90’s, when the gas flow equations were being used for analyzing or forecasting data
affected by reservoir depletion, it was realized that the Agarwal definition of pseudo-time for
buildups, while adequate for transient flow, was inappropriate for boundary dominated flow
(depleting systems). Moreover, the Agarwal pseudo-time definition did not solve the problem,
because it was using a simplified version of the total system compressibility (ct).
Blasingame et al. introduced a new definition of pseudo-time to account for these depletion
effects. Instead of defining the pseudo-time transformation in terms of wellbore conditions like
Agarwal did, they defined it in terms of the average reservoir pressure. This pseudo-time is best
described as material balance pseudo-time, which is appropriate for rate transient analysis.
With the introduction of pseudo-time, the gas flow equation can be written in a manner similar
to the liquid equation. Therefore, the liquid flow solution can be used for gas well test analysis
and forecasting provided pressure is replaced by pseudo-pressure, and time is replaced by
pseudo-time as presented below.
t p
dt p
ta (t )   gi Cgi  and m( p )  2  dp
0
− −
 g ( p) C g ( p) pbase
z
Q (t )
Litterature also defines Material Balance Equivalent time as : te 
q (t )
A.3.2 PDA Straight Line and Derivative Analysis

Straight line analysis for incompressible / slightly compressible fluids are represented by the
following equations.
Derivative analysis for incompressible / slightly compressible fluids are represented by the
For gas (compressible fluids), replace pressure (p) with pseudo-pressure Ψ(p) and time (t) by
pseudo-time ta.
A.4 Flowing Material Balance (FMB)

The "flowing material balance" formulation has been cited numerous times in one form or
another (Palacio and Blasingame 1993; Mattar and McNeil 1998; and Agarwal et al. 1999).
Flowing material balance formulation is derived using the solution for the diffusivity equation
during boundary-dominated flow (or pseudo steady state flow) regime. A significant advantage
of flowing material balance type models is that they do not require static or shut-in pressures.
Contacted volume is determined from an extrapolation of a straight line drawn through the data
exhibiting boundary-dominated flow regime. The major disadvantage of material balance type
methods is that they are strictly limited to boundary-dominated flow regime. For tight gas and
shale gas wells exhibiting very long periods of transient flow regime, material balance type
methods would not be applicable.
The flowing material balance addresses this by using stabilized flowing pressures. Once flowing
pressure stabilizes, the FMB produces a trend that, when extrapolated, provides an estimate of
GIIP/OIIP without the requirement of shutting in the well.
FMB assumes the late time data is in the boundary dominated flow and therefore points towards
a GIIP/OIIP. Conversely, for data that is transient or linear, it provides an estimate of contacted /
connected GIIP/OIIP.
A.5 Specialized Analysis

Specialized analysis is functionality similar to that of FMB. The x-axis of the plot is a time function
calculated based on the model that the user chooses (either constant pressure solution or
superposition time) while the y-axis is normalized pressure. Choosing the superposition option
will account for changes in operating conditions (i.e., accounting for variable pressures).
Specialized analysis should be used in conjunction with other plots to confirm the presence of
linear flow. A non-zero, positive y-axis intercept indicates the presence of skin.
The specialized plots can be accessed in the Data Validation/Diagnostic section of the PE²
Essentials Production Data Analysis tool as the ΔP/q or ΔΨ/q plot.
A.6 Production Diagnostics Plots

58 PE Basic Essentials
PE Basic Essentials
Basic Essentials includes tools (equations and routines) that can be used in the analysis of all
conventional and unconventional reservoirs and are comprised of the following:
• Reservoir Fluid Properties

• Relative Permeability Curve Generation
• Basic EOS PVT Simulator
• Monte Carlo Simulation: Volumetrics
• PE Graph - plots output from PE² Essentials tools
The reservoir fluid properties tool is always authorized for use in PE² Essentials.
Oil/Gas/Water PVT and Rel Perm Tool

Reservoir fluids are comprised of gas, oil and water. Without exception, the properties of these
fluids at reservoir conditions or at flowing conditions are required to perform an analysis.
Reservoir conditions would be defined as the temperature and pressure of the reservoir. Flowing
conditions are dynamic but represent a flowing temperature and pressure at a defined point in
time and location – bottom hole or wellhead.
In most cases, fluid properties are generated by the use of published empirical correlations that
have been developed over time. This allows the determination of fluid properties with a
minimum of input requirements.
All the PVT data and relative permeability data can be saved to a text file with the option of saving
the data in a format that can be read by standard (Eclipse, OPM Flow, etc) reservoir simulators.
To generate fluid properties, the ‘Gas/Oil/Water PVT & Rel Perm’ tool is run as shown in Figure
PVT-1.
This tool generates gas (Section PVT.1), oil (Section PVT.3) and water/rock (Section PVT.4)
properties based on industry standard correlations as well as fluid properties based on gas
components (Section PVT.2). Tables of calculated properties can be generated and saved for use
in other programs such as reservoir simulators.
The tool can also be used to generate relative permeability tables for use in reservoir simulators
(Section PVT.6).
Either US Oilfield or Metric units can be specified (Figure PVT-2).

Figure PVT-1: PE² Essentials - Gas/Oil/Water PVT & Relative Permeability Tool
Figure PVT-2: US Oilfield/Metric Output Properties
Selecting the ‘Export Tables’ button brings up a screen with options for saving calculated PVT
property tables (Figure PVT-3). These tables are useful for input into other programs, for
60 Oil/Gas/Water PVT and Rel Perm Tool
example, PE² Essentials Chart. The data is saved in a comma delimited file format (CSV). The tool
can also save the relative permeability data in a csv file.
Figure PVT-3: PVT Table Output Properties Options
Include tables for reservoir simulation can also be exported by the tools from the relevant screens
using through the ‘Save Simulator Deck’ button.
The PE² Essentials ‘Gas/Oil/Water PVT & Rel Perm’ tool includes empirical correlations for gas,
oil and water and also includes component generation routines for gas and oil to determine gas
shrinkage, use in EOS modeling, etc.
The tool also includes routines to calculate oil, gas and water relative permeability data (Section
PVT.6). An example of the relative permeability output table in simulator format is shown in
Figure PVT-4.
Figure PVT-4: Simulator Table of Relative Permeability

The main screen (Figure PVT-1) is used to calculate single values for the oil, gas and water PVT
parameters. To calculate ranges of parameters, the ‘Generate Oil PVT Tables’; ‘Generate Gas PVT
Tables’; and ‘Generate Water/Rock PVT Tables’ buttons are clicked.
For a gas well, entering a CGR value of ‘-1’ will use the internal correlation to estimate CGR. If ‘0’
is entered, a dry gas is assumed and if a positive value for CGR is entered, then the CGR
calculations will be calibrated to the entered value at the given temperature and pressure.
PVT.1 Gas Properties

There are several standard correlations that can be used to generate gas properties. All
correlations require gas gravity, reservoir pressure and reservoir temperature as input. PE²
Essentials will also apply corrections based on the acid gas (N₂, CO₂, H₂S) concentrations.
The gas correlations incorporated into PE² Essentials, as well as their range of validity, are
presented in the following sections.
Clicking the ‘Generate Gas PVT Tables’ button on the main screen will bring up the screen to
calculate a table of gas PVT values (Figure PVT-5).
Figure PVT-5: Gas PVT Parameters
Once the PVT table has been generated, the data can be saved to an industry-standard reservoir
simulator file format by clicking the “Save Simulator Deck’ button.
PVT.1.1 Gas Tc and Pc

The correlations for pseudocritical temperature, Tc in ˚R; pseudocritical pressure, Pc in psia; and
for miscellaneous gases, is from Standing, M. B., Volumetric and Phase Behavior of Oil Field
Hydrocarbon Systems, 1977, and includes the Wichert-Aziz correction for acid gas (Wichert, E.,
Aziz, K., “Calculate Z’s for Sour Gases”, Hydrocarbon Proceedings, May 1972).
yn2 = Input_%N2/100
yco2 = Input_%CO2/100
yh2s = Input_%H2S/100
Corr_SG = (SG - 0.9672yn2 - 1.5195yco2 - 1.1765yh2s) (PVT-1)
(1 - yn2 - yco2 - yh2s)
Tc’ = 168 + 325Corr_SG - 12.5Corr_SG²
Pc’ = 677 + 15Corr_SG - 37.5Corr_SG²
Tc’’ = (1 - yn2 - yco2 - yh2s)Tc’ + 227.3yn2 + 547.6yco2 + 672.4yh2s
Pc’’ = (1 - yn2 - yco2 - yh2s)Pc’ + 493yn2 + 1071yco2 + 1306yh2s
Corr_wa = 120((yco2 + yh2s)0.9 - (yco2 + yh2s) 1.6) + 15(yh2s0.5 - yh2s4)
Tc = Tc’’ - Corr_wa (PVT-2)

Pc = Pc’’ Tc____________ (PVT-3)
(Tc’’ + yh2s (1 - yh2s) Corr_wa)
Range of validity: 0 ≤ %N2 < 100

0 ≤ %CO2 < 100
0 ≤ %H2S < 100
0 ≤ %N2 + %CO2 + %H2S < 100
0 ≤ %CO2 + %H2S < 80
Gas correlations use pseudocritical pressure and temperature in a form termed pseudoreduced
pressure and temperature as follows:
Ppr = P (PVT-4)
Pc
Tpr = T + 460 (PVT-5)
Tc
Where: P and T are the pressure (psia) and temperature (˚F) of interest and SG is gas specific
gravity.
PVT.1.2 Gas Z Factor

The correlation for gas Z factor is from Dranchuk, P. M., Purvis, R. A., and Robinson, D. B.,
"Computer Calculations of Natural Gas Compressibility Factors Using the Standing and Katz
Correlation", Institute of Petroleum Technical Series, No. IP 74-008, 1974.
a = 0.064225
b = 0.535308Tpr - 0.612320
c = 0.315062Tpr - 1.04671 - 0.578327/Tpr2
d = Tpr
e = 0.681570/Tpr2
f = 0.684465
g = 0.27Ppr
ρ = 0.27Ppr 'Initial guess
Tpr
ρold = ρ
Iterate on ρ until Abs((ρ - ρold)/ρ) < 0.00001
fρ = aρ6 + bρ3 + cρ2 + dρ + eρ3(1 + fρ2)exp(-fρ2) - g
dfρ = 6aρ5 + 3bρ2 + 2cρ + d + eρ2(3 + fρ2(3 - 2fρ2))exp(-fρ2)
ρ = ρ - fρ
dfρ
Z = 0.27Ppr
(PVT-6)
ρTpr
Range of validity: 1.05 < Tpr < 3.0

0 < Ppr < 30
PVT.1.3 Gas Viscosity

The correlation for gas viscosity is from Lee, A. L., Gonzalez, M. H., and Eakin, B. E., "The Viscosity
of Natural Gases", Journal of Petroleum Technology, August, 1966.
MW = 28.97SG (PVT-7)
a = 10000 (9.379 + 0.01607MW)(T + 460)1.5
(209.2 + 19.26MW + (T + 460))
b = 3.448 + 986.4 / (T + 460) + 0.01009MW

c = 2.447 - 0.2224b
ρ= P MW______ (PVT-8)
669.8 Z (T + 460)
µg = a exp(bρc) (PVT-9)
Range of validity: 40 < T < 460 ˚F
14.7 < P < 10,000 psi
Where: µg is in centipoise (cp) and P and T are the pressure (psia) and temperature (˚F) of interest
and SG is gas specific gravity.
PVT.1.4 Gas Isothermal Compressibility

The correlation for gas isothermal compressibility is from Meehan, D. N., and Lyons, W. K.,
“Programmable Calculations for Gas Compressibility”, Oil and Gas Journal, Oct. 8, 1979.
T’ = 1 / Tpr
alpha = 0.06125T’exp(-1.2(1 – T’)2)
y = alphaPpr
Z
dfdy1 = 1 + 4y + 4y2 - 4y3 + y4
(1 - y)4
dfdy2 = y(29.52T’ - 19.52T’2 + 9.16T’3)
dfdy3 = (2.18 + 2.82T’)(90.7T’ - 242.2T’2 + 42.4T’3)y(1.18 + 2.82*T’)
dfdy = dfdy1 - dfdy1 + dfdy1
dzdp = alpha (1 _ alphaPpr)
Ppr (y y2dfdy )
cg = 1 - dzdp (PVT-10)
P Z
Range of validity: 1.05 < Tpr < 3.0

0 < Ppr < 30
Where: cg is in psi-1 and P is the pressure (psia) of interest.

PVT.1.5 Gas Formation Volume Factor

When dealing with oil reservoirs the gas formation volume factor is more commonly denoted as
Bg and has the units of rbbl/scf. For gas reservoirs the gas formation volume factor is presented
as a gas expansion factor Eg , or 1/Bg, and has the units scf/rcf (5.615/Bg).
Bg = 0.15879Z(T + 460) (PVT-11)

P
Where: Bg is in rbbl/scf and P and T are the pressure (psia) and temperature (˚F) of interest.
PVT.1.6 Condensate-Gas Ratio

The following is an empirical technique present by Meehan and Vogel for calculating the CGR
behaviour of volumetric gas-condensate reservoirs (Meehan, D. N., and Vogel, E. L, HP-41
Reservoir Engineering Manual, PennWell Books, 1982).
The technique requires the %mole of C4+ and C5+ in the reservoir fluid. If these values are not
available, they can be estimated from the surface CGR as follows.
Rd = 1000/CGR
C4+ = 6.547 + 25.52SGres + 30.38/Rd + 0.02633Rd - 30.3OilSG - 0.00417T (PVT-12)
C5+ = -8.53 + 7.83SGres + 56.26/Rd + 0.0109Rd + 0.07286API - 0.00424T (PVT-13)
Where: CGR is the surface condensate-gas-ratio (bbls/mmscf) for pressure greater than or equal
to the dew point pressure, OilSG is the surface specific gravity of oil (141.5/[131.5 + ˚API]), SGres
is the surface specific gravity of the reservoir gas, and T is the reservoir temperature in ˚F.
If dew point pressure of the gas is unknown, the initial reservoir pressure can be used as an
estimate. If a gas composition analysis was performed to estimate dew point pressure, the C4+
and C5+ and dew point pressure from that analysis should be used for the CGR forecast. CGRp, for
a given pressure, is forecast as follows.
Rd = 1000/CGR
R50% = exp[31.49 - 0.0001085Pd - 92.03C4+/T + 110.8C5+/T
+ 0.0215T + 6.833SGres - 26.98/Rd - 6.632ln(T)]
Log(R) = 2(1 - P/Pd) log(R50%/Rd) + log(Rd)
CGRp = 1000/R (PVT-14)
Where: CGR is the surface condensate-gas-ratio (bbls/mmscf) for pressure greater than or equal
to the dew point pressure, CGRp is the condensate-gas-ratio in bbls/mmscf at pressure P (psi), Pd
is dew point pressure in psia, SGres is gas specific gravity and T is the reservoir temperature in ˚F.
When using this technique, a number of points need to be considered:

- For P < 0.3Pd, R should be set to the value at P = 0.3Pd

- For P >= Pd, R = Rd
- If R50% > Rd, no retrograde behaviour, R = Rd
When the initial CGR is unknown, a method to predict CGR for any pressure was presented in
2007 (Ovalle, A.P., Lenn, C.P., and McCain Jr, W.D.; “Tools to Manage Gas/Condensate Reservoirs;
Novel Fluid-Property Correlations on the Basis of Commonly Available Field Data”, SPE112977,
SPE Reservoir Evaluation & Engineering, December, 2007).
This method is used in the PVT tool to predict CGRp, at, and below, the dew point pressure.
CGRp = 3.684 + 0.61967 Zt + 0.015359 Zt2 (PVT-15)
Z1 = 20.809 - 6.7095Ln(P) + 0.5136(Ln(P))2

Z2 = 11.175 - 1.2965 API + 0.042311 API2 - 0.0005438 API3 + 2.4889e-6 API4
Z3 = -13.365 + 27.652 SGres - 18.598 SGres 2 + 4.3658 SGres 3
Z4 = -1.5309 + 0.0058453 T + 1.4035e-6 T2
Zt = Z1 + Z2 + Z3 + Z4
Where: CGRp is the condensate-gas-ratio in bbls/mmscf at pressure P (psi), Z are the correlation
parameter equations, SGres is gas specific gravity and T is the reservoir temperature in ˚F.
PVT.1.7 Hydrate Formation Temperature

Gas hydrates are a recurring problem in the gas industry since they can plug tubing, production
and transmission lines. Numerous empirical correlations are available to predict the hydrate
forming temperature at various pressures. The most accurate correlation appears to be Omole
et al correlation published in 2009 (Omole, O., Falode, O.A., and Arinkoola, A.O.; Development of
empirical correlations for predicting formation of gas hydrate, International Journal Oil, Gas and
Coal Technology, Vol. 2, No. 1, 2009). This correlation is like other correlations but includes the
water vapour pressure which appears to enhance the accuracy of the correlation.
The first component of the correlation is the generation of the water vapour pressure for the
given reservoir temperature. The Buck correlation (Buck,,A.,L.; "New equations for computing
vapor pressure and enhancement factor”, Journal of Applied Meteorology, 20: 1527–1532)
appears to generate accurate values over the highest range of temperatures. It is presented as
Equation PVT-16.
Pvw = 0.088648 [exp(18.678 – 0.004264 Tc) (Tc / (257.14 + Tc)] (PVT-16)
Where: Pvw is the water vapour pressure in psi and Tc is the reservoir temperature in ˚C.
The hydrate temperature, Th in ˚F, at pressure P (psi) is estimated from Equation PVT-17.
Th = C1 + C2 log(P) + C3 (log(SG))2 + C4 SG log(Pvw) + C5 SG Pvw2 + C6 SG2 (log(P))2

(PVT-17)
+ C7 SG3 log(P)+ C8 SG4 log(P) + C9 (log(P))3 Pvw3
C1 = 37.42042
C2 = 8.920842
C3 = -43.1429
C4 = 22.62722
C5 = 143.7678
C6 = -1.90037
C7 = 12.98294
C8 = -2.86029
C9 = -0.07458
PVT.1.8 Converting SG to Recombined Reservoir Specific Gravity

In most circumstances, only wellhead/separator gas SG is available. If condensate volume is also
available, the surface SG can be converted to reservoir condition (recombined) SG using the
following equation (Craft, B. C., and Hawkins, M.F., Applied Petroleum Reservoir Engineering,
Prentice Hall, Inc. 1959).
Mo = 5954/(˚API - 8.8) (PVT-18)

OilSG = 141.5/(131.5 + ˚API) (PVT-19)
Rs = 1000000/CGR (PVT-20)
SGres = Rs SG + 4600 OilSG (PVT-21)
Rs + 133300(OilSG/Mo)
Where: CGR is condensate-gas-ratio in bbls/mmscf, API is stock tank condensate gravity, SG is

surface gas gravity and SGres is reservoir gas gravity.
PVT.2 Gas Component Generation and Analysis

The PE2 Essentials PVT tool includes the capability of generating pseudo components for oil and
gas fluids. The gas components are used to generate raw versus sales gas properties and the
pseudo oil components can be input to the EOS tool for further analysis. Components can be
manually entered or generated by the tool.
After completion of the component analysis, the final components can be saved to the PE Tools
Database for use in other tools.
PVT.2.1 Gas Pseudo Component Generation

When analysing gas fields, it may be necessary to predict shrinkage in order to determine sales
gas volumes. Shrinkage of the raw gas occurs as the heavier components of the gas become
liquid. The gas specific gravity (SG) measured at surface will always be lower than the value of
specific gravity in the reservoir. Since gas properties are a function of SG, without a bottomhole
sample it will be necessary to estimate reservoir gas SG based on the value measured at surface.
To do this, the PVT tool includes an option to work with gas components (Figure PVT-6).
The gas component generation tool is accessed through the ‘Generate Pseudo Gas Composition’
button. When the button is clicked, the input gas SG is converted to a pseudo composition and
the gas components table is populate. Entering ‘0’ for the seed allows manual entry of the
components. The raw gas component values can then be changed until the desired sales gas SG
is obtained. Calculations will not be performed until the sum of the components equals 100%.
Figure PVT-6: Gas Composition Analysis
The technique used to generate the pseudo gas components is an EPCI-developed technique and
is based on the assumption that the gas gravity predominately determines the C1 value which in
turn determines the C2 to C7+ values as follows:
C1 = (1.6254 - GasG) / 0.010783 (PVT-22)
C2 to C7+ = (96.7271 – C1) + (3.27299 - (H2S+N2+CO2)) (PVT-23)
The individual C2 through C6 values are determined through random sampling and the C7+ value
is modified to ensure the sum equals 100%. The C7+ molecular weight is defaulted to 150.
If the Seed is set to -1, new components will be generated every time the ‘Generate Gas
Composition’ button is clicked. It is possible to click the button until a desired sample is
generated. To fix the composition, copy the ‘Equiv Seed’ to the Seed box.
The equations in the following sections use the relevant property values from Table PVT-1 (GPA
Midstream Standard 2145-16, Table of Physical Properties for Hydrocarbons and Other
Compounds of Interest to the Natural Gas and Natural Gas Liquids Industries, 2017) to generate
the gas properties based on the generated pseudo components.
SG GHV GHV
Comp Mol Wt Tc (˚R) Pc (psia) SG (gas) 3
(liquid) BTU/scf MJ/m
H2S 34.08 671.58 1305.3 0.7989 1.1767 637.1 23.9
N2 44.01 227.14 492.5 0.8069 0.9672 - -
CO2 28.01 547.43 1070.0 0.8172 1.5195 - -
CH4 16.04 343.01 667.1 0.3000 0.5539 1010.0 38.0
C2H6 30.07 549.58 706.7 0.3563 1.0382 1769.7 66.0
C3H8 44.10 665.80 616.6 0.5072 1.5225 2516.1 94.0
iC4H10 58.12 734.06 526.3 0.5628 2.0068 3251.9 121.8
nC4H10 58.12 765.23 550.6 0.5842 2.0068 3262.3 121.4
iC5H12 72.15 828.63 489.9 0.6251 2.4911 4000.9 149.7
nC5H12 72.15 845.46 488.8 0.6307 2.4911 4008.7 149.3
C6H14 86.18 913.47 436.9 0.6641 2.9754 4755.9 177.6
C7H16 100.20 972.23 396.8 0.6882 3.4597 5502.6 205.4
Table PVT-1: Gas Physical Constants (GPA 2017)
A missing value in the table is the SG for the C7+ component. This value is estimated from the
input C7+ molecular weight based on a derived MW-SG correlation. The MW-SG correlation was
derived from a plot of molecular weight versus SG for components C3 to C30.
A plot of the MW-SG data is shown in Figure PVT-7. The data was fitted to a 6th order polynomial
with a resulting R2 of 0.9985.
Figure PVT-7: Molecular Weight versus Specific Gravity (C3 to C30)

The resulting correlation (Equation PVT-12) is used to estimate SG for the C7+ MW.
SGC7+ = aMW6 + bMW5 + cMW4 + dMW3 + eMW2 + fMW + g (PVT-24)
a = -1.6884x10-15
b = 2.7757x10-12
c = -1.8574x10-9
d = 6.4967x10-7
e = -1.2678x10-4
f = 1.3644x10-2
g = 1.0234x10-1
The specific gravity for the combined C5, C6 and C7+ components are used to calculate the API
value for the condensate.
Plots (Figure PVT-8) were also constructed for molecular weight versus Tc and Pc for C3 to C30.
The plots were fitted to a 3rd order polynomial and had an R2 of 0.9986 for Tc and 0.99998 for Pc.
Figure PVT-8: Molecular Weight versus Tc and Pc (C3 to C30)
The resulting correlations (Equations PVT-25 and PVT-26) are used to estimate Tc and Pc for the
C7+ MW.
Pc = -1.6967x10-5MW3 + 0.016615MW2 - 5.6594MW + 819.31 (PVT-25)

Tc = 1.9349x10-5MW3 - 0.019129MW2 + 7.3861MW + 390.78 (PVT-26)
Where: Tc is in ˚R and Pc is in psia.
PVT.2.2 Gas Shrinkage Factor

The gas component generation tool performs a Joule-Thompson (J-T) separation of the raw gas
stream to calculate the shrinkage factor. For the J-T process, it is assumed that 25% of C3, 50% of
C4 and 99.5% of C5+ will be recovered as liquid. The C5+ is termed condensate. It should be noted
that if a “deep cut” process is being used, then the C3 and C4 recovery factors used in this
calculation will be too low. The raw-to-sales gas shrinkage factor is calculated as follows.
C3Rec=0.25
C4Rec=0.5
C5Rec=0.995
Shrinkage = 100 - [C1+C2+(1-C3Rec)C3+(1-C4Rec)(iC4+nC4)+(1–C5Rec)(iC5+nC5+C6+C7+)]
Where: all component values (C1, C2, etc) are in mole percent.
PVT.2.3 Gas Liquids

Once the raw gas stream has been processed, the sales gas parameters are calculated, and the
liquid content of the gas stream is calculated. Note that all component values are in %mole.
The following equations use the data in Table PVT-1 and the calculated SG for the C7+ component
to derive the liquid recovery (in gallons per mscf) from the gas stream.
C3Liq = 0.003151C3(C3_MW/C3_SG)
C4Liq = 0.003151[iC4(iC4_MW/iC4_SG) + nC4(nC4_MW/nC4_SG)]
C5Liq = 0.003151[iC5(iC5_MW/iC5_SG) + nC5(nC5_MW/nC5_SG)]
C6Liq = 0.003151C6(C6_MW/C6_SG)
C7plusLiq = 0.003151C7+(C7plus_MW/ SGC7+)
The liquid recovery (gals/mscf) is then converted to bbls per mmscf.

C3Liq(bbls/mmscf) = 1000 C3Rec C3Liq /42 (PVT-27)
C4Liq(bbls/mmscf) = 1000 C4Rec C4Liq /42 (PVT-28)
C5+Liq(bbls/mmscf) = 1000 C5Rec(C5Liq + C6Liq + C7plusLiq)/42 (PVT-29)
Where: all component values (C3, iC4, etc) are in mole percent.
PVT.2.4 Gas GHV

The Gross Heating Value (GHV) for the sales gas is a direct summation of the %mole of the
component multiplied by the corresponding component heating value from Table PVT-1.
𝑛
(%mole(i)∗GHV(𝑖))
GHV = ∑ (PVT-30)
𝑖=0 100
Where: i represents C1 to C7 of the sales gas. The GHV for C7+ is irrelevant since the %mole of the
C7+ component in the sales gas is zero.
PVT.2.5 Gas Dew Point Pressure

Some gas reservoirs have initial temperature and pressure conditions that result in what is
termed as retrograde condensate reservoirs, sometimes called dew point reservoirs. Initially the
fluid in the reservoir is a single-phase gas but as the pressure reduces, the pressure falls below
the dew point and liquid condenses in the reservoir. The initial condensate that forms is
comprised of the denser components of the gas and the result is that the producing condensate-
gas-ratio (CGR) decreases. This will also have an effect on the production characteristics of the
well as the relative permeability to gas in the area of the wellbore is reduced because of the liquid
in the pores. A condensate “bank” may also form, further impacting gas rate.
When obtaining bottomhole samples of gas for analysis, the pressure should be above the dew
point pressure to ensure that a representative sample is obtained. To confirm that the sample is
representative of the reservoir fluid, the dew point pressure of the sample is determined. If
estimated dew point pressure is above or similar to the sampling pressure, the sample may not
be representative.
Dew point pressure is calculated using the correlation developed by Nemeth and Kennedy
(Nemeth, L. K., and Kennedy, H. T., “A Correlation of Dew Point Pressure with Fluid Composition”,
Journal of Petroleum Technology, June 1967).
Pd1 = a(H2S + CO2 + 0.2N2 + 0.4C1 + C2 + 2(C3 + iC4 + nC4) + iC5 + nC5 + C6)
Pd2 = bSGC7+ + c( C1 )
(C7+ + 0.2)
Pd3 = dTr + em + fm2 + gm3 + hn + in2 + jn3 + k
Pd = exp(Pd1 + Pd2 + Pd3) (PVT-31)

a = -0.020623054
b = 6.6259728
c = -0.0044670559
d = 0.00010448346
e = 0.032673714
f = -0.0036453277
g = 0.000074299951
h = -0.11381195
i = 0.00062476497
j = -0.0000010716866
k = 10.746622
m = 0.01 C7+ MW_C7plus
n = MW_C7plus/SGC7plus
Tr = (T + 460)
Where: Pd is the dew point pressure in psia, T is temperature in ˚F, SGC7+ is the specific gravity of
the C7+ component and all component values (C1, C2, etc) are in mole percent.
PVT.3 Oil Properties

A number of standard correlations are available for generating oil properties (Figure PVT-9). All
correlations require oil density, gas gravity, reservoir pressure and reservoir temperature and
either gas-oil-ratio or bubble point pressure as input.
Figure PVT-9: Oil PVT Properties
Once the PVT properties are generated, a simulator input file can be saved by clicking the ‘Save
Simulator Deck’ button.
PVT.3.1 Oil Bo and Pb Correlations

The bubble point pressure (Pb) and oil formation volume factor correlations (Bo), as well as their
validity ranges, available in the PVT tool are listed in Table PVT-2.
Table PVT-3 presents the Bo equation for each correlation and Table PVT-4 presents the
corresponding coefficients for the equations.
Bob Range Pb Range Temp Range Rs Range API Range GasG Range
Bob and Pb Correlation
Low High Low High Low High Low High Low High Low High
Standing (1947) 1.024 2.150 130 7000 100 258 20 1425 16.5 63.8 0.590 0.950
Vasquez and Beggs (1980) 1.028 2.226 15 6055 75 294 0 2199 15.3 59.3 0.511 1.350
Glasø (1980) 1.032 2.588 165 7142 80 280 90 2637 22.3 48.1 0.650 1.280
Al-Marhoun (1988) 1.032 1.997 20 3573 74 240 26 1602 19.4 44.6 0.750 1.370
Abdul-Majeed and Salman (1988) 1.028 2.042 75 290 0 1664 9.5 59.5 0.510 1.350
Dokla and Osman (1992) 1.216 2.493 590 4640 190 275 181 2266 28.2 40.3 0.800 1.290
Lasater (1992) 48 5780 82 272 3 2905 17.9 51.1 0.570 1.200
Macary and El-Batanoney (1992) 1.200 2.000 1200 4600 130 290 200 1200 25.0 40.0 0.700 1.000
Petrosky and Farshad (1993) 1.118 1.623 1574 6523 114 288 217 1406 16.3 45.0 0.580 0.850
Omar and Todd (1993) 1.085 1.954 790 3851 125 280 142 1440 26.6 53.2 0.612 1.320
Kartoatmodjo and Schmidt (1994) 1.007 2.144 15 6055 75 320 0 2890 14.4 58.9 0.380 1.710
Almehaideb (1997) 1.142 3.562 501 4822 190 306 128 3871 30.9 48.6 0.750 1.120
Al-Shammasi (2001) 1.020 2.916 32 7127 74 342 6 3299 6.0 63.7 0.510 3.440
Table PVT-2: Available Bob and Pb Correlations and Validity Ranges
Bo Correlation Bo Correlation Equation

Standing a1 + a2 * (RS * (GasG / SGoil) ^ a3 + a4 * t) ^ a5
Vasquez and Beggs 1 + a1 * RS + a2 * (t - 60) * (OilAPI / GasG) + a3 * RS * (t - 60) * (OilAPI / GasG)
Glasø 1 + 10 ^ (a1 + a2 * LOG(RS * (GasG / SGoil) ^ a3 + a5 * t) / LOG(10) - a3 * (LOG(RS * (GasG / SGoil) ^ a4 + a5 * t) / LOG(10))^2)
Al-Marhoun a1 + a2 * (t + 460) + a3 * (RS ^ a5 * GasG ^ a6 * SGoil ^ a7) + a4 * (RS ^ a5 * GasG ^ a6 * SGoil ^ a7) ^ 2
Dokla and Osman a1 + a2 * (t + 460) + a3 * (RS ^ a5 * GasG ^ a6 * SGoil ^ a7) + a4 * (RS ^ a5 * GasG ^ a6 * SGoil ^ a7)^2
Petrosky and Farshad a1 + a2 * (RS ^ a3 * (GasG ^ a4 / SGoil ^ a5) + a6 * t ^ a7) ^ a8
Omar and Todd a1 + a2 * (RS * (GasG / SGoil) ^ a3 + a4 * t) ^ (a6 + a7 *(OilAPI / GasG) + a8 * GasG)
Almehaideb a1 + a2 * RS * t / SGoil ^ 2
Macary and El-Batanoney (a1 + a2 * t) * EXP(a3 * RS + a4 * (SGoil / GasG))
Kartoatmodjo and Schmidt a1 + a2 * (RS ^ a3 * GasG ^ a4 / SGoil ^ a5 + a6 * t) ^ a7
Al-Shammasi 1 + a1 * (RS * (t - 60)) + a2 * (RS / SGoil) + a3 * ((t - 60) / SGoil) + a4 * (RS * GasG / SGoil)
Table PVT-3: Bo Correlation Equations
Bo Correlation a1 a2 a3 a4 a5 a6 a7 a8
Standing 0.972 0.0001472 0.5 1.25 1.175
Vasquez and Beggs (<30 API) 0.0004677 0.00001751 -1.8106E-08
Vasquez and Beggs (>30 API) 0.000467 0.000011 1.337E-09
Glasø -6.58511 2.91329 0.27683 0.526 0.968
Al-Marhoun 1 0.497069 0.00086296 0.00182594 3.181E-06 0.74239 0.323294 -1.20204
Dokla and Osman 0.0431935 0.00156667 0.00139775 3.8053E-06 0.773572 0.40402 -0.882605
Petrosky and Farshad 1.0113 7.2046E-05 0.3738 0.2914 0.6265 0.24626 0.5371 3.0936
Omar and Todd 0.972 0.0001472 0.5 1.25 1.175 1.1663 0.000762 -0.0399
Almehaideb 1.122018 0.00000141
Macary and El-Batanoney 1.0031 0.0008 0.0004 0.0006
Kartoatmodjo and Schmidt 0.98496 0.0001 0.755 0.25 1.5 0.45 1.5
Al-Shammasi 5.53E-07 0.000181 0.000449 0.000206
Table PVT-4: Bo Correlation Equation Parameters
Table PVT-5 presents the Pb equation for each correlation and Table PVT-6 presents the
Pb Correlation Pb Correlation Equation

Standing a1 * ((RS / GasG) ^ a2 * 10 ^ (t * a3 - OilAPI * a4) - a5)
Vasquez and Beggs ((a1 * RS / GasG) * 10^(-a3 * OilAPI / (460 + t)))â2
Glasø 10 ^ (a1 + a2 * LOG((RS / GasG) ^ a4 * t ^ a5 * OilAPI ^ a6) / LOG(10) - a3 * ((LOG((RS / GasG) ^ a4 * t ^ a5 * OilAPI ^ a6) / LOG(10)) ^ 2))
Al-Marhoun a1 * RS ^ a2 * GasG ^ a3 * SGoil ^ a4 * (t + 460) ^ a5
Dokla and Osman a1 * RS ^ a2 * GasG ^ a3 * SGoil ^ a4 * (t + 460) ^ a5
Petrosky and Farshad a1 * ((RS ^ a2 / GasG ^ a3) * 10 ^ (a5 * t ^ a6 - a7 * OilAPI ^ a8) - a4)
Omar and Todd a10 * ((RS / GasG) ^ (a1 + a2 * Bo + a3 * GasG + a4 * Bo ^ 2 + a5 * GasG ^ 2 + a6 / (Bo * GasG)) * 10 ^ (t * a7 - OilAPI * a8) - a9)
Almehaideb a1 + a2 * RS * SGoil / (GasG * Bo ^ a3) + a4 * t
Macary and El-Batanoney a1 * (EXP(a4 * t - a5 * OilAPI - a6 * GasG)) * (RS ^ a2 - a3)
Kartoatmodjo and Schmidt (RS / (a1 * GasG ^ a2 * 10 ^ (a3 * OilAPI / (460 + t)))) ^ a4
Al-Shammasi SGoil ^ a1 * EXP(a2 * SGoil * GasG) * (RS * (460 + t) * GasG) ^ a3
Table PVT-5: Pb Correlation Equations
Pb Correlation a1 a2 a3 a4 a5 a6 a7 a8 a9 a10
Standing 18.2 0.83 0.00091 0.0125 1.4
Vasquez and Beggs (<30 API) 27.64 0.914328 11.172
Vasquez and Beggs (>30 API) 56.18 0.84246 10.393
Glasø 1.7669 1.7447 0.30218 0.816 0.172 -0.989
Al-Marhoun 0.0053809 0.715082 -1.87784 3.14437 1.32657
Dokla and Osman 8363.86 0.724047 -1.01049 0.107991 -0.952584
Petrosky and Farshad 112.727 0.5774 0.8439 12.34 0.00004561 1.3911 0.0007916 1.541
Omar and Todd 1.4256 -0.2608 -0.4596 0.04481 0.236 -0.1077 0.00091 0.0125 1.4 18.2
Almehaideb -620.592 6.23087 1.38559 2.89868
Macary and El-Batanoney 204.257 0.51 4.7927 0.00077 0.0097 0.4003
Kartoatmodjo and Schmidt (<30 API) 0.05958 0.7972 13.1405 0.9986
Kartoatmodjo and Schmidt (>30 API) 0.0315 0.7587 11.2895 0.9143
Al-Shammasi 5.527215 -1.841408 0.783716
Table PVT-6: Pb Correlation Equation Parameters
Where: Pb is bubble point pressure (psi), RS is the solution gas-oil ratio (scf/bbl), GasG is gas
specific gravity, OilAPI is oil gravity (˚API), t is temperature (˚F), SGOil is specific gravity of oil, Bo
is oil formation volume factor at the bubble point.
PVT.3.2 Oil Viscosity Correlations

The viscosity (µobp) correlations for oil at the bubble point, as well as their validity ranges, which
are available in the PVT tool are listed in Table PVT-7.
µo-dead Range Temp Range Rs Range Pb Range API Range

µobp Correlation
Low High Low High Low High Low High Low High
Chew & Connally (1959) 0.377 50 72 292 51 3544 132 5645

Beggs & Robinson (1975) 70 295 20 2070 132 5265 16.0 58.0
Table PVT-7: Available µobp Correlations and Validity Ranges
Table PVT-8 presents the µobp equation for each correlation and Table PVT-9 presents the
µobp Correlation µobp Correlation Equation

Chew & Connally (10 ^ (RS * (a1 * RS - a2)) * uod ^ (a6 / (10 ^ (a3 * RS)) + a7 / (10 ^ (a4 * RS)) + a8 / (10 ^ (a5 * RS)))
Beggs & Robinson a1 * (RS + a5) ^ a2) * uod ^ (a3 * (RS + a6) ^ a4)
Table PVT-8: µobp Correlation Equations
µobp Correlation a1 a2 a3 a4 a5 a6 a7 a8
Chew & Connally 2.2E-07 0.00074 0.0000862 0.0011 0.00374 0.68 0.25 0.062
Beggs & Robinson 10.715 -0.515 5.44 -0.338 100 150
Table PVT-9: µobp Correlation Equation Parameters
Table PVT-10 presents the dead oil viscosity, µod, equation for the correlations which are available
in the PVT tool and Table PVT-11 presents the corresponding coefficients for the equations.
µod Correlation µod Correlation Equation

Beal (a1 + a2 / OilAPI ^ a3) * (a4 / (t + a5)) ^ (10 ^ (a6 + a7 / OilAPI))
Beggs & Robinson 10 ^ ((10 ^ (a1 + a2 * OilAPI)) * t ^ a3) - 1
Glasø a1 * t ^ a2 * (LOG(OilAPI) / LOG(10)) ^ (a3 * LOG(t) / LOG(10) + a4)
Table PVT-10: µod Correlation Equations
µobp Correlation a1 a2 a3 a4 a5 a6 a7 a8
Chew & Connally 2.2E-07 0.00074 0.0000862 0.0011 0.00374 0.68 0.25 0.062
Beggs & Robinson 10.715 -0.515 5.44 -0.338 100 150
Table PVT-11: µod Correlation Equation Parameters
Table PVT-12 presents the live oil viscosity, µoi, equation – for P > Pb - for the correlation which is
included in the PVT tool and Table PVT-13 presents the corresponding coefficients for the
equation.
µoi Correlation µoi Correlation Equation (P > Pb)

Vasquez and Beggs uobp * (P / Pb) ^ (a1 * (P ^ a2) * (10 ^ ((a3 * P) + a4)))
Table PVT-12: µoi Correlation Equations
µoi Correlation a1 a2 a3 a4
Vasquez and Beggs 2.6 1.187 -0.000039 -5
Table PVT-13: µoi Correlation Equation Parameters
Where: P is the pressure (psi), Pb is bubble point pressure (psi), RS is the solution gas-oil ratio
(scf/bbl), OilAPI is oil gravity (˚API), t is temperature (˚F).
PVT.3.3 Gas Gravity Correction for Separator Conditions

Oil property correlations assume that the surface gas specific gravity value is referenced to a
separator pressure of 114.7 psia and temperature of 60˚F. If actual separator conditions are
different, the gas specific gravity can be corrected using the following equation.
SG = SGsep[1 + 5.912x10-5API SepT Log(SepP/114.7)] (PVT-32)
Range of validity: 76 < SepT < 150 ºF

30 < SepP < 535 psia
Where: SepP is separator pressure in psi, SepT is separator temperature in ˚F, SGsep is gas specific
gravity measured at separator conditions, API is the density of the stock tank oil and SG is the
corrected specific gravity.
PVT.3.4 Solution Gas-Oil-Ratio

Knowing the bubble point pressure, the correlation for solution gas-oil-ratio, RS, of the oil as
presented by Vasquez, M., and Beggs, H. D., “Correlations for Fluid Physical Property
Predictions”, Journal of Petroleum Technology, June 1980, can be estimated. The constants to
use in the correlation are dependent on the oil density. Note if the gas-oil-ratio is calculated at
initial reservoir conditions, it is normally labeled RSI.
RS1 = exp[c API / (460 + T)]
RS = RS1(a SG Pbb) (PVT-33)
If the oil density is greater than 30˚API, then the constants are as follows:
a = 0.0178
b = 1.187
c = 23.931
For other values of API, the constants are:
a = 0.0362
b = 1.0937
c = 25.724
Range of validity:
For 15 < ˚API < 30:
0.511 < SG < 1.351
14.7 < Pb < 4,542 psia
For 30 < ˚API < 59.5:
0.530 < SG < 1.259
14.7 < Pb < 6,025 psia
Where: Pb is bubble point pressure in psia, RS is the solution gas-oil-ratio in scf/bbl, SG is the
corrected gas specific gravity, and T is temperature in ˚F.
PVT.3.5 Oil Isothermal Compressibility

When considering oil compressibility, there are separate correlations available to calculate
compressibility above the bubble point and below the bubble point. The different correlations
result in a discontinuity at the bubble point (Figure PVT-10).
Figure PVT-10: Oil Compressibility
The reason for the discontinuity is that the definition of oil compressibility below the bubble point
contains a δRS/δP term representing the change in the amount of dissolved gas in the oil. As a
result, the change in volume below the bubble point includes a liquid and a gas component which
has a significant impact on compressibility. In addition, the RS increases up to the bubble point
then becomes constant at, and above, the bubble point causing a discontinuity.
Table PVT-14 presents the isothermal compressibility, co, equations for undersaturated oil (P >
Pb) for the correlations which are available in the PVT tool and Table PVT-15 presents the
co Correlation co Correlation Equation

Vasquez and Beggs (a1 + a2 * RS + a3 *t + a4 * GasG + a5 * OilAPI) / (a6 * P)
Petrosky and Farshad a1 * RS ^ a2 * GasG ^ a3 * OilAPI ^ a4 * t ^ a5 * P ^ a6
Table PVT-14: co Correlation Equations
co Correlation a1 a2 a3 a4 a5 a6
Vasquez and Beggs -1433 5 17.2 -1180 12.61 100000
Petrosky and Farshad 1.705E-07 0.69357 0.1885 0.3272 0.6729 -0.5906
Table PVT-15: co Correlation Equation Parameters

Where: Co is in psi-1, P is the pressure in psia, Pb is the bubble point pressure in psia, RS is the
solution gas-oil-ratio in scf/bbl, GasG is the gas specific gravity, OilAPI is the oil gravity in ˚API and
T is the temperature in ˚F.
The PVT Tool estimate for oil isothermal compressibility below the bubble point uses the
correlation presented by McCain, Rollens and Villena (McCain, W. D., Rollens, J. B. and Villena, A.
J., “The coefficient of Isothermal Compressibility of Black Oils at Pressures Below the Bubble
point”, SPE Formation Evaluation, September 1988), shown below.
Cob = exp[a + bln(P) + cln(Pb) + dln(T + 460) + eln(API) + fln(RS)] (PVT-34)

a = -7.573
b = -1.450
c = -0.383
d = 1.402
e = 0.256
f = 0.449
Where: Cob is in psi-1, P is the pressure in psia, Pb is the bubble point pressure in psia, RS is the
solution gas-oil-ratio in scf/bbl at the bubble point pressure, SG is the corrected gas specific
gravity, API is the oil gravity in ˚API and T is temperature in ˚F.
The range of validity for all the compressibility correlations are as follows:
111 < P < 9,485 psia
76 < SepT < 150˚F
30 < SepP < 535 psia
15.3 < ˚API < 59.5
0.511 < SG < 1.351
PVT.3.6 Pseudo Oil Components

The PE2 Essentials PVT tool includes the capability of generating pseudo components for oil fluids.
The pseudo oil components can be input to the EOS tool for further analysis. Components can be
manually entered or generated by the tool.
After completion of the component analysis, the final components can be saved to the PE Tools
Database for use in other tools.
The oil component generation tool is accessed through the ‘Generate Pseudo Oil Composition’
button. When the button is clicked, the input oil API and solution GOR are converted to a pseudo
composition and the oil components table is populate (Figure PVT-11). Entering ‘0’ for the seed
allows manual entry of the components. The oil component values can then be changed until the
desired oil properties are obtained. Calculations will not be performed until the sum of the
components equals 100%.
The technique used to generate the pseudo oil components is based on the assumption that most
conventional oil components fall within a limited range of mole concentration values: C1 – 0.4 to
0.5; C2 – 0.05 to 0.06; C3 – 0.045 to 0.055 ; iC4 – 0.007 to 0.015; nC4 – 0.02 to 0.03; iC5 – 0.007
to 0.015; nC5 – 0.01 to 0.02; C6 – 0.015 to 0.025; C7 – 0.3 to 0.4; C2 – 0.05 to 0.06.
The ranges are randomly sampled. The generated random distribution is then normalized to take
the impurities into account and the properties are then modified to match API and solution GOR.
Figure PVT-11: Gas Composition Analysis
To correct to the solution GOR (RSI) in mscf/bbl, the actual C7/C1 ratio is modified until the
expected C7/C1 ratio is obtained, where:
Expected C7/C1 Ratio = 2.4525 * (RSI/1000)2 - 4.5405 * RSI/1000 + 2.6265
The C7Plus MW and C7Plus SG are then caclulated as follows:
C7Plus MW = 237*EXP(0.137*a)
C7Plus SG = (0.0366*API2 - 8.369*API + 1115.7) / 1000
Where:
a = a1 + a2 + a3
a1 = -1.47*(Log(RSI))6 + 52.16*(Log(RSI))5 - 768.34*(Log(RSI))4 + 6017.87*(Log(RSI))3 -

26431.59*(Log(RSI))2 + 61730.18*Log(RSI) - 59893.7
a2 = 0.0000047*API5 - 0.0008*API4 + 0.0535*API3 - 1.76318*API2 + 28.78*API - 186.6162
a3 = -325.12*GasSG6 + 2142.8*GasSG5 - 5777.4*GasSG4 + 8144.1*GasSG3 - 6323.3*GasSG2 +

2562.9*GasSG - 423.9504
PVT.4 Water and Rock Properties

The correlations used to generate water properties require water salinity, reservoir pressure and
reservoir temperature as input (Figure PVT-12). The correlations incorporated into PE² Essentials
are presented in the following sections.
Figure PVT-12: Water PVT Properties
Once the PVT properties are generated, a simulator input file can be saved by clicking the ‘Save
Simulator Deck’ button.
PVT.4.1 Water Density

The density of water is a function of the salinity of the water. The CRC Handbook of Chemistry
and Physics (56th ed.), CRC Press, Ohio, 1975-1976 published properties of NACL-water (14.7 psia
and 60˚F) which has been plotted as shown in Figure PVT-13.
Figure PVT-13: Water Properties
The following correlation (Equation PVT-35) is a representation of Figure PVT-12 plot (R2=0.9998).
ρw = aSalt2 + bSalt + c (PVT-35)

SGw = ρw/62.368
Salt = ppmNACL/10000
a = 0.00224483
b = 0.41643
c = 62.368
Where: ρw is the density of water in lb/ft3 at 14.7 psia and 60˚F, ppmNACL is the salinity of the
water in ppm, and SGw is specific gravity of the water.
To correct to reservoir conditions, divide ρw by Bw (Section PVT.4.5).
PVT.4.2 Solution Gas-Water Ratio

Correlations for solution gas-water ratio only consider the solubility of methane in water. The
effect of solubility of the heavier components are two to three times less than the solubility of
methane and are considered to be outside the accuracy limits of the correlation and are normally
disregarded. The correlation included in PE² Essentials was published by McCain (McCain, W. D.,
The Properties of Petroleum Fluids, Second Edition, PennWell Books, 1990), as shown in the
following equation:
RSw = RSw’ log-1(-0.0840655 Salt T-0.285854) (PVT-36)

RSw’ = a + bP + cP2
a = a0 + a1T + a2T2 + a3T3
b = b0 + b1T + b2T2 + b3T3
c = (c0 + c1T + c2T2 + c3T3 + c4T4)x10-7
Where: a0 = 8.1539; a1=-6.12265x10-2; a2=1.91663x10-4; a3=-2.1654x10-7

b0 = 1.01021x10-2; b1=-7.44241x10-5; b2=3.05553x10-7; b3=-2.94883x10-10
c0 = -9.02505; c1=0.130237; c2=-8.53425x10-4; c3=2.34122x10-6; c4=-2.37049x10-9
Range of validity: 70 < T < 340˚F

1,000 < P < 10,000 psia
0 < Salt < 30% by weight
Where: RSw’ is the solution gas-water ratio for fresh water in scf/bbl, RSw is the solution gas-water
ratio for saltwater in scf/bbl, ppmNACL is the salinity of the water in ppm, P is pressure in psia
and T is temperature in ˚F.
PVT.4.3 Water Isothermal Compressibility (Gas Saturated)

The correlation for water isothermal compressibility only calculates values for undersaturated
water (no free gas in the water). Two correlations are available, one for temperature up to 250˚F
with pressures up to 6000 psia and one for higher temperatures and pressures. The correlations
for the lower temperature can be found in a number of references: Craft, B. C. and Hawkins, M.
F., Applied Petroleum Reservoir Engineering, Prentice-Hall, 1959; Meehan, D. N., “A Correlation
for Water Compressibility”, Petroleum Engineer, November, 1980; and Numbere, D., Brigham,
W. E., and Standing, M. B., Correlations for Physical Properties of Petroleum Reservoir Brines,
Petroleum Research Institute, Stanford University, November, 1977. The equation is:
Cw‘= (a + bT + cT2)x10-6
Dissolved gas correction: Cw‘’= Cw’(1 + 0.0089RSw)
Cw = Cw’’[(-0.052 + 2.7x10-4T – 1.14x10-6T2 + 1.121x 10-9T3)Salt0.7 + 1] (PVT-37)
a = 3.8546 - 1.34x10-4P
b = -0.01052 + 4.77x10-7P
c = 3.9267x10-5 - 8.8x10-10P
1,000 < P < 6,000 psia
Where: Cw is the isothermal compressibility of saturated salt water in psia-1, Cw‘ is the isothermal
compressibility of fresh, gas-free water in psia-1, RSw is the solution gas-water ratio in scf/bbl,
ppmNACL is the salinity of the water in ppm, P is pressure in psia and T is temperature in ˚F.
The correlation equation for high temperatures and high pressures was presented by Osif (Osif,
T. L., “The Effects of Salt, Gas, Temperature and Pressure on the Compressibility of Water”, SPE
Reservoir Engineering, February 1988), as shown below:
Cw = 1 / (aP + bppmNACL + cT + d) (PVT-38)
a = 7.033
b = 0.5415
c = -537.0
d = 403300
1,000 < P < 20,000 psia
0 < ppm < 250,000 ppm
Where: Cw is the isothermal compressibility of saturated saltwater in psia-1, ppmNACL is the
salinity of the water in ppm, P is pressure in psia and T is temperature in ˚F.
PVT.4.4 Water Viscosity

The correlation for water viscosity was presented by Meehan (Meehan, D. N., “Estimating Water
Viscosity at Reservoir Conditions”, Petroleum Engineer, July 1980), where:
µw = 0.02414 a b 10c (PVT-39)

a = 1 - 0.00187Salt0.5 + 0.00021Salt2.5 + (T0.5 - 0.0135T)(0.00276Salt - 0.000344Salt1.5)
b = 1 + 3.5x10-12P2(T - 40)
c = 247.8/[133.15 + 0.5556(T - 32)]
Where: µw is the water viscosity in cp at pressure P in psia and temperature T in ˚F, and ppmNACL
is water salinity in ppm.
PVT.4.5 Water Formation Volume Factor

The correlation for water formation volume factor for gas saturated water was presented by
Numbere, Brigham and Standing (Numbere, D., Brigham, W. E., and Standing, M. B., Correlations
for Physical Properties of Petroleum Reservoir Brines, Petroleum Research Institute, Stanford
University, November, 1977), as presented below:
Bw = d(a + bP + cP2) (PVT-40)
Salt = ppmNACL/10000 (PVT-41)

-5
a = 0.9911 + 6.35x10 T + 8.5x10 T -7 2
b = -1.093x10-6 - 3.497x10-9T + 4.57x10-12T2

c = -5x10-11 + 6.429x10-13T - 1.43x10-15T2
d = 1 + Salt[5.1x10-8P + (5.47x10-6 - 1.95x10-10P)(T - 60) - (3.23x10-8 - 8.5x10-13P)(T- 60)2]

1,000 < P < 5,000 psia
Where: Bw is the water formation volume factor in rbbl/sbbl, P is pressure in psia, T is
temperature in ˚F and ppmNACL is the water salinity in ppm.
PVT.4.6 Formation Isothermal Compressibility

There are two correlations available for formation isothermal compressibility. One was
generated from the correlation developed by Hall (Equation PVT-38) (Hall, H. N., “Compressibility
of reservoir Rocks”, Transactions of AIME, Vol. 198, 1953) and the other was published by
Newman (Equation PVT-39) (Newman, G.H., 1973, Pore-Volume Compressibility of Consolidated,
(PVT-42)
Friable, and Unconsolidated Reservoir Rocks under Hydrostatic Loading: Journal of Petroleum
Technology, Feb, 1973), where:
Cr = 1.782x10-6(Porosity)-0.438
Cr = 9.732x10-5(1 + 55.8721 * Porosity) (PVT-43)
Where: Cr is the formation isothermal compressibility in psi-1, and Porosity is the formation
porosity in decimal format.
PVT.5 Interfacial Tension

Interfacial tension between gas and liquid phases has minimal effect on two phase pressure
calculations but when using some of the pressure gradient prediction techniques, interfacial
tension is required to calculate certain dimensionless numbers used in the technique. Original
work was performed by Baker and Swerdloff to estimate gas-oil interfacial tension (Baker, O.,
and Swerdloff, W., “Finding Surface Tension of Hydrocarbon Liquids”, Oil and Gas Journal,
(January 2, 1956), and gas-water graphs were published by Hough (“Interfacial Tensions at
Reservoir Pressures and Temperatures”, Transactions of AIME, 1951).
In 2000 Abdul-Majeed and Al-Soof updated the Baker and Swerdloff correlation (Abdul-Majeed
and Abu Al-Soof, “Estimation of Gas-Oil Surface Tension”, Journal of Petroleum Science and
Engineering, Vol. 27).
Gas-Oil Interfacial Tension – 1 (Abdul-Majeed and Al-Soof)

Note- This gas-oil correlation is the one that has been incorporated into PE² Essentials.
For gas-oil interfacial tension, the calculation is a two-step process. The dead oil-gas interfacial
tension is calculated, and this value is corrected based on the value of RS. Abdul-Majeed and Al-
Soof presented two equations for the ratio of live oil surface tension to dead oil surface tension
based on whether RS was less than 280 scf/bbl or greater than or equal to 280 scf/bbl. The graph
they presented can be better represented by a single exponential curve as shown at
http://petrowiki.org/Interfacial_tension.
σog = σod σog/σod (PVT-44)
σod = (1.17013 – 0.001694T) (38.085 – 0.259API)
σog/σod = 0.056379 + 0.94365exp(-0.0038491RS)
Where: σog is the saturated gas-oil interfacial tension in dynes/cm, σod is the dead oil gas-oil
interfacial tension in dynes/cm at temperature T in ˚F, RS is the solution gas-oil ratio in scf/bbl
and API is oil gravity in ˚API.
It should be noted that σog will be zero at miscibility pressure. A value of 1 is normally used for
σog above miscibility pressure in the appropriate dimensionless equations.
Gas-Oil Interfacial Tension – 2 (Baker and Swerdloff)

For gas-oil interfacial tension, Baker and Swerdloff presented graphs relating dead oil surface
tension with temperature and API gravity. Only 68˚F and 100˚F were considered for the
correlation. To correct for the effect of solution gas, Baker and Swerdloff related the ratio of live
to dead oil surface tension with saturation pressure. These graphs can be represented by the
following equations which were presented by Beggs (Beggs, H. D., Production Optimization Using
Nodal Analysis, OGCI, 1991).
σog = σod exp(-0.00086306P) (PVT-45)
σod = σ68 - (T - 68)(σ68 - σ100)/32

σ68 = 39 - 0.2571API
σ100 = 37.5 - 0.2571API
Where: σog is the saturated gas-oil interfacial tension in dynes/cm, σod is the dead oil gas-oil
interfacial tension in dynes/cm at temperature T in ˚F, and P is pressure in psia.
The σog equation will be zero at a pressure of 3977 psia. A value of 1 should be used for σog above
3977 psia in the appropriate dimensionless equations. For temperatures below 68˚F, assume σog
= σ68 and for temperatures above 100˚F, assume σog = σ100.
Gas-Water Interfacial Tension – 1 (Jennings, H. Y. and Newman. G. H)

Note- This gas-water correlation is the one that has been incorporated into PE² Essentials.
Published data for gas-water systems do not agree. According to McCain, the most consistent
data was published by Jennings and Newman (Jennings, H. Y. and Newman. G. H. “The Effect of
Temperature and Pressure on the Interfacial Tension of Water Against Methane-Normal Decane
Mixtures”, Transactions of AIME, 1971). The correlations can be represented as follows.
σgw = A + BP + CP2
(PVT-46)
A = 79.1618 – 0.118978T
B = -5.28473x10-3 + 9.87913x10-6T
C = (2.33814 – 4.57194x10-4T – 7.52678x10-6T2)x10-7
Where: σgw is the gas-water interfacial tension in dynes/cm at temperature T in ˚F
Gas-Water Interfacial Tension – 2 (Baker and Swerdloff)
For gas-water interfacial tension, Baker and Swerdloff presented graphs for 74˚F and 280˚F.
These graphs can be represented by the following equations which were presented by Beggs
(Beggs, H. D., Production Optimization Using Nodal Analysis, OGCI, 1991)
σgw = σ74 - (T - 74)(σ74 – σ280)/206 (PVT-47)
σ74 = 75 – 1.108P0.349
σ280 = 53 - 0.1048P0.637
Where: σgw is the gas-water interfacial tension in dynes/cm at temperature T in ˚F, and pressure
P in psia. For temperatures below 74˚F, assume σgw = σ74 and for temperatures above 280˚F, assume
σgw = σ280.
PVT.6 Relative Permeability Data Generation

The letter k represents the absolute permeability of the reservoir (in md) and k o, kg and kw
represent the effective permeability to oil, gas and water. The fluid saturations, So, Sg and Sw must
also be specified to fully define the conditions for the value of effective permeability. Studies
have shown that a reservoir’s effective permeability in terms of the reservoir fluid is a function
of the saturation of that fluid and the wetting characteristics of the reservoir.
Since there are many possible values for saturation, effective permeability is normally reported
as relative permeability: kro, krg and krw.
kro = ko/k
krg = kg/k
krw = kw/k
Effective permeability ranges from zero to k so relative permeabilities range from zero to one.
0 < kro , krg , krw < 1.0
When all three phases are present in the reservoir, the sum of the relative permeabilities is
variable and less than or equal to one: kro + krg + krw < 1.0
The PE² Essentials PVT tool includes a relative permeability model (Figure PVT-14 and Figure PVT-
15) that will generate oil/gas/water relative permeability curves and tables. The figures show the
curves that result after selecting the ‘Default’ option This tool is executed by clicking the ‘Rel
Perm’ button on the main PVT screen.
Figure PVT-14: Oil/Gas/Water Relative Permeability
Figure PVT-15: Gas/Water Relative Permeability
The relative permeability data generated by the tool can be saved to a simulator input file by
clicking ‘Save Simulator Deck’.
Oil-Water System (Kro/Krw and Pcow)
For an oil-water sytem the relative permeability analytical equations use exponents as follows.
kro = krow ((1 - Sw - Sorw)/(1 - Swi - Sorw))No (PVT-48)

krw = krwe ((Sw - Swi)/(1 - Swi - Sorw))Nw (PVT-49)
Pcow = Pc(Swi) ((1 - Sw - Sorw)/(1 - Swi - Sorw ))Npo (PVT-50)
Where No is the exponent for oil, Nw is the exponent for water and Npo is the oil capilary
pressure exponent, krow is the kro at Swi, Swi is the initial water saturation, Sw is the desired water
saturation, Sorw is the residual oil saturation, krwe is the krw at Sorw, Pcow is the oil-water capilary
pressure and Pc(Swi) is the capilary pressure at Swi.
Gas-Oil System (Krog/Krg and Pcog)
For gas-oil sytem the relative permeability analytical equations are as follows.
krg = krg(Sorg) ((Sg - Sgc)/(1 - Swi - Sorg - Sgc))Ng (PVT-51)

krog = kroge ((1 - Sg - Swi - Sorg)/(1 – Swi - Sorg))Nog (PVT-52)
Pcog = Pc(Sgc) ((Sg - Sgc)/(1 - Swi - Sorg - Sgc))Npg (PVT-53)
Where Ng is the exponent for gas, Nog is the exponent for oil in gas and Npg is the gas capilary
pressure exponent, krg(Sorg) is the krg at Sorg, Sorg is is the residual oil saturation in gas, Sgc is the
critical gas saturation, kroge is the kro at Sorg, Swi is the initial water saturation, Sg is the desired gas
saturation, Pcog is the gas-oil capilary pressure and Pcog(Sgc) is the capilary pressure at Sgc.
Normalized water saturation is calculated as Swn = (Sw – Swi )/(1 – Swi - Sorw)
The following can be used (with caution) as a general rule of thumb.
Oil Wet:
No >= 6
Nw < 3
Krow >= 0.5
Intermediate Wet:
6 < No >= 3
5 < Nw >= 3
Water Wet:
No < 3
Nw >= 5
Krow < 0.5
90 Basic Equation of State (EOS) PVT Tool
Basic Equation of State (EOS) PVT Tool

If fluid component data is available, then the PE² Essentials ‘Basic EOS PVT – Oil’ tool (Figure EOS-
1) can be used to generate oil PVT and separator properties.
Figure EOS-1: PE² Essentials Basic EOS PVT – Oil Tool
For a complete description on solving equations of state, refer to Chapter 15 in McCain, W. D.,
The Properties of Petroleum Fluids, Second Edition, PennWell Books, 1990 and Chapter 15 in
Ahmed, T., Reservoir Engineering Handbook, Second Edition, Gulf Professional Publishing, 2001.
EOS.1 Equations of State

The PE² Essentials Basic EOS tool generates PVT properties by performing EOS analysis of the
recombined reservoir fluid. The current EOS implementation is restricted to 11 components: N₂,
CO₂, C1 to C6 and a C7+ pseudo component – hence the tool’s title of “Basic EOS”. Only the Soave-
Redlich-Kwong and Peng-Robinson EOS formulations are available in this version of the tool.
An EOS is a PVT relationship comprising an empirical cubic equation relating pressure to the
volume and temperature of the petroleum fluid. Although there are correlations available to
generate PVT properties, an EOS equation is normally required to accurately model the
volumetric and phase behaviour of petroleum fluids and to predict the performance of surface
separation facilities.
For input to EOS calculations the following component properties (Table EOS-1) are used (GPA
Midstream Standard 2145-16, Table of Physical Properties for Hydrocarbons and Other
Compounds of Interest to the Natural Gas and Natural Gas Liquids Industries, 2017).
SG Acentric
Comp Mol Wt Tc (˚R) Pc (psia) SG (gas)
(liquid) Factor
H2S 34.08 671.58 1305.3 0.7989 1.1767 0.1005
N2 44.01 227.14 492.5 0.8069 0.9672 0.0372
CO2 28.01 547.43 1070.0 0.8172 1.5195 0.2239
CH4 16.04 343.01 667.1 0.3000 0.5539 0.0114
C2H6 30.07 549.58 706.7 0.3563 1.0382 0.0995
C3H8 44.10 665.80 616.6 0.5072 1.5225 0.1521
iC4H10 58.12 734.06 526.3 0.5628 2.0068 0.1835
nC4H10 58.12 765.23 550.6 0.5842 2.0068 0.2008
iC5H12 72.15 828.63 489.9 0.6251 2.4911 0.2274
nC5H12 72.15 845.46 488.8 0.6307 2.4911 0.2515
C6H14 86.18 913.47 436.9 0.6641 2.9754 0.2986
C7H16 100.20 972.23 396.8 0.6882 3.4597 0.3494
C8H18 114.23 1023.89 360.7 0.7066 3.9440 0.3971
C9H20 128.26 1070.19 330.8 0.7222 4.4283 0.4433
C10H22 142.28 1111.86 305.0 0.7346 4.9126 0.4884
Table EOS-1: EOS Component Properties (GPA Data Tables - 2017)
In order to estimate the acentric factor, ω, of the C7+ component, a plot of Molecular Weight
versus Acentric Factor was constructed with the GPA data (Figure EOS-2).
Figure EOS-2: Molecular Weight versus Acentric Factor (GPA Data Tables - 1998)
The following correlation is a representation of the Figure EOS-2 plot (R2=0.999). Equation EOS-
1 can be used to calculate ω for the C7+ component.
ω = aMW2 + bMW + c (EOS-1)
a = -2.6179x10-6
b = 3.9927x10-3
c = -3.0363x10-2
The simplest and commonly known equation of state is the ideal gas law.
PV = RT (EOS-2)
Where V is the gas volume in ft3/mole, P is pressure in psia, T is temperature in ˚R and R is the
universal gas constant = 10.73 psi-ft3/lb-mol-˚R.
The ideal gas law was experimentally derived and is only valid for hydrocarbon gases near
atmospheric pressure, while for pressures and temperatures found in petroleum reservoirs the
physical properties calculated can lead to errors in excess of 500%. This is because real gases
(natural gases) do not behavior as an ideal gas. Basically, the magnitude of deviations of real
gases from the conditions of the ideal gas law increases with increasing pressure and
temperature and varies widely with the composition of the gas. The reason for this is that the
perfect gas law was derived under the assumption that the volume of molecules is insignificant
and molecular attraction or repulsion takes place. This is not the case for real gases.
In order to express a more exact relationship between the variables P, V, and T, a correction
factor called the gas compressibility factor (the gas deviation factor, or simply the Z-factor) is
introduced, that is equation (EOS-2) becomes PV= ZRT. This equation is valid for real gases, gases
at in situ conditions, but has limitations for condensate and oil fluids. The limitations of the real
gas law have led to a number of attempts to develop an equation of state that was valid for
describing real fluids at extended ranges of pressure and temperature. The Soave-Redlich-Kwong
(SRK) and Peng-Robinson (PR) versions of the EOS are implemented in PE² Essentials.
There are a number of components that go into an EOS formulation, one is the development of
the cubic equation itself (SRK and PR) describing properties of pure fluids, and the second is the
mixing rule used to calculate mixture parameters that are equivalent to pure substances.
All EOS equations are extensions of the Van der Waals equation of state. In 1872 Van der Waals
PhD dissertation presented an equation that attempted to remove the pressure and temperature
limitations of the ideal gas law. Van der Waals proposed equation was as follows.
P= RT _ – a_ (EOS-3)
(V - b) V2
Where V is the gas volume in ft3/mole, P is pressure in psia, T is temperature in ˚R, R is the
universal gas constant = 10.73 psi-ft3/lb-mol-˚R, and a and b are constants that characterize the
molecular properties of the components.
The a/V2 term represents a correction for the forces of attraction between the molecules which
results in a reduced pressure exerted by the real gas when compared to an ideal gas. The RT/(V-
b) term represents the force of repulsion caused by the molecules which will increase the
pressure of a real gas when compared to an ideal gas.
Expansion of Van der Waals EOS equation yields the following cubic equation.
V3 – (b + RT/P)V2 + (a/P)V – ab/P = 0
This equation has two empirical constants ‘a’ and ‘b’, and is termed a cubic equation of state.
Since Van de Waals equation was accurate only at low pressures, all subsequent EOS equation
formulations included extensions/modifications to Van de Waals equation to correct for the
pressure limitation.
The Van de Waal equation (and all EOS equations) must satisfy the following conditions at the
critical temperature and critical pressure points.
∂P/∂V = 0 and ∂2P/∂V2 = 0
Differentiating Van der Walls EOS equation with respect to volume at the critical point results in
the following values for a and b for a pure component.
a = cons1 R2Tc2/Pc (EOS-4)
b = cons2 RTc/Pc (EOS-5)
Where cons1 and cons2 are dimensionless pure component parameters and have values that are
dependent on the actual EOS formulation. For a multi-component system, a mixing rule must be
applied (Section EOS.4) to determine the ‘a’ and ‘b’ terms.
The two most common EOS equations used in the industry were developed by Redlich and Kwong
with a modification by Soave (Redlich, O. and Kwong, J. N. S., “On the Thermodynamics of
Solutions. V-An Equation of State. Fugacities of Gaseous Solutions”, Chemical Reviews, 1949; and
Soave, G., “Equilibrium Constants from a Modified Redlich-Kwong Equation of State”, Chemical
Engineering Science, 1972), and Peng and Robinson (Peng, D. and Robinson, D. B., “A New Two-
Constant Equation of State”, Industrial & Engineering Chemistry Fundamentals, 1976).
EOS.2 Soave-Redlich-Kwong (SRK) EOS

Redlich and Kwong showed that adjustment of the attractive force term would considerably
improve the prediction capabilities of Van der Waals EOS equation. The resulting Redlich-Kwong
EOS is as follows.
P= RT _ – a____
(V - b) V(V + b)T0.5
In 1972, Soave replaced the term a/T0.5 in the Redlich-Kwong equation to a more generalized
temperature dependent term, aT. The Soave-Redlich-Kwong EOS was the first modification of the
simple Redlich-Kwong EOS where the parameter ‘a’ was made temperature dependent in such a
way that the vapour pressure curve was more accurately reproduced.
P= RT _ – aT__ (EOS-6)
(V - b) V(V + b)
aT = acα
Where ac is the value of aT at the critical temperature and α is a dimensionless temperature-

dependent term that becomes 1 at the critical temperature (α = 1 when T = Tc). The term α is
defined as follows.
α = [1 + m(1 – Tpr0.5)]2 (EOS-7)
Where Tpr is the reduced temperature and the parameter m is defined as follows (Grabowski M.
S. and Daubert, T. E., “A Modified Soave Equation of State for Phase Equilibria Calculations 1.
Hydrocarbon Systems”, Industrial & Engineering Chemistry Process Design and Development,
1978, Volume 17).
m = 0.48508 + 1.5517ω – 0.15613ω2 (EOS-8)
Where ω is the acentric factor and is obtained from the GPSA tables.
Expanding the SRK equation and differentiating with respect to volume at the critical point results
in the following values for ac and b for any pure component.
ac = 0.42747 R2Tc2/Pc (EOS-9)
b = 0.08664 RTc/Pc (EOS-10)
Where b modifies the actual molar volume, V, to account for high pressure effects, and a c is a
measure of the intermolecular attractive forces.
The EOS requires three input parameters per pure compound: Tc, Pc and ω. For a multi-
component system, a mixing rule must be applied (Section EOS.4) to determine the ac and b
terms.
EOS.3 Peng-Robinson (PR) EOS

In 1976 Peng and Robinson performed an in-depth study of the capability of the SRK EOS to
predict the behaviour of hydrocarbon systems and published an improved model as follows.
P= RT _ – aT _______
(EOS-11)
(V - b) V(V + b) + b(V – b)
aT = acα
Where ac is the value of aT at the critical temperature and α, which is equivalent to the SRK term,
is a dimensionless temperature-dependent term that becomes 1 at the critical temperature (α =
1 when T = Tc). The term α was defined as Equation EOS-8.
α = [1 + m(1 – Tpr0.5)]2 (EOS-12)
Where Tpr is the reduced temperature and the parameter m is defined as follows.
m = 0.37464 + 1.54226ω – 0.26992ω2 (EOS-13)
Where ω is the acentric factor and is obtained from the GPSA tables.
Peng and Robinson recommended a modified equation for the parameter m when ω > 0.49
(heavier components) as follows.
m = 0.379642 + 1.48503ω – 0.1644ω2 + 0.016667ω3 (EOS-14)
Expanding the PR equation and differentiating with respect to volume at the critical point results
in the following values for ac and b for any pure.
ac = 0.45724 R2Tc2/Pc (EOS-15)
b = 0.07780 RTc/Pc (EOS-16)
Where b modifies the actual molar volume, V, to account for high pressure effects, and a c is a
measure of the intermolecular attractive forces.
The EOS requires three input parameters per pure compound: Tc, Pc and ω. For a multi-
component system, a mixing rule must be applied (Section EOS.4) to determine the a c and b
terms.
EOS.4 Mixing Rule and Binary Interaction Coefficients

The SRK and PR equations of state were developed for pure fluids. To extend their use to mixtures
it is necessary to apply a mixing rule. The mixing rule is simply a technique that calculates
parameters for the mixture that results in the equivalent of a pure fluid. The mixing rule applies
corrections to the aT and b terms used in the EOS formulation.
There are a number of mixing rules that have been published and they all yield slightly different
results. PE² Essentials uses the mixing rules suggested by McCain (McCain, W. D., The Properties
of Petroleum Fluids, Second Edition, PennWell Books, 1990). These mixing rules are as follows.
aT = ∑I ∑j yiyj (aTiaTj)0.5 (1 - δij) (EOS-17)
b = ∑j yjbj (EOS-18)
Where δij are the binary interaction coefficients (BIC’s) and are the result of interaction between
unlike molecules (i and j represent the component).
Values for the BIC’s are empirically derived from PVT data of binary mixtures of the compounds
of interest. BIC’s are dependent on the difference in molecular sizes of the two components. The
BIC’s have different values for each binary pair and are normally different for the different
equations of state.
The PE² Essentials EOS model includes options for three different set of BIC’s. The BIC’s and
derivation techniques were presented by Ahmed (Ahmed, T. H., Equations of State and PVT
Analysis: Applications for Improving Reservoir Modeling, Gulf Publishing Company, 2007 and in
Ahmed, T., Reservoir Engineering Handbook, Second Edition, Gulf Professional Publishing, 2001).
The PE² Essentials PR EOS includes the following set of BIC values.
Table EOS-2: Binary Interaction Coefficients for PR EOS
Note that the values for C7+ BIC’s in all PE² Essentials EOS equations are calculated using Equation
EOS-24 as described for the ALT_BIC calculations presented below.
The PE² Essentials SRK EOS includes the following set of BIC values.
Table EOS-3: Binary Interaction Coefficients for SRK EOS
There is a third option for BIC’s called ‘Alt BIC’ that allows the user to enter alternate BIC’s for
use in either EOS formulation. The Alt BIC’s are used by checking the appropriate box on the main
screen (Figure EOS-1). The Alt BIC’s are entered through the ‘ALT_BICs.Blib’ file located in the “PE
Essentials\Bin\PE Essentials Basic EOS PVT Libs” directory. To use internally calculated alternate
BIC’s, delete the Blib file.
The internal alternate BIC calculation is based on molecular weights as described by Ahmed
(Ahmed, T. H., Equations of State and PVT Analysis: Applications for Improving Reservoir
Modeling, Gulf Publishing Company, 2007).
δC1-C7+ = 0.00189T - 1.167059 (EOS-19)

δCO2 – N2 = 0.12 (EOS-20)
δCO2-hydrocarbon = 0.12 (EOS-21)
δN2-hydrocarbon = 0.10 (EOS-22)
δCi-C7+ = 0.8δC(i-1) – C7+ (EOS-23)
δCi-Cj = δCi-C7+ (Mj5 – Mi5)/(MC7+5 – Mi5) (EOS-24)
Where T is temperature in ˚R, Ci and Cj represents a component (C2, C3, C4, etc), and Mi and Mj is
the molecular weight of component Ci and Cj.
The internally calculated alternate BIC’s, which are included in the PE² Essentials ALT_BICs.Blib
file (assuming 200˚F in Eqn. EOS-19), are presented in Table EOS-4.
ALT_BIC CO2 N2 C1 C2 C3 iC4 nC4 iC5 nC5 C6 C7+

CO2 0 0.12 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1
N2 0.12 0 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1
C1 0.1 0.1 0 0.000054 0.000038 0.000152 0.000152 0.000448 0.000448 0.001089 0.080341
C2 0.1 0.1 0.000054 0 0.000026 0.000117 0.000117 0.000354 0.000354 0.000867 0.064273
C3 0.1 0.1 0.000038 0.000026 0 0.000073 0.000073 0.000262 0.000262 0.000673 0.051418
iC4 0.1 0.1 0.000152 0.000117 0.000073 0 0 0.000152 0.000152 0.000481 0.041134
nC4 0.1 0.1 0.000152 0.000117 0.000073 0 0 0.000121 0.000121 0.000384 0.032908
iC5 0.1 0.1 0.000448 0.000354 0.000262 0.000152 0.000121 0 0 0.000211 0.026326
nC5 0.1 0.1 0.000448 0.000354 0.000262 0.000152 0.000121 0 0 0.000169 0.021061
C6 0.1 0.1 0.001089 0.000867 0.000673 0.000481 0.000384 0.000211 0.000169 0 0.016849
C7+ 0.1 0.1 0.080341 0.064273 0.051418 0.041134 0.032908 0.026326 0.021061 0.016849 0
Table EOS-4: Internal Alternate Binary Interaction Coefficients for PR and SRK EOS
It is possible to enter alternate BIC’s through the ‘Import BIC’s’ button on the main screen. The
BIC values to be imported must be in an Excel file in the format shown in Figure EOS-3.
The file shown in Figure EOS-5 is called ALT_BICs.xlsx and is included in the “PE
Essentials\Example Input Files\Basic EOS” directory.
Figure EOS-5: Excel File Containing Alternate BIC’s
Figure EOS-6 shows the import of the Alternate BICs from the Excel spreadsheet.
Figure EOS-6: Import of Alternate BIC’s
Importing the BIC’s will make them available for use in the EOS equations. To use these BIC’s in
future runs, they should be saved to Blib library through the ‘Save Alt BIC’s to BLIB Library’ button.
The alternate BIC’s are selected through the ‘Alt BIC’ check box (Figure EOS-1).
EOS.5 EOS Basic Example and EOS Validation

As a basic example and to validate the PE² Essentials implementation of the EOS equations, the
fluid study included in the McCain book, The Properties of Petroleum Fluids, was used. The
sample analysis included on page 260 in the McCain book was entered into the EOS tool. The
actual bubble point pressure for this fluid was reported to be 2620 psi at 220˚F. The EOS tool was
used to calculate bubble point pressure for the PR and SRK formulations, as well as for the Alt BIC
and the results are shown in Figure EOS-7. This EOS model is available in the example PE Tools
database.
Figure EOS-7: EOS Bubble Point Pressure Validation – McCain Example
Calculated, untuned bubble point pressures were consistent for all formulations and ranged from
2505 psi to 2564 psi.
EOS.6 EOS Tuning

When the bubble point pressure is not matched to a satisfactory degree, it is possible to 'tune'
the EOS equation to the known bubble point pressure and thereby improve the accuracy of the
predictions generated by the EOS.
It has been found that slight changes to the characterization of the C7+ fraction can have a
dramatic effect on the EOS-predicted PVT properties of the fluid. As a result, it is possible to ‘tune’
the EOS equation by modifying just the C7+ properties. The usual technique to tune the EOS
equation is to modify the critical properties (Tc and Pc) and the BIC’s of the C 7+ component until
a match is achieved.
Since Tc and Pc for the C7+ fraction are calculated from correlations based on molecular weight
(Equations PVT-13 and PVT-14), tuning the PE² Essentials EOS model is restricted to modification
of the C7+ BIC’s. It should be noted that binary interaction coefficients are the big ‘fudge factors’
in EOS analysis since they are not unique and normally determined through empirical means.
Specialty EOS programs incorporate tuning techniques for all laboratory measured data so that
the final EOS equation can regenerate the complete lab data set. The PE² Essentials EOS model is
a basic implementation and tunes on bubble point pressure only. If a more in-depth EOS
formulation is required, other programs should be used.
Based on the bubble point pressure results for the PR EOS (Section EOS.5), the PR EOS was tuned
to the lab-derived bubble point pressure of 2620 psig (2634.5 psia) with the results presented in
Figure EOS-8. The resulting tuned BIC’s are presented in Table EOS-5.
Figure EOS-8: Tuned Bubble Point Pressure – McCain Example

Table EOS-5: BIC’s for Tuned Bubble Point Pressure – McCain Example
As confirmed in Table EOS-5, only the C7+ BIC’s were modified to achieve the bubble point
pressure match. The modification was to increase the C7+ BIC’s by a factor of 1.07029 as shown
on Figure EOS-8.
EOS.7 EOS Flash Calculations – CCE, DLE, Separator Test

All parameters generated by equations of state are based on flash calculations. Flash calculations
are the basis for reservoir and process engineering calculations. A flash calculation is required to
determine the number of moles of gas and liquid coexisting in a reservoir or a vessel (separator)
at a given pressure and temperature. A flash calculation is also required to determine the
composition existing in the resulting gas and liquid phases.
The calculations required for estimation of moles of gas, moles of liquid, composition of the gas
and composition of the liquid require an estimate of the equilibrium ratios, also referred to as K-
factors. The correlation published by Wilson is incorporated into the PE² Essentials EOS model
(Wilson, G., “A Modified Redlich-Kwong Equation of State Applicable to General Physical Data
Calculations,” Paper No15C, 65th AIChE National meeting, May, 1968).
Ki = Pr’ exp[5.371(1 + ωi)(1 – Tr’)] (EOS-25)

Pr’ = Pci/P
Tr’ = Tci/(T + 460)
Where Ki is the equilibrium ratio of component i, Pci is the critical pressure of component i in
psia, Tci is the critical temperature of component i in ˚R, T is the temperature in ˚F, P is the
pressure in psia and ωi is the acentric factor of component i.
The three main calculations performed with an EOS are the constant composition expansion
(CCE), the differential liberation expansion (DLE) and separator tests.
EOS.7.1 Constant Composition Expansion

The CCE test is performed to determine bubble point pressure, isothermal compressibility of the
single-phase fluid, compressibility factors of the gas phase and total hydrocarbon volume as a
function of pressure. In the CCE, the total hydrocarbon volume is determined at a number of
different pressures through a flash calculation. For this technique, no hydrocarbon material is
removed from the system, which leads to the constant composition for the system. The total
volume at the bubble point pressure is termed the reference volume and the final CCE data is
reported as a ratio of the total volume at a specified pressure to the reference volume (the
volume at the saturation pressure) and is termed the relative volume.
The compressibility of the saturated oil can be calculated from the CCE relative volume using
Equation EOS-26.
Co = -1 ∂Vrel = -1 (Vrel2 – Vrel1) (EOS-26)

Vrel ∂P Vrel2 (P2 – P1)
Where Co is the compressibility in psi-1, Vrel1 is the relative volume at the higher pressure P1 and
Vrel2 is the relative volume at the lower pressure P2.
For the McCain example, using the tuned EOS model, the CCE is presented in Figure EOS-9.
Figure EOS-9: CCE Comparison – McCain Example
The CCE process is indicative of a reservoir’s pressure depletion process. As the reservoir pressure
is reduced, the gas is liberated but remains in contact/equilibrium with the oil in the reservoir
until the gas is produced.
EOS.7.2 Differential Liberation Expansion

The DLE test yields solution gas oil ratio, oil shrinkage, properties and composition of the
liberated gas, specific gravity and density of the remaining oil as function of pressure. During the
DLE test, the liberated gas is removed from the system before establishing a new equilibrium
with the remaining liquid. The fluid is maintained as a single phase, gas saturated liquid for each
pressure point.
The DLE test is performed to atmospheric pressure and 60˚F in order to determine residual, stock
tank density of the oil. The results of the DLE test are presented in Figure EOS-10 and a
comparison with the lab data is shown in Figure EOS-11.
Overall, the match to the McCain lab data is reasonable for this basic EOS system. To obtain a
closer match to the lab data, additional tuning could be attempted by modifying the bubble point
pressure.
Figure EOS-10: DLE Results – McCain Example
Figure EOS-11: DLE Comparison – McCain Example
he DLE process is indicative of production where reservoir fluid is flashed into the wellbore and
produced to the separator. In this process, the gas and oil become independent fluids and the
compositions of gas and liquid hydrocarbons in the system will vary. DLE also describes the
reservoir process after liberated gas begins to flow, independent of the oil.
EOS.7.3 Separator Test

Separator tests are run to determine the optimum production conditions that will maximize the
volume of stock tank oil. The volumetric behaviour of the stock tank fluid is dependent on the
operating conditions of the separator: operating pressure, operating temperature, number of
stages. The Basic EOS model will perform a single stage separator test to determine the
properties of the stock tank oil at the given separator operating conditions.
The separator test, when combined with the DLE test results, will yield the Bo and RS data
required for petroleum engineering calculations. Separator tests start at the bubble point
pressure and reservoir temperature, then proceeds to the separator conditions and finally to
stock tank conditions. The separator test is a differential liberation test in that the liberated gas
is removed from the system prior to the next step.
The result of the separator test for the McCain example is presented in Figure EOS-12.
Figure EOS-12: Separator Test Results – McCain Example
By way of comparison, the laboratory data for this separator test yielded a GOR of 768 scf/bbl
versus the EOS value of 577 scf/bbl; an oil API of 40.7 versus 40.1, a Bo of 1.474 versus 1.362. The
separator gas composition from the EOS and the McCain laboratory data is shown in Table EOS-
6 and indicates that the big difference between the compositions is in the mole% of C1.
Table EOS-6: Separator Test Gas Composition Comparison – McCain Example
Following the separator test, the DLE results are adjusted to the separator conditions for use in
petroleum engineering calculations (Section EOS.7.5).
EOS.7.4 Determining Optimum Separator Operating Conditions

Tests are run at a number of separator conditions to determine the optimum separator operating
conditions that maximizes recovery of oil. Several separator tests were run using the tuned PR
EOS with the McCain example data. The comparative results are presented in Figure EOS-13 for
separator operating temperature of 75˚F.
Figure EOS-13: Separator Optimization – McCain Example
From the EOS separator test results, operating the separator at 150 psi and 75˚F will maximize
oil production. Figure EOS-14 presents the EOS results for the separator test at 150 psi and 75˚F.
Figure EOS-14: Optimum Separator Test – McCain Example
EOS.7.5 Adjusting CCE/DLE Data to Optimum Separator Conditions

After determining the optimum separator operating conditions, the CCE and DLE data are
adjusted to the separator test results. These adjustments are required in order to use the data in
material balance calculations or for input into reservoir simulation models.
First, adjust the formation volume factor, Bo, above the bubble point pressure (CCE) and below
the bubble point pressure (DLE) using the following equations.
Above bubble point: Bo = SepBo RelVol
Below bubble point: Bo = SepBo DLEBo/DLEBobp
Where Bo is the value at the relevant pressure, SepBo is the value generated by the separator
test, RelVol is the CCE relative volume at the relevant pressure, DLEBo is the DLE Bo at the relevant
pressure and DLEBobp is the DLE Bo value at the bubble point pressure.
Example: at P=5062 psi, SepBo=1.3602, RelVol= 0.9555 then Bo=1.2997

Example: at P=1884 psi, SepBo=1.3602, DLEBo= 1.3462, DLEBobp= 1.4441 then Bo=1.2679
Next, adjust the solution gas-oil ratio, RS, using the following equation.
RS = SepGOR – (DLEGORbp – DLEGOR) SepBo / DLEBobp
Where RS is the value at the relevant pressure, SepGOR is the value generated by the separator
test, DLEGORbp is the DLE GOR at the bubble point pressure, DLEGOR is the DLE GOR at the
relevant pressure, SepBo is the Bo value generated by the separator test and DLEBobp is the DLE
Bo value at the bubble point pressure.
Example RS for P=1884 psi:
SepGOR = 574.4, DLEGORbp = 683.8, DLEGOR = 486.8
SepBo = 1.3602, DLEBobp = 1.4441
So, RS=388.8 scf/bbl
These calculations are performed for all the data points. Figure EOS-15 presents the results for
the McCain example.
Figure EOS-15: Adjusted Oil Properties – McCain Example
The adjusted oil properties can then be used in material balance calculations or for input into
reservoir simulation models. It should be noted that the CCE, DLE and adjusted PVT data can be
exported to a CSV file and imported into PE² Essentials Chart for basic plotting (Figure EOS-16).
Use the ‘Export Adj PVT Data’ menu button to export the data.
Figure EOS-16: PE² Essentials Chart – EOS Data
EOS.8 Gas-Oil Recombination for EOS

A gas-oil recombination routine is included in the PE² Essentials EOS tool and can be used to
generate the recombined reservoir fluid from composition analysis of the oil and gas. The
recombined fluid compositions can then be transferred to the EOS model for analysis.
Once the liquid and gas compositions are entered, recombination is performed by molar addition
of the fluids based on the inputs for gas and oil rates (Figure EOS-17). This allows the regeneration
of reservoir fluids for different separator rates.
Figure EOS-17: Gas-Oil Recombination
Following recombination, the reservoir fluid can be transferred to the EOS model through the
‘Transfer to EOS’ button.
108 Monte Carlo Simulation: Volumetrics
Monte Carlo Simulation: Volumetrics

Monte Carlo simulation is a technique that relies on repeated random sampling to obtain
numerical results. The purpose is to use randomness to solve problems that might be
deterministic in principle. In a Monte Carlo simulation, random values are selected for each of
the input variables, within a defined range and distribution of values. The model is calculated
based on these random values and the model is recorded. The simulation can be repeated
hundreds or thousands of times, each time using different randomly selected values to generate
a large number of results, each based on random input values.
When calculating volumetrics, the volume calculation is deterministic, but the input values have
a range of possible values. Use of Monte Carlo simulation will generate a range of possible
outcomes which can be used in decision analysis. An excellent reference for this subject is the
Project Economics book by Mian (Mian, M. A., Project Economics and Decision Analysis Volume
II: Probabilistic Models, PennWell, 2011).
PE² Essentials includes a ‘Monte Carlo Volumetric’ tool (Figure MCV-1a and 1b) that can be used
to generate probabilistic values for oil or gas in-place and recoverable oil/gas volumes.
Figure MCV-1a: PE² Essentials Monte Carlo Volumetrics – Gas Reservoir Tool
Figure MCV-1b: PE² Essentials Monte Carlo Volumetrics – Oil Reservoir Tool
The tool includes all the parameters that go into the calculation of in-place and recoverable
volumes, not just the reservoir parameters but the fluid parameters as well. Since fluid properties
are calculated at each sampling step, the fluid properties are consistent with the reservoir
properties at each sample step.
Input parameters to Monte Carlo simulation have a continuous probability distribution. This
means that the parameter can have any number within the range of the distribution. The most
common distributions used in the oil industry are normal, lognormal, uniform and triangular. For
the Monte Carlo Simulation model included in PE² Essentials, normal, triangular and uniform can
be used. A skewed normal distribution can also be input. Figure MCV-2 shows a normal and a
skewed normal probability distribution.
It is possible to use a triangular distribution for the input parameters. The triangular distribution
is similar to a skewed normal distribution but does not have the high and low tails that a normal
distribution has. This reduces the number of very high and very low samples.
Figure MCV-2: Probability Distributions
For a normal distribution, the absolute difference between the mid value and the low and high
values are equal (P50 – P90 = P10 – P50) but for the skewed normal distribution the differences
are not equal (P50 – P90 < > P10 – P50). For a uniform distribution, all values are equal (P90 =
P50 = P10).
Note that the nomenclature used in the oil industry is that P90 is the lower value and P10 is the
higher value. What this means is that for the P90 value, there is a 90% probability that the actual
value will be higher (10% probability it will be lower) than that value, for the P50 value there is
an equal probability that the actual value will be higher or lower than that value, and for the P10
value there is a 10% probability that the actual value will be higher (90% probability it will be
lower) than that value. Occasionally, the P95 or P99 value is used when evaluating disaster
scenarios.
Executing a Monte Carlo simulation will result in a cumulative distribution function (CDF) which
can be used to extract the value for any given probability (Figure MCV-3).
Figure MCV-3: Cumulative Distribution Function

To calculate oil/gas volumetrics, the distribution of the input parameters is entered, the CDF’s
are calculated and sampled using a random number generator. The fluid properties are
calculated based on the randomly sampled reservoir parameters and the volumes are calculated
(Figures MCV-4, MCV-5 and MCV-6). If the ‘Seed’ is entered as ‘-1’, the tool will present an
equivalent seed that can be used to regenerate the current results.
Figure MCV-4: Monte Carlo Simulation – Gas Input Parameters
Figure MCV-5: Monte Carlo Simulation – Input Distributions

Figure MCV-6: Monte Carlo Simulation – Gas Simulation Results
Once the volumetric calculations are performed, the values are stored as an output CDF. The
P90, P50 and P10 values are extracted from this output CDF and the deterministic values for the
specific realization are extracted for use in other calculations. PE² Essentials also reports
deterministic models for P1 and P99 along with the P90, P50, P10 and expected value (EV)
deterministic models (Figure MCV-7). It should be noted that these deterministic models are just
one realization and if the Monte Carlo simulation is run again, different deterministic models will
be generated. Figure MCV-8 presents an example of deterministic oil models.
Figure MCV-7: Monte Carlo Simulation - Deterministic Gas Models
Figure MCV-8: Monte Carlo Simulation - Deterministic Oil Models

Computers do not generate truly random numbers instead they generate pseudo-random
numbers based on an initial seed for the random number generator. As a result, to generate the
same probabilistic profile, and corresponding deterministic models, a constant ‘Seed’ value can
be entered in the ‘Seed (Random=-1)’ input box. By entering a specific seed value, the same
random number sequence will be generated for the Monte Carlo simulation and the output
results will be the same. Entering '-1' for the seed will generate random seed numbers based on
an internal timer so the forecast will be different each time a simulation is run.
The Monte Carlo simulation results (‘Save Simulation’) as well as the deterministic models (‘Save
Model’) and the CDF profiles (‘save SDF Results’) can be saved as ‘csv’ files. The CDF data can be
imported into PE² Essentials Chart but the x and y axis will be reversed as shown in Figure MCV-
9. To properly plot the data, it should be imported into Excel.
Figure MCV-9: Monte Carlo Simulation – PE² Essentials Chart Plot

114 PE Graph Tool
PE Graph Tool
The PE² Essentials ‘PE Graph’ Tool imports PE Essentials generated CSV files and can construct
plots with user defined x- and y- axes. Multiple data can be plotted on the y-axis.
The tool is made up of three components: the control section (Figure GRP-1); the plotting section
(Figure GRP-2); and the data import section (Figure GRP-3).
Figure GRP-1: PE² Essentials Graph Tool - Control

Figure GRP-2: PE² Essentials Graph Tool - Plotting

116 PE Graph Tool
Figure GRP-3: PE² Essentials Graph Tool – Data Import
GRP.1 Data Import

Most CSV files generated by the PE Essentials tools can be imported into PE Graph for plotting of
the data.
To import the data, open the import window either using the ‘File’ menu or the ‘Load CSV File’
button on the Plotting window. After opening the import window, drag and drop the csv file onto
the window. This will open the file allowing it to be reviewed before importing the data into PE
Graph.
The data is loaded into PE Graph by clicking the ‘Import Data button. This will import the data
headings and units into the ‘Select x-Axis Data’ and ‘Select y-Axis Data’ tables and the data set
name into the ‘Select Data Set’ table.
Note – multiple CSV files can be imported into the tool, but they have to be the same. For
instance, all the Raw CSV data files generated by the PDA tool can be imported, but an Analyzable
CSV data file cannot be imported with a Raw CSV file. The first file imported determines the
remaining files that can be imported. This is because the x-Axis and y-Axis data tables are
imported once, and different CSV files have different data stored in the files.
To make import of PDA and Production Database generated CSV files easier, these tools save all
exported CSV files to the ‘PE Essentials 2021\CSV Output Files’ directory. All other tools store the
CSV files in a user-specified directory.
It is possible to import a user-built CSV file into PE Graph but when generating the file, the format
should follow the PE Essentials CSV format of the appropriate tool. Specifically, the first 4 lines of
the CSV file are used to confirm that the file can be imported into PE Graph.
GRP.2 Plotting
Up to nine individual plots can be built with the PE Graph tool. To modify a specific plot, click on
the plot and a blue border will indicate the selected plot. To modify a specific curve, click on the
curve and it will be highlighted. In both cases after the plot/curve has been highlighted, the plot
parameters can be modified from the control window.
After the plots have been finalized, the results can be saved to a png file (File/Save Plots to File…)
or sent to the printer (File/Print Plots…) – Figure GRP-4.
Figure GRP-4: PE Graph – Saving and Printing Plots
Figure GRP-5 is an example of multi-data plotting of PDA data. The plot was built with the
parameters shown in Figure GRP-6, then the plot was saved to a PNG file.
Figure GRP-5: PE Graph – Example Plot of PDA Data
Figure GRP-6: PE Graph – Parameters for Plot

118 PE Graph Tool
Figures GRP-7 and GRP-8 present an example of multi-well, multi-graph plotting of Production
Database data.
Figure GRP-7: PE Graph – Example Plot of Production Database Data
Figure GRP-8: PE Graph – Parameters for F-12 H Well Plot
Note – when increasing/decreasing the number of displayed graphs, some of the control
parameters are reset to the base values. The scales and the headings for each graph are retained.
GRP.3 PE Graph Import File Format

The PE Graph tool can be used to import and plot non-PEE csv files as long as the format of the
csv file is set up to the PE Graph import format.
A normal PE Essentials csv file has the format shown in Figure GRP-9.
Figure GRP-9: PE Graph – Parameters for F-12 H Well Plot
The file contains the following:

Line 1: File information and units
Line 2: Well name/model name information
Line 3: Date contained in the columns in the file
Line 4: Units for the data in the file
Line 5+: Data
PE Graph reads the data in the second line, up to the first comma, to assign the name for the
imported data.
The third and fourth lines are used to set up the x-data and y-data tables.
The data starting from the fifth line is read into the plotting arrays.
120 PE Well Essentials
PE Well Essentials
Well Essentials includes the following tools:
• THP-to-BHP for gas wells
• Quick Log Analysis
• Hydraulic Fracture Design
• Artificial Lift Design
• Pressure Transient Analysis (Build-up)
• Volumetric Analysis
• Hydrate Analysis
THP – BHP Tool: Gas Well

The ‘THP-BHP Gas Well’ tool will model pressure drop in vertical and horizontal wells as well as
gas flow through the tubing-casing annulus (Figure THPG-1). For a horizontal well, pressure drop
is calculated to the top of the lateral (assumes constant pressure throughout the lateral).
Figure THPG-1: PE² Essentials THP-BHP Gas Well Tool

The main references for the tubing routines are by Beggs (Beggs, H. D. Production Optimization
Using NodalTM Analysis, OGCI Publications, 1991) and by Guo, Sun and Ghalambor (Guo, B., Sun,
K. and Ghalambor, A., Well Productivity Handbook, Gulf Publishing Company, 2008).
Gas flow in a well is governed by the conservation of energy (first law of thermodynamics) which
yields a differential equation for steady state flow in a pipe. For gas wells, the only significant
terms are the terms representing the hydrostatic gradient and the friction gradient.
∇P = ∇Phyd + ∇Pfric
There are 2 options for gas well pressure drop correlations included in PE² Essentials: Average TZ
(for gas and condensate) and Guo-Ghalambor (for gas, condensate, water. It should be noted
that only the Guo-Ghalambor correlation includes the option for modeling sand in the flow
stream.
THPG.1 Average Temperature/Z-Factor Technique

The pressure gradient in an angled gas well, where φ is the well angle measured from the vertical
(calculated from the MD and TVD data), can be simplified to the following differential equation.
dP = ρgcosφ + fρv2
dL gc 2gcd
This equation can be directly integrated to a solution as follows.
P12 = P22exp(S1) + 0.000667 fmQge2Za2Ta2(exp(S2) – 1) / ID5 (THPG-1)

S1 = 0.0375 SGwTVD/(TaZa)
S2 = S1MD/TVD
SGw = (GCR SGg + 4584 SGc) / (GCR + GEc)
GEc = 133000 SGc / MWc
MWc = 6084 / (API - 5.9)
SGc = 141.5 / (API + 131.5)
Qge = Qg(1 + GEc/GCR)
fm = [1.14 - 2Log(δ/ID + 21.25/Re0.9)]-2
Re = 20011 SGw Qeg / (ID µg)
Ta = Tfa + 460
Where: P is pressure in psia, Qge is equivalent gas rate (gas + condensate) in mscf/d, Za is average
Z based on P1 and P2, Ta is average temperature in ˚R for the depth increment, ID is pipe ID in
inches, SGg is gas specific gravity, SGc is condensate specific gravity, MWc is condensate molecular
weight, GCR is producing gas-condensate ratio in scf/bbl, API is the ˚API of the condensate, δ is
the absolute roughness of the tubing in inches, fm is the Moody friction factor, Re is the Reynolds
122 THP – BHP Tool: Gas Well
number and µg is the average gas viscosity in cp based on P1 and P2 and Tfa is average temperature
in ˚F.
Since Z and gas viscosity are functions of pressure, the pressure calculations are done in steps.
The pressures are assumed as P1 is bottomhole and P2 is tubing head. If performing the
calculations by hand, two to four depth increments may be sufficient. The PE² Essentials routines
uses 100 depth increments to calculate the pressure.
THPG.2 Guo-Ghalambor Technique

In 2005 Guo and Ghalambor developed a four-phase (gas-oil-water-sand) pressure drop model
(Guo, B. and Ghalambor, Natural gas Engineering Handbook, Gulf Publishing Company, 2005).
The model was an extension of the original formulation by Guo for coal bed methane wells. The
solution is a complex equation that requires iteration to solve. This model is included since sand
production in hydraulically fractured well may be a common occurrence.
It should be noted that the Guo-Ghalambor model is a no-slip model which limits its validity to
mist flow. As a result, it is not valid for gas wells that produce significant water.
a(cosφ + d2e)MD = 144b(P1 – P2) + X1 - X2 (THPG-2)
X1 = (0.5-bM) ln{[(144P1 + M)2 + N]/[(144P2 + M)2 + N]}

X2 = M + bN/c - bM2N-0.5 {tan-1[(144P1 + M)N-0.5] - tan-1[(144P2 + M)N-0.5]}
a = (0.0765 SGgQg + 350 SGoQo +350 SGwQw + 62.4 SGsQs) / (4.07TaQg)
b = (5.615Qo + 5.615 Qw + Qs) / (4.07TaQg)
c = 1.2431TaQg / ID2
d = 0.30436 (5.615Qo + 5.615 Qw + Qs) / ID2
e = 0.18651fm/ID
fm = [1.74 - 2Log(2δ/ID)]-2
M = cde / (cosφ + d2e)
N = c2e cosφ / (cosφ + d2e)2
Ta = Tfa = 460
Where: P1 is bottomhole pressure in psia, P2 is tubing head pressure in psia, φ is the well angle
measured from the vertical, Qg is gas rate in scf/d, Qo is oil rate in bbl/d, Qw is water rate in bbl/d,
Qs is sand rate in ft3/d, SGg is gas specific gravity, SGo is oil specific gravity, SGw is water specific
gravity, SGs is sand specific gravity, Ta is average temperature in ˚R for the depth increment and
ID is pipe ID in inches and Tfa is average temperature in ˚F.
The Guo-Ghalambor model is solved in one depth increment using an iterative process where an
estimate of the unknown pressure is made and then modified until the left-hand side of the
Equation THPG-2 equals the right-hand side of the Equation THPG-2.
THPG.3 Options for THP-BHP Calculations
THPG.3.1 Well Options

Pressure calculations can be performed for vertical or horizontal wells and for tubing or tubing-
casing annulus flow by checking the appropriate boxes in ‘Well Options’ (Figures THPG-2 and
THPG-3).
Figure THPG-2: Well Type and Flow Path Options
Figure THPG-3: Vertical and Horizontal Well Options
Annular flow is modeled by calculating an equivalent ID for the flow path.
IDann= (CsgID + TubOD)2 (CsgID – TubOD)3 (THPG-3)
Where: IDann is the equivalent ID of the annulus in inches.

THPG.3.2 Minimum gas Rates to Lift Liquid

The bottomhole pressure calculation routine also includes a calculation of the minimum gas rate
required to continuously lift water and sand (Figure THPG-4) for information purposes.
The minimum gas rate calculations are based on the Turner equations.
MinQg = 3060 THP Area Vmin / ((THT + 460) Z) (THPG-4)

Vmin(water) = 1.3σw0.25 (62.4 SGw - RhoGas)0.25) / RhoGas0.5 (THPG-5)
Vmin(sand) = 1.3σw 0.25 (62.4 SGs - RhoGas)0.25) / RhoGas0.5 (THPG-6)
RhoGas= 2.699 SG THP / (THT + 460)
Where: MinQg is the minimum gas rate based on the minimum velocity, Vmin to lift water or sand
in mscf/d, σw is the water-gas interfacial tension, SG is specific gravity of gas, SGw is specific gravity
of water, SGs is specific gravity of sand, Area is the cross-sectional area of the tubing in ft2, THP is
the flowing wellhead pressure in psi and THT is flowing wellhead temperature in ˚F.
Figure THPG-4: Output Listing of Forecast
THPG.3.3 ∆P Correction Factor

There is a “∆P Correction Factor” included for gas wells that will assist with matching of known
pressure data. Entering a value for this factor will multiply the friction factor (fm) by the entered
value. This will modify the friction gradient term in the pressure drop calculations. To change the
hydrostatic gradient, modify the gas specific gravity value.
To confirm the pressure match, the ‘Single Point THP>>BHP’ is selected on the main screen
(Figure THPG-5).
Figure THPG-5: Converting a Single TH Pressure to BH Pressure
With this option the best correlation can be chosen, and the correction factor can be evaluated.
While this option screen is active, the correction factor can be changed and then clicking the
‘Calculate’ button will re-calculate the BHP.
THPG.3.4 THP/CHP to BHP Conversion

It is possible to import a table of pressure/rate data and convert it to BHP. This is done by clicking
the ‘Import THP Table’ menu button. This will open the ‘THP to BHP Conversion’ page (Figure
THPG-6). The pressure/rate data is imported from the PE Tools database or an Excel spreadsheet
by clicking on the ‘Load Data’ button. Figure THPG-7 shows the import of data from the database.
Figure THPG-6: Converting a Table of THP/CHP Data to BHP

Figure THPG-7: Loading and Converting THP to BHP Data
The data presented in Figure THPG-6 is included in an Excel file called ‘THP to BHP Example
Data.xlsx’ located in the “PE Essentials\Example Input Files\Excel Files” directory.
After the data has been imported and converted to BHP, it is possible to save the data to a CSV
file by clicking the ‘Export BHP Data to CSV File’. After exporting the BHP data, the results can be
imported into PE² Essentials Chart for plotting.
The THP data can also be saved to the PE Tools database by clicking ’Save THP Data to PE Tools
db’.
It is possible to modify the parameters for the well and recalculate the BHP without re-importing
the THP data – for example, ‘∆P Correction Factor’ can be modified on the main screen while the
conversion page remains opened. Check the ‘Convert THP>>BHP’ button on the conversion page
to update the BHP.
THPG.4 IPR Well Operating Point

Entering the Jones IPR parameters of c and n (Equation THPG-7 and Figure THPG-8) will enable
the generation of the operating point for the well based on the intersection of the IPR curve and
the tubing performance curve.
Q = (Pi2 – Pwf2)n (THPG-7)

Figure THPG-8: Well IPR and Performance Plot
Figure THPG-9: Well Performance Listing
Figure THPG-9 presents the actual well operating points at the different tubing conditions.
Clicking the ‘Save IPR and Solution’ will save the well performance data to a CSV file.
128 Quick Log Analysis Tool
Quick Log Analysis Tool

There are numerous log analysis books available. The definitive reference for log analysis
techniques is Crain. E. R., The Log Analysis Handbook, Volume 1, Quantitative Log Analysis
Methods, PennWell Publishing Company, 1986. In addition, some of the techniques and
equations presented here were also published in LeBlanc, D. P., “Enhanced Shaly Sands and
Carbonate Analysis on the HP-41c”, Canadian Well Logging Society Journal, December, 1983 and
LeBlanc, D. P., “Shaly Sands and Carbonate Analysis on the HP-41c”, Canadian Well Logging
Society Journal, December, 1982.
PE² Essentials ‘Quick Log Analysis’ tool (Figure QLA-1) is a 'quick and dirty' log analysis tool that
can be used to evaluate up to six reservoir intervals as well as pressure-depth data.
Figure QLA-1: PE² Essentials Quick Log Analysis Tool
Commercial log analysis packages can load raw log data into the system and can clean-up the
data prior to performing log analysis on a sample basis. The purpose of the Quick Log Analysis
tool is to be able to perform a quick analysis of a potential reservoir interval so average log values
are entered and used in the calculations.
QLA.1 Log Analysis Models

A number of petrophysical models make up a log analysis system. PE² Essentials Quick Log
Analysis tool includes models for shale, porosity, water saturation and permeability (Figure QLA-
2).
Figure QLA-2: PE² Essentials Quick Log Analysis Tool - Model Options
Although there are many models and techniques available, the following models are some of the
most common used models for log analysis.
QLA.1.1 Shale Models

Shale volume is required to correct porosity and water saturation results for the effects of shale.
Shale volume is also an indicator of reservoir quality and is used to indicate net pay.
The most common models used to calculate shale volume are based on the Gamma Ray (GR) log,
the Spontaneous Potential (SP) log, and the Neutron-Density logs (Figure QLA-3).
Figure QLA-3: Shale Model Options
The ‘Gamma Ray’ and ‘Clavier’ options both use an index (ISH) calculated from the GR log.
For the ‘Gamma Ray’ option, the assumption is that the shale volume, V sh, follows a linear
relationship from the clean GR value, GR0, to the 100% shale GR value, GR100.
ISH = (GRlog - GR0) / (GR0 - GR100) (QLA-1)

Vsh = IGR (QLA-2)
Shlumberger and Dresser-Atlas (now Baker-Hughes) published optional equations that changed
the Vsh calculation to a non-linear function of IGR. Shlumberger’s equation is included as the
“Clavier” option as follows.
Vsh = 1.7 - (3.38 - (ISH + 0.7)2)0.5 (QLA-3)
The Dresser-Atlas equations are as follows.
Vsh = 0.083 (23.7ISH - 1) :Tertiary Rocks (QLA-4)

Vsh = 0.33 (22ISH - 1) :Older Rocks (QLA-5)
Note that the Dresser-Atlas equations are not included in the PE² Essentials Quick Log Analysis
model.
The ‘SP’ option is similar to the Gamma Ray option in that the assumption is that the shale
volume, Vsh, follows a linear relationship from the clean SP value, SP0, to the 100% shale SP value,
SP100.
ISH = (SPlog - SP0) / (SP0 - SP100) (QLA-6)

Vsh = ISH (QLA-7)
The ‘N-D CrossPlot’ option uses the Neutron (φN) and Density (φD) logs to determine Vsh.
Vsh = (φNlog - φDlog) / (φNSH - φDSH) (QLA-8)
The Vsh calculated by N-D crossplot may be impacted by the assumed value of the rock matrix,
the existence of gas and the rugosity of the borehole.
Choosing the ‘Minimum’ option will choose the minimum Vsh calculated using all models.
QLA.1.2 Porosity Models

There are three basic porosity logging tools – sonic, density and neutron. Each tool can be used
as an independent indicator of porosity or the Neutron (φN) and Density (φD) logs can be used
simultaneously (crossplotted) to determine the porosity (Figure QLA-4).
Porosity calculated from logs without applying a shale correction is termed apparent or total
porosity. Effective porosity is the resulting porosity after applying a shale correction.
Figure QLA-4: Porosity Model Options

The sonic log records the sonic travel time, ∆t, of a small interval of rock. The equation for sonic
porosity, φs, is as follows.
φs = (∆tlog - ∆tma) / (∆tw - ∆tma) (QLA-9)
The equation for sonic-based porosity, with shale correction is as follows.
φ = (∆tlog - ((1 - Vsh) ∆tma) - (Vsh*∆tsh)) / (∆tw - ∆tma) (QLA-10)

Where: φ is the porosity in decimal, ∆tlog is the sonic log reading in the interval of interest, Vsh is
the volume of shale, ∆tma is the rock matrix travel time (Table QLA-1), ∆tsh is the sonic travel time
in 100% shale and ∆tw is the sonic travel time in water.
Table QLA-1: Sonic ∆t Values
One caution to be kept in mind is what is termed as the ‘sonic compaction factor’, Cp. This is only
an issue if ∆tsh is greater than 100µsec/ft (328µsec/m), which could occur for depths less than
3000ft (1000m).
Cp = ∆tsh / 100 or Cp = ∆tsh / 328 (QLA-11)
φ = φs / Cp (QLA-12)
The Quick Log Analysis tool assumes that Cp is one; i.e. no compaction correction applied to φs.
The density log measures the bulk density, ρb, of the rock and records either the density, ρb, or
the density porosity, φD. The equation for density porosity is as follows.
φD = (ρma – ρb) / (ρma – ρf) (QLA-13)
The equation for density-based porosity, with shale correction is as follows.
φ = φD - Vsh φDsh (QLA-14)

φDsh = (ρma – ρsh) / (ρma – ρf) (QLA-15)
Where: φ is the porosity in decimal, φD is the density porosity, Vsh is the volume of shale, φDsh is
the density porosity of 100% shale, ρb is the recorded log bulk density, ρma is the rock matrix
density (Table QLA-2), ρsh is the density of 100% shale and ρf is the density of mud filtrate (water).
Table QLA-2: Density Values
The existence of gas may cause the calculated porosity to be too high because a low bulk density
is recorded. This can be corrected by modifying ρf as follows.
ρf = (1 – Sxo)ρg + Sxoρmf (QLA-16)

Where: Sxo is the water saturation in the invaded zone, ρg is the gas density and ρmf is the density
of the mud filtrate.
Note that this correction is not incorporated into the Quick Log Analysis model.
The neutron log measures hydrogen index of the rock and presents it as the neutron porosity,
φN, of the rock.
The caveat for the neutron log is the matrix on which the log is recorded. Most analysis and
interpretation charts assume that the neutron log is recorded on a limestone matrix. For this
situation, the neutron porosity can be corrected to a different lithology. As an example, for
Schlumberger CNL logs the following corrections apply.
- Limestone to sandstone: φNss = 0.222φNLS2 + 1.021φNLS + 0.039 (QLA-17)

- Limestone to dolomite: φNdol = 1.4φNLS2 + 0.389φNLS - 0.01259 (QLA-18)
In general, if the neutron porosity is presented on a limestone scale, add 0.04 to the log reading
to yield neutron porosity for a sandstone matrix, and vice-versa.
The equation for neutron-based porosity, with shale correction is as follows.
φ = φN - Vsh φNsh (QLA-19)

Where: φ is the porosity in decimal, φN is the neutron porosity, Vsh is the volume of shale and
φNsh is the neutron porosity of 100% shale.
The existence of gas may cause the neutron porosity read too low because of the reduced
hydrogen index in the gas. This effect is not a constant and is a function of the density and
wetness of the gas. Dry, low-pressure and high temperature gases have larger impact on the
neutron porosity. Shale will reduce the gas effect since shale φNsh tend to be high values.
If the existence of gas is suspected, using the neutron-density crossplot model is preferred to
calculate porosity. The ‘N-D CrossPlot’ model will take an average of the neutron and density log
porosities.
φ = (φN + φD) / 2 (QLA-20)
If the neutron porosity is more than 0.02 less than the density porosity, a gas effect is assumed
to exist and the porosity calculation is as follows.
φ = ((φN2 + φD2) / 2)0.5 (QLA-21)
This weights the porosity towards the density porosity and is valid in a clean sandstone or a
carbonate reservoir.
The ‘N-D Shaly Sand’ model takes an average of the shale-corrected neutron and density
porosities.
φ = (φNc + φDc) / 2 (QLA-22)
φNc = φN - Vsh φNsh (QLA-23)
φDc = φD - Vsh φDsh (QLA-24)
If the shale-corrected neutron porosity is more than 0.02 less than the shale-corrected density
porosity, a gas effect is assumed to exist and the porosity calculation is as follows.
φ = ((φNc2 + φDc2) / 2)0.5 (QLA-25)
This weights the porosity towards the density porosity.
QLA.1.3 Water Saturation Models

There are two shaly sand water saturation models as well as the Archie water saturation model
included in the Quick Log Analysis model (Figure QLA-5).
Figure QLA-5: Water Saturation Model Options
The Archie water saturation model is the original standard used by the oil industry but is now
only used for clean sandstones or carbonates. The equation is as follows.
Sw = (a Rw / φm Rt)1/n (QLA-26)
Where: a is the tortuosity factor (1 for carbonate and 0.62 for sandstone), m is the cementation
exponent (2 for carbonate and 2.15 for sandstone), n is the saturation exponent (ranges from 1.8
to 2.5, normally set equal to 2), Rw is the formation water resistivity and Rt is the true formation
resistivity.
For the flushed zone, the Archie equation is as follows.
Sxo = (a Rmf / φm Rxo)1/n (QLA-27)

Where: Rmf is the mud filtrate resistivity and Rxo is the flushed zone resistivity.
If the value for Sxo is greater than the value for Sw, this may be an indication that movable
hydrocarbons exist in the reservoir.
The Archie equation is inaccurate when shale is present in the reservoir. One of the most
commonly used shaly sand water saturation model is the Simondoux model.
The Simondoux water saturation equation is as follows.
Sw = [(A2 + B)0.5 – C]2/n (QLA-28)

A = [(1 - Vsh) Vsh aRw/2φmRsh]
B = (1 - Vsh) aRw/φmRt
C = 0.5(1 - Vsh) Vsh aRw/φmRsh
to 2.5, normally set equal to 2), Rw is the formation water, Rsh is the resistivity of the shale, and
Rt is the true formation resistivity.
An alternative shaly sand model is the empirical Poupon-Leveaux model, also referred to as the
Indonesian model. The Poupon-Leveaux water saturation equation is as follows.
Sw = [(A + B)2 Rt]-1/n (QLA-29)

A = [Vsh(2- Vsh) / Rsh]0.5
B = (φm / aRw)0.5
to 2.5, normally set equal to 2), Rw is the formation water, Rsh is the resistivity of the shale, and
Rt is the true formation resistivity.
QLA.1.4 Permeability Models

There are two permeability models included in the Quick Log Analysis model (Figure QLA-6).
Figure QLA-6: Permeability Model Options
The Wyllie-Rose permeability model is the original method used to calculate permeability from
logs. It is accurate when calibrated to core data. The equation is as follows.
k = 65000Cor φ6 / Sw2 (QLA-30)

Where: k is permeability in md, Cor is the correlating parameter equal to 1 for oil and 0.1 for gas.
The Timur permeability model is similar to the Wylie-Rose formulation but incorporates different
constants. The equation is as follows.
k = 3400Cor φ4.4 / Sw2 (QLA-31)

Where: k is permeability in md, Cor is the correlating parameter equal to 1 for oil and 0.1 for gas.
The Cor constant is entered as the ‘Calibration’ input on the main screen. This factor can also be
used to calibrate the log reading to the core data.
QLA.2 Log Analysis Parameters and Results

A number of parameters are required in order to generate a log analysis. Formation water
resistivity, Rw, is one of the main parameters required for all water saturation models. There are
two main techniques to derive Rw; calculate directly from a water bearing interval, or estimate
the value based on water salinity (Figure QLA-7).
Figure QLA-7: Input Formation Water Resistivity
Estimating Rw from a water zone is done using the Archie equation with input R t and φ. For this
calculation the a, m and n parameters need to be entered as well (Figure QLA-8). To estimate Rw
based on salinity, the ppm NACL and the temperature have to be entered. If both calculations
are performed, then the Rw to be used in the water saturation calculations has to be specified.
The net pay cutoffs and the reservoir parameters are optional inputs but are normally entered
so that net reservoir and hydrocarbon (oil and/or gas) volumes can be calculated.
Input of shale parameters are required if shaly sand analysis is to be performed (Figure QLA-9).
Figure QLA-8: Input Analysis Parameters

Figure QLA-9: Input Shale Parameters
Once the model/analysis parameters are entered, the log data is entered. Note that the Quick
Log Analysis routine is not a foot-by-foot analysis of the log data, but instead uses average
interval data. Up to six intervals can be entered (Figure QLA-10). There is no option to enter net-
to-gross values so, for shaly sands, sands with distributed shale or interbedded sands in
laminated sand/shale sequences, averages for the interval should be entered.
It is assumed that the neutron and density porosity values are calibrated to the proper matrix.
There is no internal matrix correction applied to these log values.
Figure QLA-10: Input Log Parameters
After all analysis parameters have been entered, the log analysis models have been chosen and
the log data has been entered, log analysis is initiated by clicking the ‘Run Log Analysis’ button
(Figure QLA-11).
Figure QLA-11: Output Log Analysis Results

Included in the log analysis results (Figure QLA-11) is sonic-based and density-based total organic
carbon (TOC). These calculations are included in the analysis results but they are purely
qualitative estimates and should be used for comparative purposes only. The TOC’s are calculated
as follows (ref: https://spec2000.net/11-vshtoc.htm).
Sonic: TOC = 0.234 * (∆t + 59.433log(R)) - 31.86 (QLA-32)

Density: TOC = 45.14 -142.9(ρ – 1.014)/(log(R) + 4.122) (QLA-33)
Where: ∆t sonic travel time in µsec/ft, R is the resistivity and ρ is bulk density in g/cc.
The analysis results also include a summation of the hydrocarbon volumes contained in the
intervals identified as net pay (Figure QLA-12).
Figure QLA-12: Reservoir Volumetrics Based on Log Analysis Results
To disable either the oil volume calculation or the gas volume calculation, enter zero for the ‘Oil
Bo’ or the ‘Gas Bg’.
QLA.3 Pressure-Depth Analysis

The Pressure Gradient Analysis tool is accessed from the main menu (Figure QLA-13). Data is
loaded into the tool by clicking ‘Load Data’.
Figure QLA-13: Pressure-Depth Analysis

Data can be entered manually in the Depth/Pressure table; imported from a PE Tools database
file (‘Load PE db Data’), Figure QLA-14; or imported from an Excel file (‘Excel Import’), Figure QLA-
15. The data in the figures is from ‘Gradient Data.xlsx’ included in the ‘Example Input Files\Excel
Files’ directory.
Figure QLA-14: Pressure-Depth Analysis, Database Import
Figure QLA-15: Pressure-Depth Analysis, Excel Import

After the data is loaded, it can be saved to the PE Tools database with ‘Save to PE db’ (Figure
QLA-13). The database model will also contain the analysis parameters for the well.
This tool can be used to estimate fluid properties and fluid contacts.
Gas/water/oil gradient lines can be added to the plot. The line can be moved by clicking on the
upper left end of the line and moving it to the desired location. The gradient is then modified to
get the best fit. The equivalent reservoir fluid property, at reservoir conditions, is calculated for
the given gradient and presented in the ‘Fluid Properties’ box (Figure QLA-15).
Figure QLA-15: Pressure-Depth Analysis Example
If more than one gradient line is included on the plot, the fluid contact - GOC, OWC or GWC – can
be determined by clicking the appropriate ‘Calculate’ button.
Hydraulic Fracture Design Tool

The theory in the PE² Essentials Hydraulic Fracture Design model are referenced from SPE 98047
(Daal, J. A. and Economides, M. J., Optimization of Hydraulically Fractured Wells in Irregularly
Shaped Drainage Areas, SPE98047, 2006), the book Unified Fracture Design (Economides, M. J.,
Oligney, R. E. and Valko, P. P., Unified Fracture Design, Bridging the Gap Between Theory and
Practice, Orsa Press, 2002) and the book Modern Fracturing (Economides, M. J. and Martin, T.
Modern Fracturing Enhancing Natural Gas Production Energy Tribune Publishing, 2007).
The ‘Hydraulic Fracture Design’ tool (Figure HYD-1) is used to generate fracture parameters, for
a given hydraulic fracture design, that can be used when forecasting production from a
hydraulically fractured horizontal well. The theoretical well PI is presented so that the fracture
design parameters can be optimized to maximize the well PI.
Figure HYD-1: PE² Essentials Quick Log Analysis Tool

142 Hydraulic Fracture Design Tool
The resulting fracture parameters (xf, wf, kf, and number of frac stages) can be saved and
imported into the ‘Frac Parameters’ sheet of the Unconventional Forecast tool.
The Hydraulic Fracture Design tool enables the fracture parameters to be optimized based on a
calculation of the resulting pseudo-steady state productivity index for the well.
If ‘Deliverability Parameters’ are entered, an estimate of the stabilized well production potential
at the bottom hole ‘Flowing Pressure’ entered into the model will be reported.
The optimization routine is based on using the concept of Proppant Number, Np, as a normalizing
and descriptive parameter (Valko, P. P., Economides, M. J., “Heavy Crude Production from
Shallow Formations: Long Horizontal Wells Versus Horizontal Fractures”, SPE50421, 1998) as
presented by Daal and Economides.
In general, for any Np there exists a maximum dimensionless productivity index (JD) that
corresponds to the optimum dimensionless fracture conductivity (CfD). With this value of the
dimensionless fracture conductivity it is possible to determine the fracture width (wf) and length
(xf) depending on the properties of the reservoir and the proppant.
For reference purposes, Figure HYD-2 presents the parameters for hydraulically fractured wells.
Figure HYD-2: Hydraulic Fracture Parameters – Top View
Since one hydraulic fracture, or stage, in a horizontal well has a non-square, non-radial drainage
area, it is necessary to use the rectangular shape factors (CA) to determine the productivity index
for each fracture and by extension, for the well. The shape factors for irregular drainage areas
were presented by Daal and Economides.
For implementation in the tool, a shape factor correlation was generated from a plot of the C A
and Yed (dimensionless aspect ratio = ye/xe) data (Figure HYD-3).
Figure HYD-3: Rectangular Shape Factors
The correlation equation for the shape factor is as follows.
YeD > 0.2: CA = -367.43YeD5 + 1227.3YeD4 - 1538.2YeD3 + 833.48YeD2 - 129.65YeD + 5.3719 (HYD-1)
YeD < 0.2: CA = 23.35YeD - 2.31 (HYD-2)
YeD = ye/xe (HYD-3)
Where ye is the drainage length (fracture spacing) and xe is the drainage length (well spacing).
Fracture spacing is defined as the length of the lateral divided by the number of fractures. Note
that a square drainage has a YeD = 1 and a CA = 30.88.
The proppant number is defined as follows for a square reservoir.
Np = Ix2CfD (HYD-4)
Where Ix is the penetration ratio, defined as:
Ix = 2xf/xe (HYD-5)
CfD is the dimensionless fracture conductivity, defined as:
CfD = kf wf / k xf (HYD-6)
This results in Np as follows:
Np = 2Vfrackf (HYD-7)
Vresk
Vfrac = PropM/(PropSG 62.4) h/hf (HYD-8)

Vres = xe ye h (HYD-9)
Where Vfrac is the volume of the fracture contained within the pay and V res is the reservoir
drainage volume, kf is the fracture permeability, k is the reservoir permeability, xf is the fracture
half-length, wf is the fracture width, xe is the drainage area length, ye is the drainage area width,
PropM is the mass of proppant injected in lbm, PropSG is the specific gravity of the proppant, h
is the reservoir thickness and hf is the total propped height of the fracture within the pay.
For rectangular, and square, reservoir drainage, Np becomes:
Np = Ix2CfD / YeD (HYD-10)
Daal and Economides presented a number of plots showing the relationship between CfD, JD and
Np for a square drainage area. Figure HYD-4 presents Figures 2 and 9 from Daal and Economides
showing the occurrence of the maximum value of CfD for each Np in a square drainage area. Daal
and Economides also presented a plot of YeD versus JDmax and Np (Figure HYD-5).
Figure HYD-4: Maximum CfD (Figure 2 and Figure 9 from Daal and Economides)
Figure HYD-5: Maximum JD (Figure 13 from Daal and Economides)

Daal and Economides extended the analysis for a square drainage area to include non-square
reservoirs and presented the correlation for the optimum value of CfD for Np greater than, or
equal to, 0.1 as follows:
CfDopt = 0.01 (100Yed - CfD,0.1) (Np - 0.1) + CfD,0.1 (HYD-11)
CfD,0.1 = 1.6 1 > Yed > 0.25

CfD,0.1 = 4.5Yed + 0.25 0.1 < Yed < 0.25
Incorporating productivity index (J) in its dimensionless form, JD. Economides, Oligney and Valko
presented the following expression for JD in a square reservoir and Daal and Economides
presented the expression for maximum JD (JDmax) for non-square reservoirs.
JD = [-0.63 – 0.5ln(Np) + 0.5ln(CfD) + f]-1 (HYD-12)
JDmax = [-0.63 – 0.5ln(Np) + Fopt]-1 (HYD-13)
Where f is the Cinco-Ley and Samaniego pseudo skin function for hydraulic fractures and Fopt is a
function to describe the optimum fracture behavior. Daal and Economides defined the function,
Fopt, as follows.
Fopt = (9.33YeD2 + 3.9YeD + 4.7) / 10YeD Np<0.1 and 0,25>YeD>0.1 (HYD-14)

Fopt = (a + buopt + cuopt2 + duopt3) / (a’ + b’uopt + c’uopt2) Np>0.1 (HYD-15)
uopt = ln(CfDopt)
a’= 10
b’ = 36
c’ = 33
The constants a, b, c, and d had been presented as a table of values dependent on YeD. For ease
of use, the constants have been plotted and correlations were generated for each constant as
follows.
For YeD<0.25:
a = 146.67YeD2 + 29.133
b = 306.67YeD2 - 398YeD + 126.33
c = -146.27YeD2 + 53.72YeD + 66.054
d = 13.28YeD - 0.127
For YeD>0.25:
a = -89.481YeD3 + 235.53YeD2 - 205.1YeD + 76.252
b = 55.704YeD3 - 141.21YeD2 + 114.74YeD + 25.27
c = -98.074YeD3 + 240.43YeD2 - 196.61YeD + 106.76
d = 16.9 YeD>0.5
d = 4.24YeD + 14.78 0.25>YeD<0.5
After CfDopt is calculated based on Np and Yed, the fracture parameters are calculated as follows.
Ixopt = (NpYeD / CfDopt)1/2 (HYD-16)
xfopt = 0.5 Ixopt xe (HYD-17)

wopt = 12 CfDopt k xfopt / kf (HYD-18)
Where xfopt is the optimum fracture half-length in ft and wfopt is the optimum fracture width in
inches.
JDmax is calculated using the values for Fopt and Np then converted to a horizontal well, JDhmax, for
an individual fracture as follows:
JDhmax = [1/JDmax + Sc]-1 (HYD-19)
Sc = kh/kfwf [ln(h/2rw) – 1.5708] (HYD-20)
Where Sc is the choked skin in the fracture and rw is the wellbore radius in ft.
With the value JDhmax and the number of fracture stages, the productivity index is calculated for
a gas well and an oil well as follows.
PIgas= 1000 Stages JDhmax k h / (1424.0 µ Z (T + 460)) (HYD-21)
PIoil= Stages JDhmax k h / (141.2 µ Bo) (HYD-22)
Where Stages are the number of hydraulic fracture stages placed in the well, PIgas is in
mmscfpd/psi2 and PIoil is in bopd/psi.
Artificial Lift Design Tool

A survey performed by World Oil and published in February 2012 indicated that 95% of active oil
wells in the US were produced by some form of artificial lift. A market research firm (Welling &
Company) publishes data on artificial lift in their “Worldwide Survey of the Market of Artificial
Lift Equipment”. Figure ALD-1 presents the percentage of artificially lifted oil wells, on land,
around the world and Figure ALD-2 presents the percentage of artificially lifted gas wells.
Figure ALD-1: Worldwide Artificial Lift Utilization – Land-Based Oil Wells
Figure ALD-2: Worldwide Artificial Lift Utilization – Land-Based Gas Wells

148 Artificial Lift Design Tool
The type of artificial lift used on the land-based wells are presented in Figure ALD-3.
Figure ALD-3: Worldwide Artificial Lift Utilization – Land Wells
(Source for Figures ALD-1 to ALD-3: Welling & Company 2017, Worldwide Survey of the Market
of Artificial Lift Equipment)
The preferred type of artificial lift used is a function of many parameters. Table ALD-1 and Figure
ALD-4 show the operating parameters for a number of artificial lift systems.
Plunger Foam Lift PCP Rod Lift Jet Pump ESP Gas Lift
Maximum 200 bpd 500 bpd 5,000 bpd 6,000 bpd 35000 bpd 60,000 bpd 75,000 bpd
Volume 32 m³/d 80 m³/d 790 m³/d 950 m³/d 5,560 m³/d 9,500 m³/d 12,000 m³/d
Maximum 19,000 ft 22,000 ft 8,600 ft 16,000 ft 20,000 ft 15,000 ft 18,000 ft
Depth 5,791 m 6,705 m 2,621 m 4,878 m 6,100 m 4,572 m 5,486 m
Maximum 550°F 400°F 250°F 550°F 550°F 482°F 450°F
Temp 288°C 204°C 121°C 288°C 288°C 250°C 232°C
Gas
Excellent Excellent Good Fair to Good Good Fair Excellent
Handling
Solids
Fair Good Excellent Fair to Good Good Sand<40ppm Good
Handling
Fluid
>15°API >8°API 8°<API<40° >8°API >8°API >6°API >15°API
Gravity
Prime
Well Energy Well Energy Gas/Electric Gas/Electric Gas/Electric Electric Compressor
Mover
System
N/A N/A 50% to 75% 45% to 60% 10% to 30% 35% to 60% 10% to 30%
Efficiency
Table ALD-1: Artificial Lift Selection Criteria
Figure ALD-4: Artificial Lift Operating Parameters
The PE² Essentials Artificial Lift Design tool (Figure ALD-5) can be used help determine artificial
lift parameters and to compare the benefits of the different artificial lift options.
Figure ALD-5: PE² Essentials Artificial Lift Design Tool
The PE² Essentials Artificial Lift Design tool includes design parameters for rod pumps, plunger
lift, hydraulic jet pumps and electrical submersible pumps (ESP). The definitive reference for
artificial lift are the series of books by Brown, K., et al, The Technology of Artificial Lift Methods,
Volumes 1, 2a, 2b and 4, PennWell Books, 1980 – 1984.
ALD.1 Rod Pumps
ALD.1.1 Overview
Sucker rod pumping is an artificial lift technique that provides mechanical energy to lift oil from
bottom hole to surface. Both “Rod Lift” and “PCP” artificial lift systems use sucker rods but only
rod lift is included in the PE² Essentials Artificial Lift Design tool.
Rod pumps can pump a well down to very low pressure by placing the pump barrel near the
perforations and thereby maximize oil production rate. This pump system is applicable to slim
holes, multiple completions, high bottom hole temperatures and viscous oils. The pump system
is easy to change to other wells with minimal cost.
The major disadvantages of rod pumping include excessive friction in crooked/deviated holes,
solids sensitivity, low efficiency in gassy wells, limited depth due to rod capacity, and bulky
surface equipment. Figure ALD-6 is a diagram of a sucker rod pumping system (ref: Golan, M. and
Whitson, C.H., Well Performance, Prentice Hall, 1991).
Figure ALD-6: Conventional Sucker Rod Pump

The polished rod and stuffing box combine to maintain a good liquid seal at the surface forcing
the fluid to flow into the Tee connection just below the stuffing box. Conventional pumping units
are available in a wide range of sizes, with stroke lengths varying from 12in (0.3m) to 200in
(5.1m). The stroke lengths are achieved by varying the position of the pitman arm connection on
the crank arm.
Walking beam ratings are expressed in allowable polished rod loads (PRL’s) and vary from
approximately 3,000lb (13,345N) to 35,000lb (155,690N). Counterbalance for conventional
pumping units is accomplished by placing counterweights directly on the beam of smaller units
or by attaching weights to the rotating crank arm for larger units.
There are two major types of pumping units: Conventional; and Mark II / Air-Balanced Units.
(Figure ALD-6 shows a conventional unit and Figure ALD-7 shows an air-balanced unit). Instead
of using counterweights, air cylinders are used in the air-balanced units to balance the torque on
the crankshaft.
Figure ALD-7: Mark II / Air Balanced Sucker Rod Pump
The American Petroleum Institute (API) has established designations for sucker rod pumping
units using a string of characters containing four fields. Table ALD-2 is an excerpt from a table of
pump information. A Lufkin pump catalogue is included in the “\Public\Artificial Lift Catalogues”
directory for reference purposes.
Table ALD-2: Pumping Unit API Designations
As an example, the API Unit Designation, C-912D-365-168, represents the following (refer to
Figure ALD-8 for labels). The first field is the code for the type of pumping unit: C is for
conventional units; A is for air-balanced units; B is for beam counterbalance units; and M is for
Mark II units. The second field (912D) is the code for peak torque rating in thousands of inch-
pounds and gear reducer information: D stands for double-reduction gear reducer. The third field
(365) is the code for PRL rating in hundreds of pounds – 36,500lb in this example. The last field
(168) is the code for maximum stroke length in inches.
Figure ALD-8: Sucker Rod Pump Labels

The information in the API Geometry table used in PE² Essentials Artificial Lift Design are the “A”
pump dimension, the “C” pump dimension and the “R” crank dimension. The “Crank-to-Pitman”
ratio is also required. This value can be calculated as R/P where “P”, the pitman arm dimension,
is obtained from the geometry table. For the example used above the Crank-to-Pitman ratio
would be 47/148.5 or 0.3165. This ratio is used to determine maximum pump speed.
One factor in rod pump design is the “Maximum Allowable Acceleration Factor”. There is a
limiting relationship between stroke length and cycles per minute. The maximum value of the
downward acceleration occurs at the top of the pump stroke. If the maximum acceleration
divided by g (gravitational acceleration) exceeds one, then the downward acceleration is greater
than the free-fall acceleration of the rods at the top of the stroke. This leads to severe pounding
when the polished rod shoulder falls onto the hanger and leads to failure of the rod at the
shoulder. As a result, a Maximum Allowable Acceleration Factor (the downward acceleration
divided by g) is normally limited to approximately 0.5 or to a value determined by experience in
a particular field.
Volumetric efficiency of the plunger is dependent on the rate of oil slippage past the pump
plunger and the solution–gas ratio at pump conditions. Metal-to-metal plungers are commonly
available with plunger-to-barrel clearance -0.001, -0.002, -0.003, -0.004, and -0.005. For example,
the -0.001 means the plunger’s outside diameter is 0.001 inches smaller than the barrel’s inside
diameter. For low viscosity oils (<20 cp), a plunger-to-barrel clearance 0.001 inches can be used.
For high viscosity oils, the clearance can be increased and if sand/solids are expected a plunger-
to-barrel clearance 0.005 inches can be used.
A pump’s volumetric efficiency is mainly affected by the slippage of oil and the free gas volume
below the plunger. Both effects are difficult to quantify and pump efficiency can vary over a wide
range but are commonly around 70–80%.
The procedure and equations for rod pump design are presented in Section ALD.1.2. The
derivations of the equations are not presented here but can be found in the references.
ALD.1.2 Rod Pump Design

The pump design procedure is as follows:
Step 1 - Calculate the weight of the fluid, Wf:
DAp
W f = S f (62.4) (ALD-1)
144
Sf = (1-WCut/100) * 141.5/(131.5+OilAPI) + WCut/100 * Swater
Ap = 0.25 π ODp2
Where: Wf is the weight of the fluid in lbs, Sf is the specific gravity of the produced fluid, D is the
pump setting depth in ft, Ap is the cross-sectional area of the plunger in in2, WCut is the water
cut in %, OilAPI is oil gravity in °API, Swater is the specific gravity of the produced water, and OD p
is the diameter of the plunger in inches.
Step 2 - Calculate the weight of the rods, Wr:

 s DAr
Wr = (ALD-2)
144
Ar = 0.25 π ODr2
Where: Wr is the weight of the rods in lbs, γs is the specific weight of the steel rod (=400lb/ft3), D
is the pump setting depth in ft, Ar is the cross sectional-area of the rods in in2, and IDr is the
diameter of the rods in inches. Note for a tapered string, the value for OD r is calculated as the
depth-weighted average of the rod diameters.
Step 3 - Calculate the actual downhole plunger stroke, Sp:
12D   1 1  SN 2 M Wr 
Sp = S − W f  +  −  (ALD-3)
E   Ar At  70500 Ar 
E = 30 x 106 lb/in2
At = 0.25 π (ODt2 – IDt2)
M = 1 + Unit * R/P
Where: Sp is the stroke length of the plunger in inches, S is the stroke length of the polished rod
in inches, D is the pump setting depth in ft, E is the modulus of elasticity of steel in lb/in 2, Wf is
the weight of the fluid in lbs, Ar is the cross sectional-area of the rods in in2, At is the cross
sectional-area of the tubing in in2 (note: this is set to 0 for an anchored tubing), N is pump speed
in SPM, M is the pittman ratio, Wr is the weight of the rods in lbs, Ar is the cross sectional-area of
the rods in in2, ODt is outer diameter of the tubing in inches, IDt is inner diameter of the tubing
in inches, Unit is +1 for a conventional pump and -1 for a Mark II / air balanced pump, R is the
crank dimension in inches, and P is pitman arm dimension in inches.
Step 4 - Calculate the production rate, q:
Ap NS p Ev
q = 0.1484 (ALD-4)
Bo
Where: q is the fluid production rate in stb/d, Ap is the cross-sectional area of the plunger in in2,
N is the pump speed in SPM, Sp is the stroke length of the plunger in inches, Ev is the pump
volumetric efficiency, and Bo is the oil formation factor (Bt is used for oil/water production).
Step 5 - Calculate the required prime mover power, Ppm:
Ppm = Fs ( Ph + Pf ) (ALD-5)
Ph = 7.36 10 −6 qS f LN
ptf
LN = H +
0.433S f
Pf = 6.3110 −7 Wr SN
Where: Ppm is the required prime mover (surface) power in hp, Fs is the safety factor (1.25 – 1.5),
Ph is the hydraulic power required to lift the fluid in hp, Pf is the power required to overcome the
friction in the system in hp, q is the production rate in stb/d, S f is the specific gravity of the
produced fluid, LN is the net lift in ft, H is the depth to the average fluid level in the annulus, Ptf is
the flowing tubing head pressure in psi, Wr is the weight of the rods in lbs, S is the stroke length
of the polished rod in inches, and N is the pump speed in SPM
Step 6 – Calculate the maximum polished rod load, PRLmax:
PRL max = W f − S f (62.4)

Wr
+ Wr + Wr F1 (ALD-6)
s
S m N m (1 + Unit RP )
2
F1 =
70500
A
S m = 2R
C
70500L
Nm =
S m (1 − RP )
Where: PRLmax is the maximum polished rod load in lbs, Wf is the fluid weight in lbs, Sf is the
specific gravity of the produced fluid, Wr is the weight of the rods in lbs, γs is the specific weight
of the steel rod (=400lb/ft3), F1 is the maximum upward acceleration factor, Sm is the maximum
stroke length of the polished rod in inches, Nm is the maximum allowable pump speed in SPM,
Unit is +1 for a conventional pump and -1 for a Lufkin II / air balanced pump, R is the crank
dimension in inches, P is pitman arm dimension in inches, A is the API pump dimension in inches
and C is the API pump dimension in inches and L is the maximum allowable acceleration factor.
Step 7 – Calculate the minimum polished rod load, PRLmin:
PRL min = − S f (62.4)

Wr
+ Wr − Wr F2 (ALD-7)
s
S N (1 − Unit RP )
2
F2 = m m
70500
Where: PRLmin is the minimum polished rod load in lbs, Sf is the specific gravity of the produced
fluid, Wr is the weight of the rods in lbs, γs is the specific weight of the steel rod (=400lb/ft3), F2 is
the minimum upward acceleration factor, Sm is the maximum stroke length of the polished rod
in inches, Nm is the maximum allowable pump speed in SPM, Unit is +1 for a conventional pump
and -1 for a Lufkin / air balanced pump, R is the crank dimension in inches, and P is pitman arm
dimension in inches.
Step 8 – Calculate the counterweight requirements, CW:

CW = 1
2
(PRL max + PRL min ) (ALD-8)
Where: CW is the recommended counterweight in lbs, PRLmax is the maximum polished rod load
in lbs, PRLmin is the minimum polished rod load in lbs.
ALD.1.3 Rod Pump Example

Table ALD-3 and Figure ALD-9 present an example of a rod pump design.
This example uses a C-320D-213-86 pump. The API specs for this pump are as follows: A = 111in,
C = 96.05in, R = 37in and P = 114in. From experience, a maximum acceleration factor of 0.4 was
used.
The design input parameters are as follows:
Reservoir Pressure = 4500psi Tubing OD = 3.5in Conventional Pump

Reservoir Temperature = 200°F Tubing ID = 2.992in Pump Volumetric Efficiency = 0.8
Oil Density = 45°API Pump Setting Depth = 3500ft Safety Factor for Prime Mover = 1.35
Bubble Point Pressure = 500psi Annulus Liquid Depth = 3500ft API Pump Dimension A = 111in
Gas Gravity = 0.6 No Tubing Anchor API Pump Dimension C = 96.05in
Water Salinity = 35,000ppm Plunger Diameter = 2.25in API Pump Dimension R = 37in
Solution GOR = 87scf/bbl Rod Diameter = .875in API Pump Dimension P = 114in
Bo = 1.134 Rod Length = 3500ft Maximum Allowable Acceleration = 0.4
Water Cut = 0% Polish Rod Stroke = 86in
Tubing Head Pressure = 100psi Pumping Speed = 18
Table ALD-3: Sucker Rod Pump Design – Input Data
Figure ALD-9: Sucker Rod Pump Design - Example
The design calculations and results are as follows:
Step 1: Ap = 3.976, Sf= 0.802, Wf = 4834.6

Step 2: Ar= 0.601, Wr = 7161.6
Step 3: M = 1.3246, At = 2.59, Sp = 80.86
Step 4: q = 605.9 blpd
Step 5: LN = 3788.1, Pf = 6.995 hp, Ph = 15.36 hp, Ppm = 30.18 hp
Step 6: Sm = 85.52, Nm = 17.7, F1 = 0.502, PRLmax = 14,860.6 lbs
Step 7: F2 = 1.35, PRLmin = 4,597.0 lbs
Step 8: CW = 9,728.8 lbs
ALD.2 Plunger Lift
ALD.2.1 Overview
Plunger lift systems are applicable to high gas–liquid ratio (GLR) wells. As an added benefit, the
plunger automatically keeps the tubing clean of paraffin and scale. Plunger lift systems (Figure
ALD-10) are good for low-rate liquid wells producing less than 200 bpd and are commonly used
to lift water and condensate from gas wells.
Figure ALD-10: Plunger Lift Pumping System
The main advantages of plunger lift are that it requires no external energy source, a rig is not
required for installation, it is low cost, can be used in gas wells, works in deviated wells and can
produce wells to depletion. Disadvantages include the required GLR’s, it is a low volume pump
and it cannot handle solids.
The design techniques for a plunger lift system are referenced from Foss, D. L. and Gaul, R. B.:
“Plunger Lift Performance Criteria with Operating Experience - Ventura Field, “Drilling and
Production Practice, API (1965), 124-140 and Mower, L.N; Lea, J.F., Beauregard, E., and Ferguson,
P.L.: “Defining the Characteristics and Performance of Gas-Lift Plungers”, SPE Paper 14344.
Originally, plunger lift was used to produce oil wells. Currently, plunger lift has become more
common in gas wells for de-watering purposes. As shown in Figure ALD-11 (Source Khamehchi,
E; Khishvand, K; Abdolhosseini, H: A case study to optimum selection of deliquification method
for gas condensate well design: South Pars gas field) high-pressure gas wells may produce gas,
water and/or condensate in the form of mist. As the gas flow velocity reduces, the carrying
capacity of the gas decreases and the liquid builds up in the bottom of the well.
Figure ALD-11: Liquid Loading of a Gas Well
When the gas velocity drops to a critical level, liquid begins to accumulate in the well and the
flow will change to a slug flow regime. The accumulation of liquids (liquid loading) increases
bottom hole pressure which further reduces gas production rate. Low gas production rate results
in lower gas velocity and eventually the well may stop producing as the liquid level rises in the
well.
The purpose of plunger lift is to remove liquids from the wellbore allowing the well to be
produced at a low bottom hole pressure. The plunger is basically a length of steel and is dropped
down the tubing to the bottom of the well and allowed to travel back to the surface. The plunger
provides a piston-like interface between liquids and gas in the wellbore. The well’s energy is used
to lift the plunger and the liquids out of the wellbore.
There are two main requirements for plunger lift operation: a minimum GLR and sufficient well
pressure. For the plunger lift to operate, there must be a sufficient quantity of gas per barrel of
liquid for a given well depth. The gas in the tubing-casing annulus is used as the source of gas. It
is possible to augment the gas supply and increase casing pressure by injecting additional gas into
the annulus.
The Foss and Gaul design model was originally designed for oil well operations that assumed the
well would be shut-in after plunger arrival. In general, this model overpredicts required casing
pressure. If a well meets the Foss and Gaul criteria, it is almost certainly a candidate for plunger
lift.
One of the required input items for the plunger lift design model is the fall times of the plunger
in both liquid and gas. Table ALD-4 is an example of a manufacturer’s recommended fall times (A
PCS Ferguson catalogue is included in the “\Public\Artificial Lift Catalogues” directory).
Fall Times
Plungers, Weatherford
(ft/min)
39-150 Pad / Conventional Plungers in Fluid
150-400 Pad / Conventional Plungers in Gas
1500-1800 Continuous Flow In Gas, Shut in
600-1000 Continuous Flow against flow
Table ALD-4: Typical Fall Times for Plungers
ALD.2.2 Plunger Lift Design

The plunger lift design procedure is as follows:
Step 1 – For a given slug, calculate the minimum pressure required to lift the plunger:
D
PC min = [ Pp + 14.7 + Ptf . + ( Plh + Plf )Vslug ](1 + ) (ALD-9)
K
P p = W p / At
At = 0.25 π IDt2
Where: PCmin is minimum casing pressure in psi, Pp is the plunger weight pressure in psi, Ptf is the
flowing tubing head pressure in psi, Plh is the hydrostatic liquid gradient of the slug in psi/bbl, P lf
is the flowing friction gradient of the slug in psi/bbl, Vslug is the volume of the slug in bbls, D is the
depth to the plunger in ft, K is a gas friction factor constant in feet, Wp is the weight of the plunger
in lbs, At is the internal cross sectional area of the tubing in in2, IDt is the internal diameter of the
tubing in inches.
Foss and Gaul suggested an approximation where K and Plh + Plf are constant for a given tubing
size and a plunger velocity of 1,000 ft/min. Table ALD-5 presents the Foss and Gaul values.
Tubing Plh +Plf
K (ft)
Size (in) (psi/bbl)
2.375 33,500 165
2.875 45,000 102
3.5 57,600 63
Table ALD.4-5: Foss and Gaul Parameters
The PE² Essentials plunger lift design tool calculates the value for Plh, Plf and K as follows:
Plh = 0.433Ls S f
2
fricl S f LsVrise
Plf =
44585IDt
fric l = 0.0056 + 0.5(25.794 S f IDtVrise ) −0.32

875292.6 IDt (Tavg − 460)
K= 2
fric g S g Vrise
fric g = 0.0056 + 0.5(135972 S g IDtVrise /(Tavg + 460)) −0.32
Where: Plh is the tubing hydrostatic liquid gradient of the slug in psi/bbl, L s is the slug length per
barrel in the tubing in feet, Sf is the specific gravity of the produced fluid, WCut is the water cut
in %, Swater is specific gravity of the water, Plf is the flowing friction gradient of the slug in psi/bbl,
fricl is the liquid friction factor, IDt is the internal diameter of the tubing in inches, Vsise is the rising
velocity of the plunger in ft/min, K is the gas friction factor constant in feet, Tavg is the average
wellbore temperature in °F, fricg is the gas friction factor and Sg is the gas gravity.
Step 2 - Calculate the maximum pressure that the casing must reach:
 A + At 
Pc max = Pc min  a  (ALD-10)
 Aa 
Aa = 0.25 π (ODt2 – IDc2)
Where: PCmax is maximum pressure must reach to operate the system in psi, PCmin is minimum
casing pressure when the plunger reaches the surface in psi, A a is casing-tubing annular cross
section in in2, At is the inner cross section of the tubing in in2, ODt is the outer diameter of the
tubing in inches, IDc is the internal diameter of the casing in inches, and IDt is the internal
diameter of the tubing in inches.
Step 3 - Calculate the minimum required GLR to lift the slug and the plunger:
GLRmin = 1000 Vg / Vslug (ALD-11)

37.14Fgs PCavgVt
Vg =
Z (Tavg + 460)
FGS = 1 + 0.02 D / 1000
 A 
Pcavg = Pc min 1 + t 
 2 Aa 
Vt = At (D − Vslug Lt )
Lt = 808.56 / At
Where: GLRmin is minimum gas-liquid ratio required to lift the slug and plunger in scf/bbl, Vg is
the gas volume required to lift the slug in mscf, Vslug is the slug volume to be lifted in bbl, Fgs is
the gas slippage factor, PCavg is average casing annulus pressure in psi, Vt is the gas volume in the
tubing in mcf, Z is the gas deviation factor, Tavg is the average wellbore temperature in °F, D is the
depth to the plunger in ft, PCmin is minimum casing pressure in psi, At is the inner cross section of
the tubing in in2, Aa is casing-tubing annular cross section in in2, and Lt is the tubing capacity in
ft/bbl.
Step 4 - Calculate the cycles per day and maximum rate:
1440
N C max = (ALD-12)
D D − Vslug Lt /(1 − Fg ) Vslug Lt
+ +
Vr V fg (1 − Fg )V fl
qmax = NCmac Vslug (ALD-13)
Where: NCmac is the maximum possible cycles per day, D is the depth to the plunger in ft, V slug is
the volume of the slug to be lifted in bbl, Lt is the tubing capacity in ft/bbl, Fg is the fraction of gas
in the slug (commonly set at 0.8), Vr is the rising velocity of the plunger in ft/min, Vfg is the plunger
fall velocity in gas in ft/min, Vfl is the plunger fall velocity in liquid in ft/min and qmax is the
maximum liquid rate in bbl/d.
Step 5 - Calculate plunger fall time for a given slug:
Lh = Lt Vslug / (1 – Fg) (ALD-14)

Ftt = Fact [(D – Lh) / Ftg + Lh / Ftl]
Where: Lh is height of the gassy liquid in the tubing in ft, Lt is the tubing capacity in ft/bbl, Vslug is
the slug volume to be lifted in bbl, Fg is the fraction of gas in the slug (commonly set at 0.8), FT t
is the total plunger fall time in minutes, Fact is a calibration factor, FTg is plunger fall time in gas
in minutes, and FTl is the plunger fall time in the gassy liquid in minutes.
ALD.2.3 Plunger Lift Example

Figure ALD-12 and Table ALD-6 show the input/output for a plunger lift design example. This is a
gas well that produces water. A rising velocity for the plunger of 1000 ft/min was assumed for
this example. The calculations are presented for a 1-barrel slug.
Figure ALD-12: Plunger Lift Pump Design - Example
The design input parameters are as follows:
Reservoir Pressure = 4000psi Producing GLR = 8000scf/bbl Tubing ID = 1.995in

Reservoir Temperature = 180°F Water Cut = 100% Tubing OD = 2.375in
Oil Density = 45°API Liquid Rate = 10bbl/d Casing ID = 4.56in
Bubble Point Pressure = 1500psi Tubing Head Pressure = 100psi Plunger Weight = 10lb
Gas Gravity = 0.7 Tubing Head Temperature = 100°F Fall Velocity in Gas = 750ft/min
Gas Z Factor = 0.9519 Pump Setting Depth = 7000ft Fall Velocity in Liquid = 750ft/min
Water Density = 1.025 Maximum Available CHP = 800psi Design Rise Velocity = 750ft/min
Table ALD-6: Plunger Lift Design- Input Data
The design calculations and results for a 1 barrel slug are as follows:
Step 1: At = 3.126, Pp = 3.199, fricg = 0.0143, fricl = 0.0210, K = 104764,

Plh = 114.8, Plf = 62.6, PCmin = 315.1 psi
Step 2: Aa = 11.901, PCmax = 397.8 psi
Step 3: Lt = 258.6, Vt = 0.146, PCavg = 356.4 psi, FGS = 1.14,
Vg = 3.867, GLR = 3866.7 scf/bbl
Step 4: NCmax = 61.98 cycles/d, qmax = 61.98 bbl/d
Step 5: Lh = 1293.3, Ftt = 16.2 min
ALD.3 Hydraulic Jet Pump
ALD.3.1 Overview
A hydraulic jet pump converts the energy in an injected power fluid (water or oil) to pressure that
lifts production fluids. The main advantages of a hydraulic jet pump are that it has no moving
parts so solids and gassy fluids present no problem to the pump, it can handle high rates, works
in deviated wells, multiple wells can be run from the same surface equipment and it is a low
maintenance pump. Disadvantages are that it has low efficiency (10–30%), and it requires high
pressure surface equipment.
A hydraulic jet pump is a dynamic displacement pump that increases the pressure in the
produced fluids through the use of a jet nozzle (Figure ALD-13 – from Volume 4 of Technology of
Artificial Lift Methods).
Figure ALD-13: Hydraulic Jet Pump

The power fluid enters the top of the pump through an injection tubing. The power fluid
accelerates through the nozzle and is mixed with the well’s produced fluid in the throat of the
pump. As the fluids mix, some of the momentum of the power fluid is transferred to the produced
fluid and thereby increases its kinetic energy (velocity head) resulting in a pressure increase in
the produced fluid. The combined fluid stream enters the casing-tubing annulus and is produced
to surface.
Selection of a jet pump is made based on a number of dimensionless variables (refer to Figure
ALD-13).
R = A j / At
M = q 3 / q1
H = (P2 – P3) / (P1 – P2)
η=MH
Where: R is the dimensionless nozzle area, M is the dimensionless flow rate, H is the
dimensionless head, η is the pump efficiency, Aj is the jet pump nozzle area in in2, At is the jet
pump throat area in in2, q3 is the well fluid rate in bbl/d, q1 is the power fluid rate in bbl/d, P2 is
the available discharge pressure from the pump in psi, P3 is the pressure at the pump inlet in psi,
and P1 is the required power fluid pressure in psi.
Table ALD-7 presents one manufacture’s selection of jet pumps.
Table ALD-7: Jet Pump Specifications
A Weatherford catalogue is included in the “\Public\Artificial Lift Catalogues” directory for

reference purposes. Pump selection is based on the pump manufacturer’s performance curves,
Figure ALD-14. The performance curve in Figure ALD-14 shows the effect of M on H and η.
Figure ALD-14: Jet Pump Performance Chart (ref: Vol. 4 of Technology of Artificial Lift Methods)
For a given jet pump specified by an R value, there exists a peak efficiency η p at a peak Hp and
Mp. For example, assuming a pump with an R value of 0.262, the peak efficiency, ηp, of 0.255 will
occur at an Mp value of 0.9 and Hp value of 0.28 (from the ‘C’ curves in Figure ALD-14).
It is good field practice to attempt to operate the pump at its peak efficiency. If Mp and Hp are
used to denote M and H at the peak efficiency, respectively, pump parameters should be
designed using these parameters.
ALD.3.2 Hydraulic Jet Pump Design

Note that the PE² Essentials Artificial Lift Design tool uses the Vogel IPR equation to calculate the
bottom hole flowing pressure for the hydraulic jet pump as well as for the electrical submersible
pump.
The hydraulic jet pump design procedure is as follows:
Step 1 - Calculate flowing bottom hole pressure and pump intake pressure:

Pwf = 0.125Pr 81 − 80(q / qmax ) − 1 (ALD-15)
P3 = Pwf − 0.433S l (TD − D) (ALD-16)
Where: Pwf is the bottom hole flowing pressure in psi, Pr is the reservoir pressure in psi, q is the
flow rate in bbls/d, qmax is Vogel IPR maximum rate in bbls/d, P3 is the pressure at the pump inlet
in psi, TD is the depth of the reservoir in feet, and D is the depth to the jet pump in feet.
Step 2 - Calculate pump fluid rates:
q1 = q3 Bo / Mp (ALD-17)
q2 = q3 Bo + q1 (ALD-18)
Where: q1 is the power fluid rate in bbl/d, q3 is the well production rate in bbl/d, Bo is the oil
formation factor (Bt is used for oil/water production), Mp is dimensionless rate at peak efficiency,
and q2 is the total return flow rate in bbl/d
Step 3 - Calculate system pressures:
P1 = P3 + S f (q1 /(1214.5 A j )) 2
P2 = ( P3 + H p P1 ) /(1 + H p )
Hp = Mp ηp
Where: P1 is the required power fluid pressure in psi, P3 is the pressure at the pump inlet in psi,
Sf is the specific gravity of the produced fluid, q1 is the power fluid rate in bbl/d, Aj is the jet pump
nozzle area in in2, P2 is the available discharge pressure from the pump in psi, Hp is the
dimensionless head at peak efficiency, Mp is the dimensionless rate at peak efficiency, and ηp is
the peak pump efficiency.
Step 4 - Calculate required pump power:
Power = 1.7e-5 q1 Ps (ALD-19)
Where: Power is the required pump power in hp, q1 is the power fluid rate in bbl/d, and Ps is the
required surface pump operating pressure calculated from the pump discharge pressure using
tubing pressure drop correlations in psi.
ALD.3.3 Hydraulic Jet Pump Example

Table ALD-8 and Figure ALD-15 show the input/output for a hydraulic jet pump design example.
Reservoir Pressure = 4000psi Producing GLR = 268.2scf/bbl Tubing ID = 1.992in

Reservoir Temperature = 200°F Water Cut = 0% Tubing OD = 2.375in
Oil Density = 40°API Liquid Rate = 758bbl/d Casing ID = 5in
Bubble Point Pressure = 1500psi Tubing Head Pressure = 100psi Pump Area Ratio, R = 0.262
Gas Gravity = 0.6 Tubing Head Temperature = 100°F Jet Nozzle Area = 0.16in2
Bo = 1.207 Pump Setting Depth = 9700ft Pump Peak Efficiency, ηp = 0.255
Water Density = 1.025 Max Rate for Vogel IPR = 8000 stb/d Rate at Peak Efficiency, M p = 0.9
Reservoir Depth = 10000ft Power Fluid Specific Gravity = 0.8
Table ALD-8: Hydraulic Jet Pump Design – Input Data
Figure ALD-15: Jet Pump Design - Example
Step 1: Pwf = 3784 psi, P3 = 3677.1 psi

Step 2: q1 = 1016.8 bbl/d, q2 = 1931.9 bbl/d
Step 3: HP = 0.283, Sf = 0.825, P1 = 3699.0 psi, P2 = 3681.9 psi
Step 4: CHP = 522.0 psi (using Hagedorn-Brown from P2),
Ps = 805 psi (using Hagedorn-Brown from P1), Power = 13.9 hp
ALD.4 Electrical Submersible Pump
ALD.4.1 Overview
Electrical submersible pumps (ESP’s) can lift extremely high volumes of fluid from highly
productive oil reservoirs. ESP’s tend to be used in offshore operations but are also used onshore.
ESP systems can deliver higher horsepower then other lift systems, they can operate in hotter
applications, can be installed as a dual pump installation, and may include down-hole oil/water
separation. New pump designs tend to be more tolerant of sand and gas production. Automation
of the ESP system includes monitoring, analysis, and control.
Limitations to ESP applications include the requirement for high voltage electricity, they are not
suitable for deep, high-temperature reservoirs, gas and solids production tend to cause issues,
and they are costly to install and repair.
Figure ALD-16 is a schematic of an ESP installation.
Figure ALD-16: Electrical Submersible Pump (ESP)

The ESP system operates like any electric pump commonly used in other applications. Electrical
energy is transported to the down-hole electric motor through electric cables. These electric
cables are attached to the side of the production tubing. The downhole electric motor drives the
pump and the pump imparts energy to the fluid in the form of hydraulic power, which lifts the
fluid to surface.
ESPs are pumps made of centrifugal pump stages. The number of stages required is determined
by the volumetric flow rate and the lift (height) required. The length of a pump module can range
from 40 to 344 inches in length. Voltage requirements can range from 420 to 4,200 V.
ESPs can operate over a wide range of conditions; at depths over 12,000 ft and volumetric flow
rates of up to 45,000 bbl/day. Some operating conditions can limit ESP applications, including:
free gas production, downhole temperature, fluid viscosity, and solids content.
ESP design incorporates pump performances curves published by the pump manufacturer (Figure
ALD-17). A Haliburton ESP catalogue is included in the “\Public\Artificial Lift Catalogues” directory
for reference purposes. Selection of a specific pump involves choosing a pump of the largest
possible diameter that can be run in the well. The required pump rate should be within the
recommended operating range of the pump and close to its peak efficiency.
Figure ALD-17: ESP Performance Plot for a 100-Stage Pump
PE² Essentials ESP design is based on the head generated per stage and the power required per
stage of the selected pump.
ALD.4.2 Electrical Submersible Pump Design

Note that the PE² Essentials Artificial Lift Design tool uses the Vogel IPR equation to calculate the
bottom hole flowing pressure for the hydraulic jet pump as well as for the electrical submersible
pump.
The ESP design procedure is as follows:
Step 1 - Calculate flowing bottom hole pressure and pump intake pressure:

Pwf = 0.125Pr 81 − 80(q / qmax ) − 1  (ALD-20)
Pint = Pwf − 0.433Sl (TD − D)
Where: Pwf is the bottom hole flowing pressure in psi, Pr is the reservoir pressure in psi, q is the
flow rate in bbls/d, qmax is Vogel IPR maximum rate in bbls/d, Pint is the pressure at the ESP inlet
in psi, TD is the depth of the reservoir in feet, and D is the depth to the ESP in feet.
Step 2 - Calculate pump parameters:
Dmin = D − ( Pwf − Ps ) / 0.433S f (ALD-21)

ΔP = PD - Pint
Head = ΔP / 0.433 (ALD-22)
Ns = Head / Hs
Power = Ns Powers (ALD-23)
Where: Dmin is the minimum setting depth for the ESP in feet, D is the depth to the ESP in feet,
Pwf is the bottom hole flowing pressure in psi, Ps is the reservoir pressure in psi, Sf is the specific
gravity of the produced fluid, ΔP is the required ESP pressure differential in psi, PD is the required
discharge pressure at the pump based on THP (calculated by tubing pressure drop correlations)
in psi, Pint is the pressure at the ESP inlet in psi, Head is the required head to be generated by the
pump in feet, NS is the required number of stages, HS is the head per stage for the selected pump
in feet, Power is the required pump power in hp, and Powers is power per stage for the selected
pump in hp.
ALD.4.3 Electrical Submersible Pump Example

Table ALD-9 and Figure ALD-18 show the input/output for an ESP design example.
Reservoir Pressure = 4350psi Reservoir Depth = 10000ft Tubing ID = 2.992in

Reservoir Temperature = 200°F Producing GLR = 394.3scf/bbl Pump Setting Depth = 9800ft
Oil Density = 32°API Water Cut = 25% Minimum Pump Suction = 200psi
Bubble Point Pressure = 2650psi Liquid Rate = 8000bbl/d Capacity of Pump = 10000bbl/d
Gas Gravity = 0.6 Tubing Head Pressure = 100psi Pump Head per Stage = 60ft
Bo = 1.252 Tubing Head Temperature = 100°F Pump Power per Stage = 6hp
Water Density = 1.032 Max Rate for Vogel IPR = 15000 stb/d Pump Efficiency = 0.72
Table ALD-9: ESP Design – Input Data
Figure ALD-18: ESP Design - Example
Step 1: Pwf = 2822.8 psi, Pint = 2744.4 psi

Step 2: Dmin = 3309.2 ft, Sf = 0.905, PD = 3525.6 psi (using M-HB for BHP)
ΔP = 781.1 psi, Head = 1804.0 ft, Ns = 30.1, Power = 180.4 hp
Pressure Transient Analysis Tool

The PE² Essentials ‘Pressure Transient Analysis’ tool is comprised of a basic pressure buildup
analysis module (Figure PTA-1) and an analytical test simulator (Figure PTA-2).
Figure PTA-1: PE² Essentials Pressure Transient Analysis Tool
There are numerous reference texts available on the subject of well test analysis. The one
reference the author uses the most often is the book by John Lee - Lee, W.J., Well Testing, SPE
Textbook Series Vol. 1, 1982, SPE. This reference contains concise information and includes a full
list of the metric versions of all the equations presented in the book. This book was update in
2003 (Lee, J,; Rollins, J.; and Spivey, J. Pressure Transient Testing, SPE Textbook Series Vol. 9,
2003, SPE) but the 1982 edition is still the most useful if you have to deal with US oilfield and
metric units.
A secondary reference, and the main reference for the Analytical Well Test Simulator, is
Streltsova, T.D., Well Testing in Heterogeneous Formations, John Wiley & Sons, 1988.
Note: only build-up test analysis is included in the PE² Essentials Pressure Transient Analysis tool.
174 Pressure Transient Analysis Tool
Figure PTA-2: PE² Essentials Analytical Well Test Simulator
Pressure drawdown and buildup tests provide an opportunity to obtain estimates of the
following well and reservoir properties:
• Permeability to the produced phase (oil, gas, or water) - the average value within the
radius of investigation achieved during the test
• Skin factor - a dimensionless measurement of the damage or stimulation done to the well
• Current average pressure within the drainage area of the well
• Indication of flow barriers (such as faults) in the reservoir
Flow tests can be useful when the reservoir is at uniform pressure, such as when a new well is
completed or when a well has been shut in for a lengthy period. Flow tests are appropriate when
a well must continue to produce revenue even though a test is needed. Analysis of flow tests is
simplest when the rate is fairly constant.
Buildup tests are appropriate at virtually any time in the life of a well because they simply require
that the well be shut in. Buildup tests have the advantage that the zero rate is much more easily
controlled than a “constant rate” flow test. For this reason, buildup tests are the preferred type
of pressure transient test.
PTA.1 Pressure Buildup Analysis

PE² Essentials build-up analysis is focused on a semi-logarithmic plot of test data, with the slope
of the straight line on the plot used to determine permeability. Figure PTA.3 is a typical semi-log
Horner plot of the buildup test data.
To generate an interpretation, the straight line on the semi-log Horner plot is moved by using the
mouse to click on and move the end of the line, to generate a best-fit through the points. When
the mouse is released, the equivalent of the semi-log straight line is placed on the log-log plot to
confirm the interpretation. When the lines are acceptable on both plots, the interpretation is
completed.
Figure PTA-3: Semi-Log / Horner Buildup Analysis Plot
The “correct” semi-log straight line used for basic analysis is indicated on the figure; the line can
be identified with the help of log-log plots (Figure PTA-4).
Figure PTA-4: Log- Log Diagnostic Plot

The log-log diagnostic plot includes a “derivative” curve. The numerical derivative at a point is
determined by finding a weighted average of the slopes preceding the point and following the
point, as illustrated in Figure PTA-5.
Figure PTA-5: Numerical Derivative at Point X
In Figure PTA-5, the parameter L defines the minimum range of the preceding and following
points. This is incorporated to smooth out "noise" in the neighborhood of the central, X, point. L
is defined as a Δ(ln(Δt)) for a well test. Experience suggests that a value of L = 0.1 to 0.3 is usually
a satisfactory compromise between being so far from the central point that detail is lost and
being so near the central point that a great amount of noise is retained in the data. The PE²
Essentials build-up analysis tool has a built in smoothing factor of 0.1.
For a Horner plot, shut-in bottomhole pressure is plotted versus the logarithm of the ratio of
producing time, tp, plus shut-in time, Δt, to shut-in time (called Horner time). Simple equations
are used to estimate permeability and skin factor once the correct semi-log straight line is
identified and its slope, m, is determined. These equations apply to both drawdown and buildup
tests. The following build-up analysis equations are used for oil wells:
162.6qµB
k= (PTA.1)
mh
 ( P1hr - Pwf )  k  
S = 1.151 − log  + 3.23

(PTA.2)
 
 ct rw
2
 m  
Ps = 0.87mS (PTA.3)
kt
rinv =
948ct (PTA.4)
Where: k is effective permeability to the produced phase in md, q is oil flow rate in stbbl/d, B is
the formation volume factor in rbbl/stbbl, m is the slope of the semi-log straight line in psi/cycle,
h is the net pay thickness in feet, S is the skin factor (dimensionless), P1hr is the theoretical change
in pressure after 1 hour of shut-in taken from the straight line fit in psi, Pwf is the final flowing
pressure in psi, ф is the porosity (fraction), μ is the fluid viscosity in cp, ct is the total
compressibility of the formation and its fluids in 1/psi, rw is the wellbore radius in feet, ΔPs is the
pressure drop caused by the skin in psi, rinv is the radius of investigation for time t in feet, and t
is time in hours.
For gas well test analysis, the equations are similar except for the use of pseudo pressure, ψ,
instead of pressure, as follows:
1632qT
k= (PTA.5)
mh
(Ψ1ℎ𝑟−Ψ𝑤𝑓) 𝑘
𝑆 = 1.151 ( 𝑚
− 𝑙𝑜𝑔 (𝜑𝜇𝑐 𝑟 2 ) + 3.23) (PTA.6)
𝑡 𝑤
Where: q is gas flow rate in mscf/d, T is the formation temperature in °R, m is the slope of the
semi-log straight line in psi2/cp/cycle.
Extrapolation of the pressure on a semi-log plot to a Horner time of 1 yields P* which provides an
estimate of original reservoir pressure in a new well or “false” pressure, which serves as the basis
for determining current drainage area pressure, Pr.
The skin factor is an indicator of the damage, or stimulation, existing in the wellbore. Following
calculation of the skin factor, the pressure drop caused by the skin can then be determined.
Finally, a radius of investigation can be determined for a given time. This radius can be used to
estimate minimum fluids in place in the reservoir.
For a well test in which rates were not constant prior to the build-up, an effective production
time for the Horner plot is calculated as follows:
24 N p
tp = (PTA.7)
qlast
Where: tp is the corrected producing time, Np is total production during the flow period in bbls,
and qlast is the final flow rate before shut-in in bbls/d.
Consider the following example of a build-up test following a multiple rate flow period. Table
PTA-1 presents the flow and pressure build-up data.
Rate Period Np Shut-in Pressure Horner
(bbls/d) (hours) (bbls) (hours) (psi) Time
200 36 300 0 1384 -
0 12 0 2 1530 121
100 48 200 3 1535 81
125 144 750 4 1538 61
5 1540 49
8 1546 31
10 1549 25
12 1551 21
19 1556 13.6
24 1559 11
36 1563 7.67
Table PTA-1: Example Build-up Test

From Equation PTA.7, the equivalent tp was 240 hrs. This value was used to calculate the Horner
Time in Table PTA-1. Figure PTA-6 is the semi-log plot of the build-up data.
Figure PTA-6: Semi-Log Plot of Data in Table PTA-1
Assuming the following parameters:
Horner Slope, m = 27.3 psi/cycle

P1hr = 1522 psi
μ = 0.8 cp
h = 15 ft
ct = 1.5x10-5/psi
ф = 0.25
rw = 0.333 ft
B = 1.25 rbbl/stbbl
Analysis yields the following:
162.6qµB 162.6(125)(0.8)(1.25)
k= = = 49.6md
mh (27.3)(15)
 ( P1hr - Pwf )  k  
S = 1.151 − log  + 3.23 = 1.151(5.055 − 8.379 + 3.23) = −0.1
  
 ct rw 
2
 m 
kt (49.6)(36)
rinv = = = 792 ft
948ct 948(0.25)(0.8)0.000015
PTA.2 Analytical Well Test Simulator

The Analytical Well Test Simulator included in the PE² Essentials Pressure Transient Analysis tool
simulates well tests based on the “Method of Images”. Imaginary wells, referred to as image
wells, are used to generate the pressure effect of the reservoir discontinuity.
For a full description of the Method of Images, refer to the Streltsova book. For purposes of this
section, rather than a derivation, only specific results are presented here. The basic solution of
the flow equation for pressure drop at any point in the reservoir is given by Equation 8.8.
qB  948ct r 2 
P = Pi + 70.6 Ei −  (PTA.8)
kh  kt 
 e −u
Ei (− x) = −  du
x u
Where: Ei is the exponential integral and all other terms are as defined previously.
At the wellbore, r becomes rw and for practical shut-in times, Δt, Ei can be approximated by a log
function, Ei (− x) = ln(1.781x) so the flowing pressure at the wellbore becomes Equation PTA-9:
qB  kt 

Pwf = Pi − 70.6 ln 2 
(PTA.9)
kh  1688ct rw 
For the Method of Images, the image well is represented by Equation 8.8. Consider a linear no-
flow boundary (a fault) in the reservoir. The configuration can be represented as shown in Figure
PTA-7 with the real well and an image well located at a distance of 2d from the real well, d being
the distance to the fault.
Figure PTA-7: Image Well configuration for a No-Flow Boundary
The image well produces at the same rate as the real well so, over time, there will be a drainage
boundary generated at a distance ‘d’ from the real well as shown in Figure PTA-8
Time 1
Time 2
Figure PTA-8: Image Well Generation of No-Flow Boundary
The equation describing the pressure at the wellbore in Figure PTA-8 is presented below:
qB   kt 

 948ct (2d ) 2 
Pwf = Pi − 70.6  ln − Ei   (PTA.10)
kh   1688ct rw 2   
 kt 
The first term in the square brackets is the pressure drop at the real well and the second term is
the pressure drop caused by the image well at a distance of 2d from the real well.
The Analytical Well Test Simulator can model the following configurations (Figure PTA-9):
Figure PTA-9: Analytical Model Configurations
Figure PTA-10 presents examples of two image well configurations: perpendicular boundaries
and parallel no-flow boundaries. For more complete descriptions and for additional
configurations, refer to the Streltsova book.
Figure PTA-10: Example Image Well Configurations
To use the Analytical Well Test Simulator, the data from the build-up analysis can be imported
into the model or, to simulate a future test, the data can be manually entered.
PTA.2 Gas Well Example

This example uses the data in the Excel file ‘ General WellTest Example_BHP Data.xlsx’ in the ‘PE
Essentials 2020\Example Input Files\Excel Files’ directory.
This is a gas well example so choose ‘Gas Well’ on the main screen and enter the gas PVT
parameters (Figure PTA-11).
Figure PTA-11: Example Gas Well PVT Data

The Excel file is linked to the tool and the pressure data is entered (Figure PTA-12 and PTA-13).
Figure PTA-12: Example Gas Well Pressure Data
Figure PTA-13: Example Gas Well

Note that the pressure data can be saved to the PE Tools database for import at a later time.
The values for net pay and wellbore radius are then entered before proceeding to analysis.
From the log-log plot, it appears that the derivative had stabilized between 3.5 hours and 12
hours. Two points are entered on the semi-log plot to place a straight line in this time interval
(Horner time of 11 and 38) – Figure PTA-14. The straight line can be changed by clicking on one
of the end points.
Figure PTA-14: Example Gas Well Analysis
The pressure derivative and the semi-log plot data appears to fall below the straight line at late
shut in times. This could be indicative of increasing pay in the reservoir or a constant pressure
boundary – the radius of investigation for this test is 736 ft.
184 Volumetric Analysis Tool
Volumetric Analysis Tool

The PE² Essentials ‘Volumetric (MB) Analysis’ tool is a diagnostic material balance analysis tool (Figure
MBA-1).
Figure MBA-1: PE² Essentials Volumetric (MB) Analysis Tool
MBA.1 General Material Balance

Material balance analysis (MBA) is an interpretation technique used to determine original oil or
original gas in-place (OOIP, GIIP) based on production and injection volumes and static reservoir
pressure data. It is based on the law of conservation of mass and makes the following
assumptions:
• The reservoir is homogenous and behaves like a tank

• Fluid production/injection occurs at single points
• There is no directional component to the flow
In real life, none of these assumptions are met by oil and gas reservoirs: reservoirs are not
homogeneous tanks, production and injection are areally distributed and occur at different times
and fluid flow is directional. Nevertheless, MBA is a commonly used analysis technique and has
been found to yield reasonably acceptable results.
The general material balance equation relates the original oil, gas, and water volumes in the
reservoir to production volumes and current pressure conditions and fluid properties. The
assumption of a tank behaviour means that the reservoir is considered to have the same pressure
and fluid properties at all location in the reservoir. Consider Figure MBA-2
Figure MBA-2: Reservoir Material Balance
The simplest way to visualize material balance is that if the measured surface volume of oil, gas
and water were returned to a reservoir at the reduced pressure, it must fit exactly into the
volume of the total fluid expansion plus any fluid influx.
The general material balance for an oil reservoir can be expressed as follows:
Net Reservoir Withdrawal = Expansion of Oil

+ Original Dissolved Gas
+ Expansion of Gas Cap
+ Reduction in Hydrocarbon Pore Volume
+ Natural Water influx
+ Gas/Water Injection Volumes
The general material balance for a gas reservoir can be expressed as follows:
Net Reservoir Withdrawal = Expansion of Gas

MBA.2 Data Input

Figure MBA-3 shows the data input options for the PE² Essentials MBA tool. Data can be input
from an Excel spreadsheet. After importing the production data, it can be saved to a DVXv file
through the ‘Save Production/PVT Data’ button on the main screen (Figure MBA-1).
Figure MBA-4 shows the PVT input screen.
Figure MBA-3: PE² Essentials Volumetric (MB) Analysis Tool, Data Input
Figure MBA-4: PE² Essentials Volumetric (MB) Analysis Tool, PVT/Reservoir Data Input
The PVT parameters can be entered manually or imported from a PE² Essentials PE Tools
database. Clicking ‘Import PE Tools db PVT Properties’ will open a sheet that will give the option
of entering PVT data stored with a well or PVT data from a stored PVT model (Figure MBA-5).
Figure MBA-5: Importing PVT Data from the PE Tools Database
Selecting the appropriate button will list the options available for the database. Click on the
relevant well/tool and the PVT data will be imported into the tool. The remaining data is entered
manually.
MBA.3 General Oil Material Balance

The general material balance equation for an oil reservoir can be expressed as the equation of a
straight line, as follows:
F = N(Eo + mEg + Efw) + We (MBA-1)
Where: F is net withdrawal at reservoir conditions in rbbl, N is the original oil in place in stbbl, Eo
is the expansion of oil and original gas in solution in rbbl/stbbl, m is the initial gas cap volume
fraction (initial hydrocarbon volume of the gas cap / initial hydrocarbon volume of the oil zone)
in rbbl/rbbl, Eg is the expansion of the gas cap gas in rbbl/stbbl, Efw is the expansion of the connate
water and reduction in the hydrocarbon pore volume due to connate water expansion and
decrease in the pore volume in rbbl/stbbl, and We is cumulative water influx from the aquifer in
rbbl.
The terms in the general oil material balance equation are as follows:
F = Np [Bo + (Rp – Rs)Bg] + Wp Bw – Wi Bw - Gi Bg (MBA-2)

Eo = (Bo – Boi) + (Rsi – Rs)Bg
Eg = Boi Bgi (Bg – Bgi)
Efw = (1 + m) Bo Ce ΔP
Ce = (Cw Swi + Cf) / (1 – Swi)
We = U S(p,t)
Where:
- F is net withdrawal at reservoir conditions in rbbl, Np is the cumulative oil production in stbbl,
Bo is the oil formation volume factor at current conditions rbbl/stbbl, Rp is the cumulative oil-gas
ratio (Gp/Np) in scf/bbl, Rs is gas in solution at current conditions in scf/stbbl, Bg is gas formation
volume factor at current conditions in rbbl/scf, Wp is cumulative water production in stbbl, Bw is
water formation volume factor at current conditions in rbbl/stbbl, Wi is cumulative water
injection in stbbl, and Gi is cumulative gas injection in scf.
- Eo is the expansion of oil and original gas in solution in rbbl/stbbl, Bo is the oil formation volume
factor at current conditions rbbl/stbbl, Boi is the oil formation volume factor at initial conditions
rbbl/stbbl, Rsi is gas in solution at original conditions in scf/bbl, Rs is gas in solution at current
conditions in scf/bbl, and Bg is gas formation volume factor at current conditions in rbbl/scf.
- Eg is the expansion of the gas cap gas in rbbl/stbbl, Boi is the oil formation volume factor at initial
conditions rbbl/stbbl, Bgi is gas formation volume factor at initial conditions in rbbl/scf, and Bg is
gas formation volume factor at current conditions in rbbl/scf.
- Efw is the expansion of the connate water and reduction in the hydrocarbon pore volume due
to connate water expansion and decrease in the pore volume in rbbl/stbbl, m is the initial gas
cap volume fraction (initial hydrocarbon volume of the gas cap / initial hydrocarbon volume of
the oil zone) in rbbl/rbbl, Bo is the oil formation volume factor at current conditions rbbl/stbbl,
Ce is the effective compressibility of the connate water and the pore volume in 1/psi, and ΔP is
the change in reservoir pressure (Pi – Pr) in psi.
- Ce is the effective compressibility of the connate water and the pore volume in 1/psi, C w is the
water compressibility in 1/psi, Swi is the initial connate water saturation, and Cf is the pore volume
compressibility in 1/psi.
- We is cumulative water influx from the aquifer in rbbl, U is the aquifer constant in rbbl/psi, and
S(p,t) is the aquifer function which is dependent on the type of aquifer in psi.
MBA.3.1 Undersaturated Oil / Depletion Drive Reservoir

For an undersaturated oil / depletion drive reservoir, the Eg and We terms in the general material
balance equation are zero, simplifying the oil material balance equation to:
F = N (Eo + Efw) = NEt (MBA-3)

This indicates that for an undersaturated oil / depletion drive reservoir, a plot of F versus E t
(Figure MBA-6) will yield a straight line going through the origin and having a slope of N. In some
cases, a reservoir will exhibit depletion drive very early in its production life.
Figure MBA-6: Volumetric Analysis – Depletion Drive Analysis
Figure MBA-5 includes data presented on page 759 in Ahmed and the data file is included as ‘PE²
Essentials Oil Well Volumetric Surveillance Data Depletion Drive Ahmed pg 759.DVXv’ in the
“Example Input Files\DVX Model Files\Volumetric Analysis” directory. The oil initially in place was
reported by Ahmed to be 257 mmstbls.
Refer to MBexamples.xlsx in the “Example Input Files\Excel Files” directory for complete
information for all the cases presented in this section.
Figure MBA-7 presents the data from a theoretical water drive oil reservoir. This reservoir had an
initial oil in place of 150 mmbbls. This data is included as ‘PE Essentials Oil Well Volumetric
Surveillance Data Water Drive.DVXv’ in the Example Input Files\DVX Model Files\Volumetric
Analysis directory.
A water drive reservoir may exhibit a depletion drive at very early times, but this is not always
the case. Caution should be used when attempting this type of analysis on a water drive reservoir.
The reservoir in Figure MBA-6 is obviously a water drive reservoir based on the non-linear trend
of the data.
Figure MBA-7: Volumetric Analysis – Depletion Drive Analysis of Water Drive Reservoir
Note that the straight line can be moved by dragging either the start point or end point of the
line.
MBA.3.2 Gas Cap Drive Oil Reservoir

For an oil reservoir containing a gas cap, the We term is zero and the Efw term can be disregarded
since the compressibility of the gas is an order of magnitude greater than the compressibility of
the water or the rock. The oil material balance equation becomes:
F = N(Eo + mEg) (MBA-4)

F/Eo = N + mN(Eg/Eo)
Equation MBA-4 is the equation of straight line having a slope of mN and an intercept of N. Figure
MBA-8 is an example of a theoretical gas cap drive oil reservoir. This reservoir had an initial oil in
place of 150 mmbbls and an initial gas cap volume factor, m, of 0.5.
Figure MBA-8: Volumetric Analysis – Gas Cap Drive Reservoir
Once the straight line is placed on the graph, the values for N and m are automatically calculated.
Note that the straight line can be moved by dragging either the start point or end point of the
line.
MBA.3.3 Water Drive Oil Reservoir

For an oil reservoir containing an aquifer, only the Eg term in the general material balance
equation is zero. The oil material balance equation becomes:
F = N(Eo + Efw) + We (MBA-5)

F/(Eo + Efw) = N + We/(Eo + Efw)
F/Et = N + We/Et
The We term is a difficult term to determine without knowledge of the aquifer type and
properties. Refer to Section 5.6 for complete information on aquifer modeling. To generate a plot
of Equation MBA-5, the We data for a given aquifer is generated and the plot is examined. If the
plot is not a straight line, a new aquifer model is used to generate a different set of W e data and
the plot re-examined.
This is a time-consuming process so in 1978 Campbell et all (Campbell, R.R. and Campbell, J.M.;
Mineral Property Economics, vol. 3, Petroleum Property Evaluation, Campbell Petroleum Series,
1978) published the Campbell Method for a water drive reservoir. For this method, F/E t versus F
id plotted. The intercept of the line will yield N and the slope is a function of the aquifer
performance. Figure MBA-9 is the same example that was presented in Figure MBA-7.
Figure MBA-9: Volumetric Analysis – Water Drive Analysis
MBA.3.4 Combination Gas Cap / Water Drive Oil Reservoir

For an oil reservoir containing a gas cap and an aquifer, all terms in the general material balance
equation are required for analysis. The problem can very quickly become unconstrained and
difficult to obtain a unique solution. The oil material balance equation becomes:
F = N(Eo + mEg + Efw) + We (MBA-6)

F/(Eo + mEg +Efw) = N + We/(Eo + mEg + Efw)
F/Et = N + We/Et
There are three unknows in Equation MBA-6; N, We and m. The Campbell method is used by
assuming a value for m and calculating N.
For the theoretical example included as ‘PE Essentials Oil Well Volumetric Surveillance Data Gas
Cap Water Drive.DVXv’ in the “Example Input Files\DVX Model Files\Volumetric Analysis”
directory, a gas cap ratio, m, of 0.5 was entered for the analysis. OOIP was estimated to be 148
mmstb. Figure MBA-10 shows the analysis plot for this data.
Figure MBA-10: Volumetric Analysis – Gas Cap / Aquifer Combination Drive Analysis
MBA.3.5 Drive Indices for an Oil Reservoir

If the oil zone withdrawal terms are separated from the remaining terms in the general material
balance equation, the following is obtained:
Np [Bo + (Rp – Rs)Bg] = N(Eo + mEg + Efw) - Wp Bw – Wi Bw - Gi Bg + We (MBA-7)
By definition, m = G Bgi / N Boi. By replacing the terms for the other expressions and dividing by
the left hand side, Equation MBA-8 is obtained. Note that the Efw has been replaced by the term
“Rock”.
N [( Rsi − Rs ) Bg − ( Boi − Bo )] + G ( Bg − Bgi ) − G p Bg + Gi Bg + We − W p Bw + Wi + Rock (MBA-8)

1=
N p Bo + N p ( R p − Rs ) Bg
From Equation MBA-8, the drive indices are defined as follows:

N [( Rsi − Rs ) Bg − ( Boi − Bo )]
SDI = (MBA-9)
G ( Bg − Bgi ) − G p Bg + Gi Bg
GDI = (MBA-10)
We − W p Bw + Wi
WDI = (MBA-11)
EDI = 1.0 − SDI − GDI − WDI (MBA-12)
Where: SDI is the solution gas drive index, GDI is the gas drive index, WDI is the water drive index, and EDI
is the compressibility expansion drive.
Since the sum of the drive indices is one, calculating the three main drive indices (SDI, GDI, WDI) will allow
the calculation of the fourth index, EDI.
Following an analysis, an evaluation of the drive index can be performed to confirm that the
results make sense. For the Figure MBA-9 example in Section MBA.3.4, a plot of the drive indices
was generated (Figure MBA-11) using the assumed value for m and the estimated value for OOIP
obtained from the analysis. The plot shows that the reservoir drive started out as solution gas
and gas cap drive, then as production progressed, water drive became more dominant.
Figure MBA-11: Volumetric Analysis – Combination Drive Reservoir, Drive Index Analysis
MBA.3.6 Water Drive Reservoir Example

An example of a water drive reservoir was history matched to a Fetkovich finite aquifer (refer to
PE Essentials Oil Material Balance Tool documentation). Table MBA-1 presents the data.
Pressure Cum Oil Cum Gas Cum Wat Pressure Cum Oil Cum Gas Cum Wat
Years Years
(psi) (mstb) (mmscf) (mbbls) (psi) (mstb) (mmscf) (mbbls)
0 3000 0 0 0 11 2698 5274.3 3711.7 207.7
1 2923 564.2 395.8 3.3 12 2680 5563.7 3916.2 237.3
2 2880 1418.8 996 15.7 13 2665 5846.5 4116 269.1
3 2848 2155.3 1513.9 32.7 14 2650 6113.6 4304.8 301.8
4 2821 2766.9 1944.2 52.3 15 2633 6386.7 4498 338.6
5 2800 3265.5 2295.2 72.2 16 2620 6642.3 4678.9 375.5
6 2780 3686.7 2591.9 92.6 17 2607 6874.9 4843.6 411.4
7 2762 4058 2853.5 113.6 18 2593 7104.3 5006 449.4
8 2744 4400.1 3094.8 136 19 2580 7328.5 5164.8 488.9
9 2730 4702.4 3308 158.1 20 2568 7539.7 5314.4 528.4
10 2713 4992.3 3512.6 182
Table MBA-1: Volumetric Analysis Example – Water Drive
The data in Table MBA-1 is included in the ‘Smith pg 12-71.xlsx’ spreadsheet located in the “Book
Examples\Example MBA\Water Drive” directory.
The data was imported and is included in the ‘PE Essentials Oil Well Volumetric Surveillance Data
Smith pg 12-71.DVXv’ file. Figure MBA-12 shows the PVT and reservoir properties for this
example.
Figure MBA-12: Volumetric Analysis Example – PVT Data

According to Smith et al and confirmed by the history match of the production history presented
in the Oil Material Balance Tool Example, the initial oil in place for this reservoir was 26.6 mmbbls.
Figure MBA-13 presents the Campbell plot for this data.
The straight line through the early time data indicated 41.6 mmbbls. This is obviously too high
which suggests that the aquifer was affecting the pressure even at early times. In order to
evaluate this data, the ‘Plot MB Model’ option was used to compare the history matched
reservoir/aquifer response with the observed response.
The first time ‘Plot MB Model’ is checked, the material balance models available in the PE Tools
database will be listed for selection (Figure MBA-14).
Figure MBA-13: Volumetric Analysis Example – Volumetric Analysis

Figure MBA-14: Volumetric Analysis Example – Selecting MB Model
After choosing a material balance model, it is plotted. This is shown in Figure MBA-15.
Figure MBA-15: Volumetric Analysis Example – Volumetric Analysis with MB Model

From Figure MBA-15, it is apparent that the green straight line should be steeper (lower N) as
expected. In addition, additional work could be performed on the reservoir/aquifer model to
improve the match, if required.
Figure MBA-16 shows the drive index analysis for this example assuming an initial oil in place of
26.6 mmbbls.
From the plot of the drive indices, the performance of the reservoir/aquifer model is very similar
to the response of the historical data.
The plots indicate that water drive was the dominant drive for the reservoir, confirming the
conclusion that water was influencing the pressure at early times and that the value of N obtained
from Figure MBA-13 was too high.
It should be noted that a close-to-perfect match in Figure MBA-15, can be obtained by assuming
a 24 mmbbls initial oil in place. Since fluid properties are different, a difference in results is
expected.
Figure MBA-16: Volumetric Analysis Example – Drive Indices with MB Model

MBA.4 General Gas Material Balance

The general material balance equation for a gas reservoir can be expressed as the equation of a
straight line, as follows:
F = G(Eg + Efw) + We (MBA-13)
Where: F is net withdrawal at reservoir conditions in rbbl, G is the original gas in place in scf, Eg
is the expansion of the gas cap gas in rbbl/scf, Efw is the expansion of the connate water and
reduction in the hydrocarbon pore volume due to connate water expansion and decrease in the
pore volume in rbbl/scf, and We is cumulative water influx from the aquifer in rbbl.
The terms in the general gas material balance equation are as follows:
F = Gwgp Bg + Wp Bw
Gwgp = Gp + Npc Fc
Eg = Bg – Bgi
Efw = Bgi Ce ΔP
Ce = (Cw Swi + Cf) / (1 – Swi)
Fc = 132.79 Sc / Mc
Sc = 141.5 / (131.5 + °API)
Mc = 6084 / (°API - 5.9)
We = U S(p,t)
Where:
- F is net withdrawal at reservoir conditions in rbbl, Gwgp is the cumulative wet gas production in
scf, Bg is gas formation volume factor at current conditions in rbbl/scf, Wp is cumulative water
production in stbbl, and Bw is water formation volume factor at current conditions in rbbl/stbbl.
- Gwgp is the cumulative wet gas production in scf, Gp is the cumulative dry gas production in scf,
Np is the cumulative condensate production in stbbl, Fc is the condensate conversion factor in
scf/stbbl, Sc is condensate specific gravity, °API is the API gravity of the condensate, and Mc is the
molecular weight of condensate.
- Eg is the gas expansion rbbl/scf, Bgi is gas formation volume factor at initial conditions in rbbl/scf,
and Bg is gas formation volume factor at current conditions in rbbl/scf.
- Efw is the expansion of the connate water and reduction in the hydrocarbon pore volume due
to connate water expansion and decrease in the pore volume in rbbl/stbbl, Bgi is gas formation
volume factor at initial conditions in rbbl/scf, Ce is the effective compressibility of the connate
water and the pore volume in 1/psi, and ΔP is the change in reservoir pressure (Pi – Pr) in psi.
- Ce is the effective compressibility of the connate water and the pore volume in 1/psi, C w is the
water compressibility at current conditions in 1/psi, Swi is the initial connate water saturation,
and Cf is the pore volume compressibility at current conditions in 1/psi.
- We is cumulative water influx from the aquifer in rbbl, U is the aquifer constant in rbbl/psi, and
S(p,t) is the aquifer function which is dependent on the type of aquifer in psi.
MBA.4.1 Gas Expansion / Volumetric Gas Reservoir

For a volumetric, depletion drive reservoir, the drive energy is the expansion of the gas and the
Efw and We terms in the general material balance equation are zero, simplifying the gas material
balance equation to:
F = G Eg (MBA-14)
Incorporating all the terms for these parameters and including the expression for Bg, the material
balance equation, for a volumetric gas reservoir, becomes the conventional P/Z equation:
P/Z = Pi/Zi (1 – Gp/Zi) (MBA-15)
From the above expression, a plot of P/Z versus Gp will yield a straight line which can be
extrapolated to 0 to yield the original gas in place, G.
Figure MBA-16 is a plot of the data included as ‘PE Essentials Gas Well Volumetric Surveillance
Data Gas SPE16484 - Case 1.DVXv’ in the “Example Input Files\DVX Model Files\Volumetric
Analysis” directory.
Figure MBA-16: Volumetric Analysis – Volumetric Gas Reservoir
This example is a volumetric gas reservoir which, based on the P/Z plot, contains 70.5 Bscf of gas
initially in place.
MBA.4.2 Water Drive Gas Reservoir

For a gas reservoir containing an aquifer, the Efw term can be neglected because it is very small
(by an order of magnitude) when compared to the Eg term. The general material balance
equation becomes:
F = GEg + We (MBA-16)
We = U S(p,t) (MBA-17)
The final form of the material balance equation for a gas reservoir containing an aquifer is:
F/Eg = G + U S(p,t)/Eg (MBA-18)
The definitions of U and S(p,t) depend on the type of aquifer.
For most gas reservoirs containing an aquifer, the P/Z versus Gp plot can be used to evaluate the
early production time data to obtain an estimate of the initial gas in place, G.
Figure MBA-17 presents the data from ‘PE Essentials Gas Well Volumetric Surveillance Data Gas
w Aquifer SPE16484 - Case 2.DVXv’ in the “Example Input Files\DVX Model Files\Volumetric
Analysis” directory.
Figure MBA-17: Volumetric Analysis – Gas Reservoir with Aquifer, Expansion Drive Analysis
In the early stages of production, the reservoir appears to behave like an expansion drive /
volumetric reservoir. A straight line on the P/Z plot indicates a gas initially in place of
approximately 1350 Bscf.
Figure MBA-18 shows the water drive analysis for this example.
To evaluate this plot, the gas in place from the previous analysis was used to direct the analysis.
This analysis yields a similar initial gas in place of 1306 Bscf.
It should be noted that this solution is not unique. More in-depth analysis may be required
whenever an aquifer is present in a reservoir.
Figure MBA-18: Volumetric Analysis Example – Gas Reservoir With Aquifer, Water Drive
MBA.4.3 Depletion Drive Gas Reservoir Example

In the Gas Material Balance Tool example, a multi-tank history match was generated for a low-
perm gas reservoir. The production data for this example was input into the Volumetric (MB)
Analysis tool for analysis (Figure MBA-19). The data file is included as ‘PE Essentials Gas Well
Volumetric Surveillance Data Multi-Tank Gas Reservoir.DVXv’ in the “Book Examples\Example
MBA\ Gas Depletion Drive” directory.
Figure MBA-19: Volumetric Analysis Example – Gas Input Data
The PVT data is shown in Figure MBA-20.
Figure MBA-20: Volumetric Analysis Example – PVT Data
A depletion drive analysis indicated 6.1 Bscf gas initially in place (Figure MBA-21) and Figure MBA-
22 confirms that the multi-tank gas material balance model is valid for this reservoir. Note the
multi-tank gas material balance model indicated 3 Bscf in tank 1 and 50 Bscf in tank 2.
Figure MBA-21: Volumetric Analysis Example – Depletion (P/Z) Analysis
Figure MBA-22: Volumetric Analysis Example – Depletion (P/Z) Analysis with Gas MB Model
Hydrate Analysis Tool

Natural gas hydrates are crystals formed by water, natural gases and associated liquids, in a ratio
of 85 mole% water to 15 mole% hydrocarbons. Hydrates can form anywhere and anytime that
hydrocarbons and water are present at the appropriate temperature and pressure, such as in
wells, flow lines, valves and meter discharges.
Low temperatures and high pressures will cause hydrate formation, but it is also a function of gas
composition. In a pipeline, hydrates usually form at the hydrocarbon-water interface, and
accumulate as flow pushes them downstream. In wells, hydrates can form at any point in which
hydrate formation conditions are met.
Three conditions are required to form hydrates:

• Free water and natural gas. Gas molecules ranging in size from methane to butane are
typical hydrate components, including CO2, N2, and H2S. The water in hydrates can come
from free water produced from the reservoir, or from condensed water resulting from
the cooling of the gas phase.
• Low temperatures are required for hydrate formation; but, even though hydrates are 85
mole% water, the temperature does not have to be below the freezing point of water for
hydrates to occur. Hydrates can easily form at 38°-40°F or higher temperatures with high
pressure.
• High pressures can cause hydrate formation. At 38°F, common natural gases form
hydrates at pressures as low as 100 psi. At 1500 psi, common gases can form hydrates at
temperatures as high as 66°F. Since pipelines typically operate at higher pressures,
hydrate prevention can be a primary consideration.
There are a number of techniques that can be used to prevent hydrate formation:
• Water removal provides the best protection for pipelines. Free water is removed through
separation, and water dissolved in the gas can be removed by drying with tri-ethylene
glycol (TEG) to obtain water contents less than 7 lbm/MMscf. Water removal is not
normally possible for a producing well, so other prevention schemes must be used.
• Maintaining high temperatures keeps the system in the hydrate-free region. High fluid
temperature may be retained by adding heat by circulating hot fluids or electrical heating,
although this may not economical in most cases.
• The pressure may be decreased below hydrate formation pressure. This leads to the
concept of adding system pressure drops at high temperature points (e.g. bottom-hole
chokes during testing). However, this may not be feasible in production situations.
• Most frequently hydrate prevention means injecting an inhibitor such as methanol
(MeOH) or mono-ethylene glycol (MEG), which decreases the hydrate formation
temperature below the operating temperature.
• Kinetic inhibitors are low molecular weight polymers dissolved in a carrier solvent which
are injected into the water phase in pipelines. These inhibitors work by bonding to the
hydrate surface and preventing crystal formation and growth for a period longer than the
free water residence time in the well/pipeline. Water is then removed at the facilities.
206 Hydrate Analysis Tool
Depressurization is often the main tool for hydrate plug removal, but the preferred solution is to
prevent the formation of hydrate plugs in the first place, through design and operating practices.
The use of many gallons of inhibitors may be costly on a continuous basis but these expenses are
easily overshadowed when considering the formation of a hydrate causing production to stop.
The PE² Essentials Hydrate Analysis tool enables the determination of the hydrate formation
temperature using simple gas specific gravity techniques as well as the more accurate gas
composition techniques. In addition, the tool includes the capability of estimating the inhibitor
volumes required to protect the wells and pipelines. The tool can model a number of hydrate
inhibitors, although the modeling of kinetic inhibitors is not included in this version of the tool.
An excellent reference is Carroll, J; Natural Gas Hydrates, A Guide for Engineers, Gulf Professional
Publishing, 2020.
HYD.1 Hydrate Prediction – Specific Gravity Methods

A total of six specific gravity (SG) based correlations are included in the Hydrate Analysis tool
(Figure HYD-1).
Figure HYD-1: PE² Essentials Hydrate Analysis Tool – Gravity Based Analysis
Although the SG method may not be highly accurate it has a high level of appeal because of its
simplicity. There are numerous SG-based correlations published in various papers, books and
journals. It was found that a number of these correlations have errors in their formulas. The most
famous problem is with the Kobayashi et al correlation published by the SPE (ref: Kobayashi, R.,
Song, K.Y., Sloan, E.,D., Bradley, H.B. (Editor), Petroleum Engineers Handbook. SPE, Richardson,
TX, 1987, 25-1, 25-28). Numerous authors, including myself, have attempted to correct the
equation but were unsuccessful.
Because of the limitations in the SG models, several correlations have been included in the
Hydrate tool to allow a range of values to be determined. Note, unless otherwise specified, the
hydrate temperature is reported in °F and pressure is in psi.
HYD.1.1 Katz (1944)

The Katz correlation is based on the charts published by Katz in 1944 (ref: Katz, D., L.: Prediction
of Conditions for Hydrate Formation in Natural Gases, Petroleum Technology, TP 1748, July 1944
– also available as SPE 945140). The Katz correlation used in the Hydrate Analysis tool was derived
from Figure HYD-2. This graph was presented in the Katz reference.
Figure HYD-2: Katz Gravity Based Correlation
Based on Figure HYD-2, the EPCI-Katz correlation equations are as follows:

SG = 0.554: T_hyd =-6.9347(Log(P))4 + 87.285(Log(P))3 - 413.58(Log(P))2 + 909.9Log(P) - 753.98
SG = 0.6: T_hyd = 4.4739(Log(P))4 - 54.971(Log(P))3 + 242.63(Log(P))2 - 426.21Log(P) + 277.59
Figure HYD-3 presents the plots of the EPCI-Katz equations. The hydrate temperature is calculated by
interpolating between the appropriate SG curves.
Figure HYD-3: Katz Gravity Based Pressure-Temperature Curves
HYD.1.2 Salufu (2013)

Salufu et al observed that the water vapor pressure had an effect on the pressure of the gas
molecule when hydrate was forming (ref: Salufu, S.O., and Nwakwo, P.; New Empirical
Correlation for Predicting Hydrate Formation Conditions, SPE 167571, 2013).
Salufu presented a correlation for hydrate temperature, T_hyd, that included the effect of vapour
pressure, Pvw, as follows:
SG<0.6: T_hyd = 16.2602 (Ln(P e(Pvw)) - Ln(105.358 SG))

0.6<SG<0.84: T_hyd = 12.1212 (Ln(P e(Pvw)) - Ln(8.7511 SG))
SG>0.84: T_hyd = 10.9529 (Ln(P e(Pvw)) - Ln(2.4196 SG))
The water vapour pressure is calculated using Buck’s equation (ref: Buck, A.L,; New Equation for
Computing Vapor Pressure and Enhancement Factor, American Meteorological Society, 1982).
Pvw = 0.088648 e((18.678 - Tc/234.5) (Tc/(257.14 + Tc)))
Where: Pvw is the water vapor pressure, psi, and Tc is the temperature, °C.
HYD.1.3 Towler (2005)

In 2005, Towler and Mokhatab presented a correlation to generate T_hyd (ref: Towler, B. F. and
S. Mokhatab, “Quickly estimate hydrate formation conditions in natural gases,” Hydrocarbon
Processing, Vol. 84, 2005). This correlation is as follows.
T_hyd = 13.47Ln(P) + 34.27ln(SG) - 1.675ln(SG)Ln(P) – 20.35

The accuracy of the Towler correlation in predicting hydrate formation temperature reduces with
increasing methane content in a binary methane-ethane system.
HYD.1.4 Makogan (1997)

In 1997, Makogan published a simple correlation for the hydrate formation pressure as a function
of temperature and gas gravity for paraffin hydrocarbons (ref: Makogon, Y.F., Hydrates of
Hydrocarbons. PennWell, Tulsa, OK. 1997).
Makogan’s correlation generates pressure in atm for a given temperature in °C. The correlation
has been solved for hydrate temperature for use in the Hydrate tool as follows:
T_hyd(°C) = (-0.0497 + [0.00247 – 0.1988 k (b - log(Patm))]0.5) / 0.0994k

b = 2.681 - 3.811SG + 1.679SG2
k = -0.006 + 0.011SG + 0.011SG2
HYD.1.5 Motiee (1991)

In 1991, Motiee published a correlation to calculate the hydrate temperature as a function of the
pressure and the gas gravity (ref: Motiee, M., Estimate Possibility of Hydrate. Hydro. Proc. 70 (7),
98-99, July 1991).
T_hyd =-124.951+48.98387 log(P)-2.66303(log(P))2+176.9101 SG-75.5873 SG2-10.4505 SG log(P)
HYD.1.6 Sun (2013)

The specific gravity method presented above are not generally valid for sour gases and there are
very few correlations that have been published specifically for sour gases. In 2013, Sun et al
published a correlation based on 60 sour gas samples for the ternary mixture of CH4, CO2, and
H2S (ref: Sun, C.,Y., Chen, G.,J., Lin, W., Guo, T.,M., Hydrate Formation Conditions of Sour Natural
Gases, J. Chem. Eng. Data 48, 600-603, 2013). The Sun correlation generates hydrate temperature
in °K from input pressure in MPa.
T_hyd(°K) = 1000 / (4.343295 + 0.0010734 P – 0.091984 ln(P) - 1.071989 SG)

HYD.2 Hydrate Prediction – Gas Composition Methods

The gas gravity methods have a high level of appeal because of their simplicity. Despite their
relative simplicity, these methods can be surprisingly accurate. For sweet natural gas mixtures,
the gas gravity method is reportedly accurate to within 20% or better for estimating hydrate
pressures/temperatures.
Whenever there is an increasing amount of heavier components present, the resultant hydrate
characteristics change. This change can have a significant effect on the hydrate forming pressure.
For example, pure methane forms a hydrate at 59°F and 1855 psi. The presence of only 1%
propane results in a mixture that forms a hydrate at 59°F and 1115 psi.
If the gas composition is available, then the more accurate technique to calculate the hydrate
temperature/pressure point is by the use of K-factor charts which represent vapor-solid
equilibrium conditions for the hydrate forming components.
The PE² Essentials Hydrate Analysis tool (Figure HYD-4) includes two composition-based
techniques for generating hydrate temperature/pressure: one based on the original Katz K-
Charts presented in the GPSA Engineering Data Book, as published by Berge in 1986; and the
second presented by Mann in her 1988 thesis and published by Poettmann et al in 1989.
Figure HYD-4: PE² Essentials Hydrate Analysis Tool – Composition Based Analysis
HYD.2.1 Berge (1986)

In 1986 Berge published a paper describing the use of a computer to calculate the Katz K-Factor
parameters (ref: Berge, B., K., Hydrate Predictions on a Microcomputer, SPE 15306, 1986). The
calculations are rather simple to implement but requires an iterative technique to generate a
result.
For this version of the Hydrate Analysis tool, the Towler SG-based correlation is used to generate
an initial estimate of the hydrate temperature, Tg, for a given pressure in psi. With the pressure
and estimated hydrate temperature, the K-Factors are calculated for each component from the
K_C1 = 1.0 - (0.014 + 15.38/P) (54.81 - 21.37 ln(P) + 2.95(ln(P))2) + (0.14 + 15.38/P) Tg
If the calculated K_C1 is greater than 1, then the following equation for K_C1 should be used:
K_C1 = 1.01 + 9.31/Tg - 1614.16/Tg2 + (625.57 - 3.31Tg + 0.023Tg2)/P + (-3.79e4 + 2.09e6/Tg -
9.82e7/Tg2 + 1.54e9/Tg3)/P2
K_C2 = e(-29.91 + 2779.25/P - 60251.4/P^2) Tg(6.94 - 471.93/P)
K_C3 = -0.094 + 21.76/P - 4039.38/P2 + 2.44e5/P3 + (-3.5 + 2191.45/(P + 1021.19)) / (Tg + (-

80.57 + 9076.44/(P + 166.9)))
If P<=600 psi then use the following equation for K_iC4:

K_iC4 = (-2.22e-6 + 6.99e-3/P + 0.74/P2) e(Tg (0.14 + 1.88/P))
If P>=800 psi then use the following equation for K_iC4:
K_iC4 = 6.66e-6 e(4.72e-4 P) e(Tg (0.16 - 1.24e-5 P))
Otherwise use the following equation for K_iC4:
K_iC4 = 1.15e-5 e(0.146Tg) + (P - 600)/200 (9.72e-6 e(0.145Tg) - 1.15e-5 e(0.146Tg))
If P<=600 psi then use the following equation for K_nC4:

K_nC4 = 0.25 - 1.7e-3 P + (-6.85e-4 + 3.34e-5 P) Tg
If P>=700 psi then use the following equation for K_nC4:
K_nC4 = 1.67 - 0.092Tg + 1.25e-3 Tg2
Otherwise use the following equation for K_nC4:
K_nC4 = -0.768 + 1.93e-2 Tg + (2.44 - 0.11Tg + 1.25e-3 Tg2) (P - 600)/100
The value for K_N2 presented by Berge was changed to a correlation presented in the Mann
thesis which was dependant on whether or not H2S was present in the gas.
For sweet gas (H2S=0), the following set of equations is used.

If Tg<=32 °F then the following equation is used for K_N2:

K_N2 = e(35.49 – 964/(Tg+460) – 0.001396P + 5.603 ln(P) – 87.46/P – 0.06486(Tg+460) – 2972.6 ln(P)/(Tg+460))
Otherwise the following equation is used for K_N2:
K_N2 = e(174.82 - 44553/(Tg+460) - 1.43e-6 P - 0.5996 ln(P) + 20.9/P - 7798/P^2 - 0.15963(Tg+460))
If Tg>32 °F then the following equation is used for K_N2:

K_N2 = e(8.83643 - 2551.94/(Tg+460) - 0.00725046P)
Otherwise the following equation is used for K_N2:
K_N2 = e(74.02 - 45325/(Tg+460) + 5.56e-5 P - 0.32788(ln(P))^2 + 815.4/P - 29564/P^2 + 1185861 ln(P)/(Tg+460)^2)
K_CO2 = -0.025+194.99/P-1.085e5/P2+(-4.29 + 269.82/P - 2.38e6/P2)/(Tg + (-57.03 - 4.99e-3 P))
K_H2S = e(11.28 - 4.34 ln(P)) Tg^(-2.15 + 0.914 ln(P))
Following the calculation of the K-Factors, the sum of ΣYi/Ki is calculated. Where the Yi is the mol
fraction of the gas component and Ki is the corresponding K-Factor. If the sum is not equal to 1,
a new estimate of hydrate temperature is made and the K-Factor calculations are performed
again.
HYD.2.2 Poettmann (1989) – Mann (1988)

In 1988, Susan Mann published a thesis describing a new set of K-Factors representing the vapor-
solid equilibrium conditions for gas-hydrate mixtures. Her results were subsequently published
by Poettmann et al in 1989 (ref1: Mann, S., L., Vapor-Solid Equilibrium Ratios for Structure II
Natural Gas Hydrates, Colorado School of Mines, 1988. Ref2: Poettmann, F., H., Sloan Jr., E., D.,
Mann, S.,L., McClure, L.,M., Vapor–solid Equilibrium Ratios for Structure I and II Gas Hydrates,
Proceedings 68th Annual GPA Convention, San Antonio, TX, 1989).
The general form of the Poettmann equation for K-Factor is as follows:
Ln(K) = A+B(SG)+C(Tg)+D/P+E/P2+F(0.001P)2+G/SG+H(0.001P)3+I(SG)(P)+J(ln(P))+L(P)+M/Tg
The following table lists a subset of the constants used in the Poettmann equation.
Table HYD-1: Example of Poettmann/Sloan Constants
For a complete list of the constants and equations, refer to the Mann thesis. The determination
of hydrate temperature is iterative and follows the technique described in Section HYD.2.1.
HYD.3 Hydrate Temperature Depression

The hydrate temperature will be naturally depressed by the presence of NaCl in the free water –
water of condensation is fresh. The main technique used to depress the hydrate temperature is
to add an inhibiter to the flow stream.
HYD.3.1 Salinity Temperature Depression – McCain (1990)

The SG-based routines include a calculation of the hydrate suppression capability of saline water.
In 1959, Katz published a curve that presented the hydrate temperature depression caused by
water containing sodium chloride. McCain published a correlation that was generated from the
Katz curve (ref: McCain, W., D., The Properties of Petroleum Fluids, PennWell Books, 1990).
The correlation generates a salinity-based hydrate temperature reduction, ΔTs, in °F, for a given
water salinity, in percent.
ΔTs = c1NACL% + c2NACL%2 + c3NACL%3

c1 = 2.20919 - 10.5746 SG + 12.1601 SG2
c2 = -0.106056 + 0.722692 SG - 0.85093 SG2
c3 = 0.00347221 - 0.0165564 SG + 0.019764 SG2
HYD.3.2 Basic Inhibitor Temperature Depression

The SG-based and composition-based hydrate temperature routines include a calculation of the
hydrate temperature suppression resulting from the addition of inhibitors. For a more rigorous
calculation of the effects of inhibitors, refer to Section HYD.4, describing the module to calculate
inhibitor volumes.
To estimate the effect of the inhibitor, the inhibitor is specified using the ‘Hydrate Inhibitor Type’
table. This version of the Hydrate Analysis tool is limited non-ionic inhibitors, namely: methanol,
ethanol, monoethylene glycol (MEG), diethylene glycol (DEG) and triethylene glycol (TEG).
The relevant properties of these inhibitors are presented in Table HYD-2.
Chemical Methanol Ethanol MEG DEG TEG Water

Chemical Formula CH4O C2H6O C2H6O2 C4H10O3 C6H14O4 H2O
Molar Mass 32.04 46.069 62.068 106.12 150.174 18.015
Density, g/cm3 0.792 0.7893 1.1132 1.118 1.1255 1
Density, lbm/gal 6.503 6.587 9.29 9.33 9.393 8.345
Constant, KH 1297 1297 1500 2222 3000 -
Constant, CM 0.21 0.21 -1.25 -8 -15 -
Table HYD-2: Inhibitor Properties
The constants listed in the table are used in the appropriate inhibitor model.
The temperature suppression of an inhibitor is a function of its mol% concentration in the water
phase. The performance of the inhibitors is presented in Figure HYD-5.
Figure HYD-5: Inhibitor Hydrate Temperature Suppression (after Carroll)
There are three inhibitor models available in the Hydrate Analysis tool for determining the
hydrate temperature suppression.
HYD.3.2.1 Hammerschmidt (1934), mod: Pedersen (1989)

In 1934 Hammerschmidt published a model to predict inhibitor performance (ref:
Hammerschmidt, E.,G. “Formation of Gas Hydrates in Natural Gas Transmission Lines”, Industrial
and Engineering Chemistry, 1934). The Hammerschmidt constant, KH, presented with the original
equation was modified by Pederson and are those presented in Table HYD-2 (ref: Pedersen, K.,S.,
Fredenslund, A., Thomassen, P., Properties of Oils and Natural Gases. Gulf Publishing, Houston,
TX, 1989).
The Hammerschmidt equation is widely used and is presented below.
ΔTi = KH wt% / MW(100 – wt%)
Where ΔTi is the resulting hydrate temperature depression, in °F, KH is the Hammerschmidt
constant listed in Table HYD-2, MW is the molecular weight of the inhibitor and wt% is the
concentration of inhibitor in weight%.
HYD.3.2.2 Nielsen-Bucklin (1983)

In 1983, Nielsen and Bucklin developed an alternative model for methanol inhibitors that was
based on first principles (ref: Nielsen, R.,B. and Bucklin, R., W., "Why not use methanol for
hydrate control?", Hydrocarbon Processing, Vol. 62, No.4, 71-78, April 1983).
Nielsen and Bucklin presented a theoretical basis for the development of the Hammerschmidt
equation for low methanol concentrations. They derived the Hammerschmidt equation using the
freezing point depression of an ideal solution and truncating the higher order terms. They
suggested that the Hammerschmidt equation was only valid at methanol concentrations up to
0.20 mass fraction. For higher methanol concentrations, Nielsen and Bucklin developed the
following equation for ΔTi, in °F, based on the mole fraction, xi, of inhibitor.
ΔTi = -129.6 ln(1 – xi/100)
xi = [wt%/MWi] / [wt%/MWi + (100 – wt%)/MWw]
Where wt% is the weight percent of inhibitor, MWi is the molecular weight of the inhibitor (Table
HYD-2) and MWw is the molecular weight of water (18.01523).
Nielsen and Bucklin claimed their equation was valid to 90 wt% methanol.
From 1983 to 1987 the Gas Processors Association sponsored research to measure hydrate points
and corresponding effects of inhibitors such as methanol and glycol. The result of this study was
that the Gas Processors and Suppliers Association (GPSA) Engineering Data Book recommends
the Hammerschmidt equation is valid up to 25 wt% methanol concentrations. The Nielsen-
Bucklin equation was recommended for methanol concentrations ranging from 25-50 wt%.
HYD.3.2.3 Carroll (2003)
The Hammerschmidt and Nielsen-Buckner models were developed to model methanol as the
inhibitor. In 2003, Carroll published a modification to the Nielsen-Buckner model that extended
the validity of the equation to all the commonly used inhibitors (ref: Carroll, J., Natural Gas
Hydrates A Guide for Engineers, Gulf Professional Publishing, 2003).
The basis for the Carroll model was the same as that for the Nielsen-Bucklin equation with the
addition of an activity coefficient to account for the type and concentration of the inhibitor. The
Carroll model yields a hydrate depression, ΔTi in °F, based on the mole fraction, xi, of inhibitor as
follows.
ΔTi = -129.6 [CM xi2 + ln(1 – xi)]
The constant CM (Table HYD-2) was called the Margules constant by Carroll.
HYD.4 Hydrate Inhibitor Volume

The PE² Essentials Hydrate Analysis Tool includes an option for calculating the volume of inhibitor
required to depress the hydrate temperature by a given amount (Figure HYD-6). This module
includes the option to include a “Safety Factor” to be considered when calculating the volumes
of inhibitor.
Figure HYD-6: PE² Essentials Hydrate Analysis Tool – Inhibitor Volume
The calculation of inhibitor volume is made up of the following: the amount of inhibitor required
in the water phase; the temperature depression resulting from salinity of the free water; the
inhibitor loss due to vaporization; and inhibitor absorption by the condensate.
The temperature depression caused by water salinity was presented in Section HYD.3.1.
HYD.4.1 Inhibitor Required for Water Phase

There are two sources of water in the production stream: free water and water of condensation
from the gas. The free water has salinity and has a partial protection for hydrate formation
(Section HYD.3.1) but still needs to be inhibited. The water of condensation is fresh and, as a
result, is prone to forming hydrates.
The water measured at surface is a combination of free plus condensed water. To determine the
amount of condensed water that is generated from the gas, set the ‘Free Water Rate’ to zero and
run the tool. The ‘Condensed Water’ value can then be used to determine the actual free water
rate to be entered for the final run.
Note that the free water rate is assumed to be constant but condensed water rate is a function
of the gas rate. The tool is not a forecasting tool so using Water-Gas Ratio for free water rate is
not warranted.
The calculation of condensed water is based on the correlations published by Moshfeghian (ref:
Moshfeghian, M., Lean Sweet Natural Gas Water Content Correlation, Tips of the Month,
PetroSkills, John M. Campbell & Co., September 2014).
Water Content of gas (Figure HYD-7), in bbls/mmscf, at a pressure, P in psi, and temperature, T
in °F, is given by the following equation for the temperature range of 41°F to 149°F.
WContent = 2.617 A/P + 0.1781 B
A = 3698.8338 e[-(0.007248 T – 2.8009)^2]

B = 5.6269 e[-(0.006418 T – 2.4165)^2]
Refer to the reference for the A and B equations for other temperature ranges.
Figure HYD-7: Water Content of Gas
The water resulting from condensation (WCondensation) is the difference between the
WContent of the gas at initial temperature/pressure and at operating temperature/pressure.
The free water is assumed to be constant and independent of the gas rate and is converted to
lbm/mmscf as follows - assuming 1 mmscf.
FreeWater [lbm/mmscf] = (bbl/d) (42 gal/bbl) (8.345 lbm/gal) / 1mmscf/d
The minimum amount of inhibitor required to supress the hydrate temperature is based on the
total volume of water (WTotal = WCondensation + FreeWater) that is to be inhibited and the type
of inhibitor.
The temperature suppression of an inhibitor is a function of its concentration in the water phase.
The performance of the inhibitors, presented in Figure HYD-5, was re-plotted as Figure HYD-8.
Figure HYD-8: Inhibitor Hydrate Temperature Suppression (after Carroll)
To simplify calculation of the minimum inhibitor required, in wt%, to suppress the temperature,
ΔT in °F, by a specified amount, the following EPCI-generated equations can be used.
Methanol:
wt% = 2.637e-5 ΔT3 - 0.008898 ΔT2 + 1.3239 ΔT + 0.2273
Ethanol:
wt% = -4.4305e-7 ΔT4 + 1.507e-4 ΔT3 - 0.02193 ΔT2 + 1.8989 ΔT + 0.218
MEG:
wt% = 1.519e-08 ΔT5 - 5.181e-6 ΔT4 + 0.0007027 ΔT3 - 0.05068 ΔT2 + 2.414 ΔT + 0.4411
DEG:
wt% = -7.262e-10 ΔT6+2.736e-7 ΔT5-4.067e-5 ΔT4+0.003066 ΔT3-0.1282 ΔT2+3.389 ΔT+1.168
TEG:
wt% = -1.3e-9 ΔT6 + 4.676e-7 ΔT5 - 6.652e-5 ΔT4 + 0.004671 ΔT3 - 0.1783 ΔT2 + 4.048 ΔT + 1.933
After the wt% of inhibitor required to suppress the hydrate temperature is determined, the
minimum volume of inhibiter required is calculated based on the total water in the system.
InhibitWater [lbm/mmscf] = wt% * WTotal / (100 - wt%)
It should be noted that the calculation of wt% includes any safety factor included in the input
parameters. The value for InhibitWater is the minimum amount of inhibitor required to protect
the system based solely on the water content.
HYD.4.2 Inhibitor Lost to the Vapor Phase

Inhibitor lost to the vapor phase is a problem when using methanol as an inhibitor. Vapor loss for
the other inhibitors (MEG, DEG, TEG) is orders of magnitude less than what occurs for methanol
and is ignored in the Hydrate Analysis tool calculations. No information is available on Ethanol,
but it is not commonly used for hydrate prevention.
The calculation of methanol vapor loss is based on the correlations published by Moshfeghian
(ref: Moshfeghian, M., A Simple Model for Estimation of Methanol Loss to Vapor Phase, Tips of
the Month, PetroSkills, John M. Campbell & Co., August 2011)
The calculation of loss to the vapor phase is based on K-values. The K-value is defined as the mole
fraction of methanol in vapor phase, yv, divided by the mole fraction of methanol in the aqueous
liquid phase, xa, K-value = yv/xa. K-value is calculated from the following correlation.
K-value = 1/P* e(5.37 (1 + omega) (1 - 1/T*))
T* = (T + 459.69) / 615
P* = P / 35
omega = 2.95 - 0.02607 P* + 8.92828e-5 P*2 - 0.851257 / T*
The estimated K-value is used to calculate the mole fraction of methanol in the vapor phase based
on the mole fraction of methanol in the aqueous phase, which is calculated from the weight% of
methanol in the aqueous phase. The vapor loss is calculated in lbm/mmscf as follows.
xa = wt%/32.04 / [wt%/32.04 + (100 – wt%)/18.015]

yv = xa K-value
VaporLoss [lbm/mmscf] = 84471.4 yv
The constant 84471.4 is a conversion factor obtained as follows.

lbm/mmscf = (yv [lbmole MeOH / lbmole of gas]) (32.04 lbm/lbmole MeOH) (lbmole / 379.3scf gas) (106
scf/mmscf gas)
Since the effect of gas composition is small, the K-value correlations are expressed in terms of
pressure and temperature (Figure HYD-9).
Figure HYD-9: Methanol K-Value for Vapor Phase
An example of the resulting value of lbm/mmscf for methanol lost to the vapor phase for a system
at 40°F is presented in Figure HYD-10.
Figure HYD-10: Methanol Lost to Vapor Phase at 40°F
Note: This calculation is valid only for methanol. No loss is assumed for the other inhibitors.
HYD.4.3 Inhibitor Lost to Condensate Phase

The calculation of the inhibitor lost to the condensate is based on data published by Ng et al (ref:
Ng, H., J., Chen, C., J., Vapour-Liquid and Vapour-Liquid-Liquid Equilibria for H2S, CO2, Selected
Light Hydrocarbons, and a Gas Condensate in Aqueous Methanol or Ethylene Glycol Solutions,”
GPA Research Report RR-149, Gas Processors Association, Tulsa, Gas Research Institute, Chicago,
1995). The published data was for methanol and MEG. Since no other published data is available
for the other inhibitors, the MEG correlation is used for all other inhibitors in the Hydrate Analysis
tool.
The calculation of inhibitor lost to the condensate phase is based on K L-values. The KL-value is
defined as the mole fraction of methanol in the condensate, yc, divided by the mole fraction of
methanol in the aqueous liquid phase, xa, KL=yc/xa.
The Ng data indicated that the KL-value was only a function of temperature (Figure HYD-11). The
correlation equations, in terms of temperature, T in °F, are as follows.
MeOH: KL-value = e[5.9 – 5404.5 (1/(T + 459.69))]

MEG: KL-value = e[4.2 – 7266.4 (1/(T + 459.69))]
Figure HYD-11: Methanol and MEG KL-Value for Condensate Phase
The estimated KL-value is used to calculate the mole fraction of the inhibitor in the condensate,
yc, based on the mole fraction of inhibitor in the aqueous phase, xa. The mole fraction in the
aqueous phase is calculated from the weight% of inhibitor in the aqueous phase, calculated in
Section HYD.4.1.
xa = wt%/MW / [wt%/MW + (100 – wt%)/18.015]

yc = xa KL-value
wt%_Inh = yc MW / (90 + yc (MW - 90))
CondLoss [lbm/mmscf] = 349.86 SGcond CGR wt%_Inh
MW is the molecular weight of the inhibitor from Table HYD-2 wt%_Inh is the weight percent of
inhibitor in the condensate and the 349.86 is a conversion factor [(42gal/bbl) (8.33lbm /gal)].
HYD.4.4 Total Inhibitor Required

After calculation of the appropriate volumes, the total inhibitor required to protect the system
to the given temperature is the sum of the volumes.
Total Inhibitor [lbm/mmscf] = InhibitWater + VaporLoss + CondLoss
The volume of inhibiter is calculated using the density of the inhibitor.
Inhibitor Volume [gal/mmscf] = Total Inhibitor [lbm/mmscf] / Inhibitor Density [lbm/gal]
Daily inhibitor injection requirements are then determined based on the gas rate.
Inhibitor Rate [gal/day] = Inhibitor Volume [gal/mmscf] Gas Rate [mmscf/d]
If a commercial inhibitor solution is being used which has an inhibitor concentration of less than
100%, the concentration of the solution needs to be considered when calculating the required
injection rates.
HYD.4.4.1 Example
A subsea gas pipeline has wellhead conditions of 195oF and 1050 psia. The gas flowing through
the pipeline has a specific gravity of 0.7 and is cooled by the surrounding water to a seabed
temperature of 38oF. There is a pipeline pressure drop to 950 psia at the outlet. Gas exits the
pipeline at a rate of 10 mmscf/d. The gas produces condensate at a ratio of 7.8 bbl/mmscf, with
an average density of 40°API. The gas well produces 5 bwpd free water with a salinity of
35,000ppm. The hydrate temperature of the gas is 65oF.
Find the rate of inhibitor injection needed to prevent hydrate formation in the pipeline. For this
example, the use of methanol will be worked out.
• From Section HYD.3.1: ΔTs = 2.5oF for a salinity of 3.5%, therefore the required ΔT for
inhibitor protection is reduced from 27oF to 24.5oF
• From Section HYD.4.1, WContent of gas at 1050 psi and 195 oF is 1.6 bbl/mmscf and
WContent of gas at 950 psi and 38 oF is 0.02 bbl/mmscf. Net water of condensation is 1.5
bbl/mmscf or 15.7 bwpd at 10 mmscf/d
• The free water rate is assumed to be a constant and assuming gas of 1 mmscf/d, then
Total water = (1.5 bbl/mmscf + 5 bbl/10 mmscf) (42 gal/bbl) (8.345 lbm/gal). So WTotal =
701 lbm/mmscf
• The weight percent of methanol required to inhibit hydrates to a ΔT of 24.5 is wt% = 27.7%
• MeOH to inhibit water = wt% WTotal / (100 - wt%). So InhibitWater = 268.6 lbm/mmscf
• Methanol lost to vapor (Section HYD.4.2) is calculated at T = 38oF and P = 950 psi: T* =
0.8092, P* = 27.143, omega = 1.2562 and xa = 0.177. Results are K-value = 0.002117, yv =
0.0003752 and VaporLoss = 31.7 lbm/mmscf
• Methanol lost to condensate (Section HYD.4.3) is calculated where CGR = 7.8, SG cond =
0.8251, KL-value = 0.00702, xa = 0.177, yc = 0.001242 and wt%_inh = 0.0004425. So
CondLoss = 0.996 lbm/mmscf
• Total inhibitor required = InhibitWater + VaporLoss + CondLoss. So total MeOH required
is 301.3 lbm/mmscf. For 10 mmscf/d, the 100% inhibitor injection would be 3013 lbm/day
or 463 gal/day or 10.5 bbl/day or 19 gal/hr. If required, these values would have to be
modified to take into account the weight percent of the commercial solution used.
224 PE Forecast Essentials
PE Forecast Essentials
The Forecast Essentials section contains the following:
• Horizontal Hydraulically Fractured Well Forecasting
• Basic Reservoir Simulator
• Stream Tube WaterFlood Simulator
• Miscible/Immiscible CO2 WAG WaterFlood Simulator
• Gas Material Balance – Multi-Tank
• Oil Material Balance
• Decline Curve Analysis
• Monte Carlo Simulation: Decline Curve Production Forecast
Unconventional Forecast Simulator Tool

The PE² Essentials ‘Unconventional Forecast’ tool includes models that are used to generate
forecasts for hydraulically fractured horizontal, unconventional oil and gas wells (Figure UNC-1).
Figure UNC-1: PE² Essentials – Unconventional Forecast Tool

The unconventional forecasting tool incorporates the following:

• Analytical Model: an unconventional, hydraulically fractured well forecasting engine
based on SPE144093 type curves.
• Numerical Model: a reservoir simulator based engine to generate forecasts for
unconventional, hydraulically fractured wells. The simulator is based on the US DOE
BOAST simulator (Ref: DOE/BC--89/3/SP, DOE/BC/14831-18 and NIPER-542).
• Both gas and depletion drive oil material balance models are included in the Analytical
Model.
• A Sales Gas model to predict wellhead (sales gas and condensate) properties of the raw
gas stream based on a surface flash.
• Gas well tubing pressure drop correlations: Average TZ for single phase gas, Guo-
Ghalambor to model mist flow of gas/condensate/oil and water and Modified Hagedorn
& Brown for high liquid yield gas wells (Note that annulus flow is not modeled in this
version).
• Oil well tubing pressure drop correlation is limited to Modified Hagedorn & Brown.
• Frac water flow back modeling for gas wells can be incorporated when using the Analytical
Model. This is an empirical pseudo model and includes an "acceleration" factor to either
increase or decrease the rate of frac water recovery and ultimate volume of frac water
recovered. This model will account for the additional tubing pressure drop caused by
production of frac water.
• A History Match tool is included that can use either the Analytical or Numerical Model to
history match an oil or gas well. This tool can be used to evaluate the net effectiveness of
the frac program (as well as history match other reservoir parameters).
• The integrated Economics Tool can be used to optimize completion parameters and
determine the economically optimum number (in terms of PIR) of fracs to place in the
well.
• PE² Essentials - Unconventional Forecast can be run in either Oilfield or Metric units (the
program will convert the units, so the numbers do not have to be re-entered, if a run in
both units is required).
Tool output (Figure UNC-1) includes frac ‘Spacing’ in feet/meters and frac ‘Interference’ time in
days. These parameters are calculated as follows.
FracSpacing = Lateral Length / #Fracs
FracInter in days = 9.875 φµc FracSpacing2 / k
The FracInter equation above assumes that FracSpacing is entered in feet. For meters, the value
of the constant, 9.875, becomes 106.3. To calculate FracInter in hours, the constant would be
237 for ft and 2551.2 for m.
226 Unconventional Forecast Simulator Tool
UNC.1 Analytical Model

The Analytical Model forecast engine is based on the Productivity Index (PI) type curves published
by LeBlanc et al (LeBlanc, D., Martel, T., Graves, D., Tudor, E. and Lestz, R., Application of Propane
(LPG) Based Hydraulic Fracturing In The McCully Gas Field, New Brunswick, Canada, SPE 144093,
June 2011). The type curves were also presented by LeBlanc et al in World Oil (LeBlanc, D.,
Huskins, L., Lestz, R., Application of Propane Based Hydraulic Fracturing in the McCully Gas Field,
New Brunswick Canada, July 2011).
The dimensionless PI type curves incorporated into the Unconventional Forecast tool’s Analytical
Model are shown in Figure UNC-2.
Figure UNC-2: Dimensionless PI Type Curves for Hydraulic Fractures
The Analytical Model includes a gas reservoir material balance (P/Z) model and an oil reservoir
material balance model that assumes natural depletion of an oil reservoir with no aquifer or gas
cap. The analytical reservoir material balance models are used to generate the reservoir pressure
forecast for the reservoir as well as the GOR forecast for an oil reservoir.
For gas reservoirs, the material balance formulation is a straightforward P/Z versus cumulative
production for a volumetric gas reservoir. For oil reservoirs, the material balance formulation is
more complex. Only a solution gas drive is considered in the current version of the
Unconventional Forecasting model.
For an oil reservoir, a krg/kro versus saturation model is required to determine the kro/krg
development as saturations change. The empirical equations for 2-phase (oil/gas) relative
permeabilities presented by Honarpour et al (Honarpour, M., M., Koederitz, L., F. and Harvey, A.,
H., “Empirical Equations for Estimating Two-Phase Relative Permeability in Consolidated Rock”,
Transactions AIME, 1982) are used to predict relative permeability variations in terms of
saturation (Figure UNC-3).
Figure UNC-3: krg/kro versus Saturation
These equations for krg/kro are as follows.
krg/kro = a / b (UNC-1)
a = 0.98372 (So / (1 - Swi))4 [(So - Sorg) / (1 - Swi - Sorg)]2
b = 1.1072 KrgSorg [(Sg - Sgc) / (1 - Swi)]2 + 2.7794 KrgSorg Sorg(Sg - Sgc) / (1 - Swi)
Where: So is the current oil saturation, Sorg is the residual oil saturation in an oil/gas system, Swi
is the initial water saturation, KrgSorg is the gas relative permeability at Sorg, Sg is the current gas
saturation and Sgc is the critical gas saturation.
Once the krg/kro value is determined for a given So, the value for kro can be calculated based on
Corey’s method (Corey, A., T., “The Interrelation Between Gas and Oil relative Permeabilities”,
Producers Monthly, 1954). This value is required for productivity calculations using the type
curves. Corey’s equation for kro in oil/gas systems as follows.
kro = [1 – Sg/(1 - Swi)]4 (UNC-2)

Where: Sg is the current gas saturation and Swi is the initial water saturation.
The material balance forecast for oil reservoir pressure and GOR performance is determined
using Tarner’s Method as presented by Mian (Mian, M., A., Petroleum Engineering Handbook for
the Practicing Engineer, PennWell Publishing, 1992).
The Tarner technique involves an iterative procedure to predict GOR at a given pressure assuming
a value for Np and Gp. The requirements for the technique are solution Gas-Oil Ratio, RS, versus
pressure correlation and krg/kro versus saturation models. RS is calculated using equations
presented in the PVT tools information.
Starting with estimated pressure and GOR for a given Np and Gp, the iterative equation is:
Npn = a/b + c/d (UNC-3)

a = N[Bo - Boi + Bg(RSi - RS)] + G(Bg - Bgi)
b = Bo - BgRS + 0.5Bg(GORn – GORn-1)
c = Bg[Gpn-1 – 0.5Npn-1(GORn + GORn-1)]
d = Bo - BgRS + 0.5Bg(GORn + GORn-1)
Where: i is initial condition, n is the current step, n-1 is the previous step, N is the OOIP, G is the
GIIP, Np is cumulative oil production, Gp is the cumulative gas production, GOR is producing gas-
oil ratio and RS is solution GOR.
After calculating Npn, Son is calculated as follows:
Son = Bob(N - Npn) (1 - Swi - Sgc) / NBoi (UNC-4)
With the calculated Son, the krg/kro value is obtained from the krg/kro model (Figure 5-3). The GOR
for this step is then calculated as follows.
GORn = RSn + (krg/kro) (µo/µg) (Bo/Bg) (UNC-5)

Where: µo, µg, Bo and Bg are evaluated at the current pressure.
If GORn and the estimated GOR are in agreement, the next step is performed; otherwise new
estimates are made, and the process is repeated.
The material balance models generate the reservoir pressure, and GOR development, for the
reservoir. The dimensionless PI type curves are then used to generate the well deliverability over
time based on the estimated reservoir pressure, GOR, and relative permeability values (kro)
generated during the material balance calculations. The type curves include the early time, non-
stabilized performance of a hydraulically fractured well as well as the transitional and stabilized
periods. As a result, the full life of the well can be modeled.
These dimensionless type curves (Figure UNC.2) were generated through the use of a commercial
reservoir simulator. Numerous forecasts were generated for a large number of reservoir
parameters, fracture conductivity, kfw, and fracture half length, xf, values. The simulation results
were converted to dimensionless parameters and used to generate type curves for the
dimensionless productivity index, JD, and dimensionless time, tD, as a function of dimensionless
fracture conductivity, FCD (Note this parameter is referred to CfD in the Hydraulic Fractur Analysis
tool).
Dimensionless productivity index, JD, is used since it is not impacted by: the length of the transient
period; whether or not the well is exhibiting radial flow; whether or not the well is hydraulically
fractured; and whether the wellbore is vertical or horizontal. Using dimensionless productivity
index, it is possible to use data from any well without additional normalization.
The dimensionless equations are as follows:
Oil: JD = PI (141.2 Bo µo / ko h) (UNC-6)

Gas (P2): JD = PI (1424 µg Z T/ kg h) (UNC-7)
tD = 0.0036 k t / (φ µ ct rw2) (UNC-8)
FCD = (kf wf) / (k xf) (UNC-9)
Where: T is temperature in ˚R, µo and µg are viscosities in cp and t is time in days.
The calculation procedure is as follows: 1) For a given time, calculate t D; 2) The JD for the
corresponding FCD is obtained from the type curves; 3) From the cumulative production from
the previous time step, the material balance data is used to obtain reservoir pressure and GOR;
4) With the parameters entered in the ‘PVT’, ‘Reservoir’, ‘Frac’ and ‘Wellbore’ models, the
productivity index (PI) is calculated from JD; 5) The production rate and flowing pressure are
calculated (‘Generate Forecast’).
During the forecast, once the ‘Minimum Rate’ is reached the well is shut in. This logic is based on
assuming only natural flow from the well. Since most oil wells need to be pumped, the oil model
generates a forecast based on a minimum BHP rather than THP. After the oil rate and flowing
bottomhole pressures are generated, the corresponding THP is calculated.
The Analytical Model generates 50-year forecasts. The resulting forecast can be saved to the PE
Tools database or to a comma-delimited file (csv) by clicking ‘Export Results to CSV’ for plotting
using PE² Essentials Chart (Figure UNC-4) or imported into Excel for plotting.
Figure UNC-4: Chart Plot of Analytical Model Forecast

UNC.2 Numerical Model

The Numerical Model (Figure UNC-5) forecast engine is reservoir simulator based and is an
implementation of the US DOE BOAST II reservoir simulator (Franchi, J., R., Kennedy, J., E., and
Dauben, D., L., BOAST II: A Three-Dimensional, Three Phase Black Oil Applied Simulation Tool, US
Department of Energy Report DOE/BC-88/2/SP, 1987). This is a public domain simulator and the
fortran code can be obtained from the US DOE.
This implementation of the simulator is a stripped-down version of BOAST II specifically

programmed to model a hydraulic fracture. For a description of a general-purpose version of this
simulator, refer to the PE2 Essentials Basic Reservoir Simulator tool.
Figure UNC-5: Unconventional Forecast Model – Numerical Model
To run the numerical model, a single fracture simulation model is built internally using the
parameters entered in the ‘PVT’, ‘Reservoir’, ‘Frac’ and ‘Wellbore’ models. The simulation time
in years is entered and the simulator is executed (‘Generate Forecast’).
If ‘Save Simulation Run Data’ is checked, a file will be generated as the run progresses that
includes time step and convergence data for every time step. This file can be examined to confirm
progress of the run.
The simulation model includes a horizontal well in a 3D rectangular grid, with the horizontal well
in the middle of the structure. The gridding is geometric and the top xy view of the model
(showing the location of the hydraulic fracture) is shown in Figure UNC-6.
Figure UNC-6: Top View – Numerical Model Grid
The Figure UNC-6 grid represents one fracture stage. The total x-distance would be the drainage
distance for this fracture stage. The total y-distance would be the drainage in the y-direction, in
other words, the well spacing. X-distance multiplied by the y-distance would be the drainage area
for the fracture stage.
In Figure UNC-6, the black line represents the completed well and the white line represents the
fracture location and its total length is 2xf. The assumption is that the well is perforated at the
fracture location but there is unstimulated reservoir communication behind casing for ~50% of
the stage length.
Figure UNC.7 presents the xz cross-sectional view along the well location, showing the well
completion in black, as well as the hydraulic fracture location in white. For the simulation model,
the fracture is assumed to exist over 100% of the pay interval and the well is placed in the middle
of the pay interval.
Figure UNC-7: Cross Section View – Numerical Model Grid
The simulation of a single-phase gas reservoir is relatively quick. Simulation of a multiphase oil
reservoir is slower because this reservoir simulator is a basic IMPES simulator. When using the
Numerical Model, the length of the forecast period has to be entered before the run will start.
For gas reservoirs, 50-year forecasts are quick but for oil reservoirs, a 20-year forecast may be
more practical.
In this version, the Numerical Model will only generate a forecast for one set of hydraulic
fractures at a time - for example, a well with 5 fracs. To generate a forecast for different fracture
scenarios, separate runs are required.
The forecast results can be saved to the PE Tools database or to a comma-delimited file (csv) by
clicking ‘Export Results to CSV’ for plotting using PE² Essentials Chart (Figure UNC-4) or imported
into Excel for plotting.
UNC.3 PVT Model

The Gas PVT Model (Figure UNC-8) includes a flash analysis to generate gas properties to allow
the forecasting of sales gas, condensate and water of vaporization from the raw gas. Refer to the
PVT Tool for additional information on gas component analysis.
Figure UNC-8: Unconventional Forecast – Gas PVT Model
The PVT gas components can be entered manually or imported from a PE² Essentials PE Tools
database. Clicking ‘Import PE Tools db Components’ will list the available component data stored
in the PE Tools database (Figure UNC-9).
Figure UNC-9: Importing Gas Components from the PE Tools Database
The entered gas components should represent the recombined raw gas stream. This information
is used to evaluate the liquid content of the raw gas, the shrinkage factor, the gas gravity and the
gross heating value of the sales gas.
If gas compositions are not available, pseudo gas components can be generated from the gas
gravity by clicking ‘Generate Gas Compositions’ (Figure UNC-10). Refer to PVT Tools for more
information on generation of gas compositions.
Figure UNC-10: Generating Gas Compositions for Gas PVT Model
To calculate the sales gas components, it is assumed that 25% of the propane, 50% of the butane
and 99.5% of the pentanes+ are recovered from the raw gas stream.
The water content of the gas is used to in the Analytical Model to generate the net water
produced at surface at wellhead pressure and temperature conditions. The forecast will be a
function of the producing pressure and temperature entered on the main screen of the
unconventional forecasting model (Figure UNC-1).
For the Gas Model, the forecast results represent sales gas volumes. To access the raw gas
volumes, ‘Export Results to CSV’ after the run and import the resulting CSV file into a spreadsheet.
For the Oil PVT Model, separator pressure and temperature are entered as well as the oil
properties to correct the gas gravity for separator conditions (Figure UNC-11).
Figure UNC-11: Unconventional Forecast - Oil PVT Model
The oil PVT parameters can be entered manually or imported from a PE² Essentials PE Tools
database. Clicking ‘Import PE Tools db PVT Properties’ will open a sheet that will give the option
of entering PVT data stored with a well or PVT data from a stored PVT model (Figure UNC-12).
Figure UNC-12: Importing Oil PVT Data from the PE Tools Database
Selecting the appropriate button will list the options available for the database. Click on the
relevant well/tool and the PVT data will be imported into the tool. Note - To disable the gas
gravity correction routine, enter 114.7psi / 790.8kPa and 60˚F / 15.55˚C for the separator
conditions.
UNC.4 Oil/Gas Reservoir Models

The inputs for the Reservoir Model are self-explanatory. Because of the large variation in valid
parameters, the program does not test the input value to determine the validity of the reservoir
parameters. It is up to the engineer to ensure that valid parameters are entered.
Figure UNC-13 presents the Gas Reservoir Model and Figure UNC-14 presents the Oil Reservoir
Model.
Figure UNC-13: Unconventional Forecast - Gas Reservoir Model
Figure UNC-14: Unconventional Forecast - Oil Reservoir Model

The reservoir properties can be imported from one of the well models in the PE Tools database.
After clicking ‘Import PE Tools dB Reservoir Properties’ a screen will open listing the available
well models in the database.
‘Reservoir Length’ is used to determine the GIIP/OIIP and to construct the x-grids for the
Numerical Model. It is not necessarily the same value as the ‘Lateral Length’ entered in the
Wellbore Model.
‘Reservoir Width’ is the sum of the distance to the two drainage boundaries from the horizontal
well. This is also equivalent to the well spacing and is used to build the y-grids in the numerical
model.
The ‘Water RSW’ parameter is calculated based on the water salinity entered into the Oil PVT
Model (Figure UNC-11) and is used in the Numerical Model.
UNC.5 Fracture Model

The Frac Model is used to enter the hydraulic fracture parameters. Figures UNC-15 and UNC-16
show the fracture models for gas and oil models when using the Analytical Model and Figure
UNC-17 presents the fracture model for the Numerical Model.
Figure UNC-15: Unconventional Forecast - Fracture Model – Gas Model, Analytical Model
Figure UNC-16: Unconventional Forecast - Fracture Model – Oil Model, Analytical Model
Figure UNC-17: Unconventional Forecast - Fracture Model – Gas and Oil Numerical Model
In all cases, the frac parameters can be imported from the PE Tools database. Figure UNC-18 is a
schematic of the hydraulically fractured horizontal well model.
Figure UNC-18: Hydraulically Fractured Horizontal Well Model
From Figure UNC-18, Nf is the total number of fractures placed in the well, Lx is the reservoir
length, Ly is the reservoir width (well spacing) and xf is the fracture half length. Drainage area for
the horizontal well is Lx x Ly.
When using the Analytical Model, minimum and maximum hydraulic fracture values can be
entered so that a range of fractures can be modeled during one forecast run. If only a single
hydraulic fracture scenario is required, the same value for the minimum and maximum number
of fractures should be entered.
The Frac Model for gas includes an option to incorporate a ‘Frac Water Flowback’ model. The
flowback model is an empirical pseudo model with no theoretical basis. It is based on information
presented in the SPE paper (144093). The model is calibrated to the well's current producing (or
assumed initial) WGR which is entered in the Frac Model. The water flowback model also includes
a ‘Flowback Acceleration Factor’ to either increase or decrease the rate of frac water recovery
and the ultimate volume of frac water recovered. If using this model, make sure that the volume
of frac water produced is QC’d since there is no upper limit built into the model – the water
production is included in the csv results file generated by ‘Save Results’. If recovered volume is
too high, reduce the acceleration factor. The Frac Water Flowback Model has been included for
development planning (water handling) purposes.
The main purpose of the Frac Water Flowback Model is to include the additional tubing pressure
drop that occurs during the flowback of frac water. If this model is used, then either the Guo-
Ghalambor or Hagedorn-Brown tubing correlations should be used to model the wellbore
pressure drop.
UNC.6 Wellbore Model

The default tubing pressure drop correlation for a gas well is the Average TZ correlation, with the
option to use Guo-Ghalambor or Modified Hagedorn-Brown correlations. For an oil well, the only
available option is the Modified Hagedorn-Brown correlation.
Since production of gas, condensate and water is included when modeling a gas well, the Guo-
Ghalambor correlation has been included for multi-phase pressure drops. Although this
correlation includes up to 4-components: gas, water, condensate and sand; sand production is
assumed to be zero in the Unconventional Forecast Model.
Single phase gas well tubing correlations, like the Average TZ correlation, are normally valid when
WGR is less than 100 bbls/mmscf (561m³/106sm³). Since Guo-Ghalambor is a multi-phase
correlation, it may extend the range of WGR validity for a gas well, but caution should be used
when accepting the results for high WGR's. If the Frac Water Flowback Model is used, then the
more rigorous Modified Hagedorn-Brown correlation may be more appropriate.
In this version of the Unconventional Forecast Model, forecasts are generated for horizontal wells
only (Figure UNC-19). Well models can be imported from the PE Tools database.
Figure UNC-19: Unconventional Forecast - Horizontal Well Model
The ‘Lateral Length’ and the number of fractures entered in the Fracture Model, are used to
generate the x-distance, or the drainage distance for the fracture stage. Both the Analytical
Model and the Numerical Model generate a forecast for a single fracture. The total well
production is the summation of production from the total number of fractures.
The Analytical Model’s Gas Model forecast also calculates minimum gas flow rate required to lift
water or condensate based on the Turner correlation (refer to THP-BHP Gas Tool) and are
included in the CSV file when the results are exported to a CSV file.
To disable the tubing calculations and perform a forecast at bottomhole conditions, enter -1 for
‘Measured Depth to Top of Lateral’ and enter the value for flowing BHP for the flowing tubing
head pressure on the main screen.
UNC.7 Forecast Plotting

The ‘Unconventional Forecast’ tool includes a very basic plotting package to view overall forecast
results (Figure UNC-20). It is possible to save the plots to a png file by selecting “Save Graph”.
For more in-depth data analysis and more flexibility with plotting, export the results to a csv file
and use PE² Essentials Chart or import the file into a spreadsheet.
Figure UNC-20: Unconventional Forecast Plotting
The Plot Window can remain open during subsequent runs. To plot the new forecast data, click
the “Update” button.
UNC.8 Unconventional Forecast Economics Model

The Unconventional Forecast Model includes an Economics Model (Figure UNC-21). This model
is set up specifically to use the current unconventional forecast results, along with the entered
capital and op costs, to generate a basic economic analysis. For a general use economics model
refer to the Scoping Economics tool or CAPE.
Figure UNC-21: Unconventional Forecast - Economics Model
If a number of frac cases are included in the forecast, then the Economics Model allows the
engineer to evaluate the optimum economic number of fracs to be placed in the well.
The capital costs (Figure UNC-22) and operating costs (Figures UNC-23 - gas and UNC-24 - oil) are
entered separately. The Gas/Oil pricing is entered in the ‘Op Costs’ model and should be entered
in $US. All other costs should be entered in the local currency. The ‘Conversion Factor’ converts
the oil/gas prices to the local currency for calculations. Output is presented in the local currency.
Escalation factors and discount factor are entered on the main screen (Figure UNC-21).
Figure UNC-22: Economics Model – Capital Costs
Figure UNC-23: Economics Model – Operating Costs (Gas Well)

Figure UNC-24: Economics Model – Operating Costs (Oil Well)
Following the economics run, the ‘Optimum Economic #Fracs’ is presented on the main screen
(Figure UNC-21). This is calculated as the number of fractures that maximizes the Profit-to-
Investment Ratio (PIR).
PIR = CumCFdisc / TotalWellCosts (UNC-10)
Where: CumCFdisc is the discounted cumulative annual cash flow and TotalWellCosts is the capital
cost of the well (assumed to be sunk costs).
UNC.8.1 Total Well Costs

The total well costs are assumed to be sunk costs so no escalation or discounting is applied to
this cost. The well cost is calculated with Equation UNC-11.
TotalWellCosts = (DailyCost)(WellMD)/ROP + (#Fracs)(FracCost) + TieInCost (UNC-11)
Where: DailyCost is the daily drilling cost, WellMD is the measured depth of the well, ROP is the
rate of penetration, #Fracs is the total number of hydraulic fractures, FracCost is the cost per frac
and TieInCost is the cost to tie-in the well, clean up the well and any other costs associated with
the well.
UNC.8.2 Escalation Factors
The cost of a number of operating parameters can be escalated over time to take inflation into
account. The following parameters can be escalated:
• Operating Expense
o Head Office Overhead
o Fixed Well Costs
o Variable Well Costs
• Gas Processing Fee
• Gas Pipeline and Transportation Expense
• Oil/Condensate Transportation Fee
• Oil/Gas Price
The Annual Escalation Rates are entered on the main economics sheet and cost is escalated as
shown in Equation UNC-12.
EscFactor = (1 + EscRate/100)t (UNC-12)
Where: EscFactor is the annual escalation, EscRate is the escalation rate entered on the main
sheet (Figure UNC-21) and t is the time in years.
UNC.8.3 Discount Rate

To take the time value of money into account, all future revenue is converted to a common
reference point in time. This is assumed to be the current year or the present (hence the term,
‘present value’). This is achieved by discounting future cash flow. Discounting converts a future
sum of money into the equivalent of present-day cash.
The rate used for discounting future cash flow is called the discount factor and is entered as the
‘Annual Discount Rate’ on the main sheet (Figure UNC-21). The annual discount factor is
calculated with Equation UNC-13.
DiscFactor = (1 + DiscRate/100)-t (UNC-13)
Where: DiscFactor is the annual discount applied to the cash flow, DiscRate is the discount rate
entered on the main sheet (Figure UNC-21) and t is the time in years.
The present value, PV, of a net cash flow, CF, received at some future time, t, is calculated with
Equation UNC-14.
PVt = CFt DiscFactort (UNC-14)
Where the subscript t is the time in years at which the cash flow is received.
UNC.8.4 Oil/Gas Prices

For this version of the Economics Model, oil/gas price is assumed to be flat except for the
escalation rate that is entered. To enter future gas/oil prices, the ‘Scoping Economics’ tool should
be used.
The oil/gas prices for an oil well or the gas/condensate prices for a gas well are entered in the
‘Operating Costs’ model (Figure UNC-25).
Figure UNC-25: Economics Model – Oil/Gas Prices
Net gas price is calculated based on the base gas price, the heat content premium and any other
gas premium (or discounts) and then converted to the local currency. Net oil/condensate price is
converted to the local currency.
NetGasPrice = [(BaseGasPrice)(HeatContent) + Premium](LocalConversion) (UNC-15)
NetOilPrice = (OilPrice)(LocalConversion) (UNC-16)
The final net gas and oil prices are presented on the Operating Cost model sheet (Figures UNC-
23 and UNC-24).
UNC.8.5 CapEx, OpEx, Revenue, Net Present Value and PIR

For this version of the Economics Model, all capital expenditures, CapEx, are assumed to be sunk
costs so there are no escalations or discounting of the well costs. The economics forecast starts
at a negative value equal to the well costs.
The net gas and oil/condensate prices are escalated prior to being discounted. This esc/disc price
is then applied to the appropriate production stream to generate the future net present value
revenue stream.
NetGasPriceesc = (NetGasPrice)(EscFactorgas) (UNC-17)

NetOilPriceesc = (NetOilPrice)(EscFactoroil) (UNC-18)
NetGasRevenuedisc = (NetGasPriceEsc)(Qg)(ProdDays)(DiscFactort) (UNC-19)
NetOilRevenuedisc = (NetOilPriceEsc)(Qo)(ProdDays)(DiscFactort) (UNC-20)

TotalNetRevenuenpv = NetGasRevenuedisc + NetOilRevenuedisc (UNC-21)
The total net present value revenue, TotalNetRevenuenpv, is calculated on an annual basis and
added together to generate the project revenue for the complete production forecast for a given
fractured well. These calculations are performed for each fractured well case and the results are
added to the results table on the Economics sheet (Figure UNC-26).
Figure UNC-26: Economics Model – Results Table
The operating costs are escalated based on the escalation rates entered into the model (Figure
UNC-21).
TotalOpCostsEsc = (HeadOffice)(EscFactorOpEx) +
(FixedCost) (EscFactorOpEx) +
(VariableCosts)(Qg)(ProdDays)(EscFactorOpEx) +
(PipeLineFee)(Qg)(ProdDays)(EscFactorPipeline) + (UNC-22)
(GasProcessFee)(Qg)(ProdDays)(EscFactorProcess) +
(OilProcessFee)(Qo)(ProdDays)(EscFactorProcess)
The total cumulative, escalated operating costs are then discounted (Equation UNC-23) and
added to the results table on the Economics sheet (Figure UNC-26).
TotalOpCostsdisc = (TotalOpCostsesc)(DiscFactort) (UNC-23)
The net present value is calculated with Equation UNC-24:
NPV = TotalNetRevenuenpv – TotalOpCostsdisc – CapEx (UNC-24)
Profit to investment ratio is then calculated with Equation UNC-25:
PIR = NPV / CapEx (UNC-25)

Both NPV and PIR are presented on the results table on the Economics sheet (Figure UNC-26).
The ‘Optimum Economic #Fracs’ is determined as the number of fractures that maximizes PIR.
This value is reported on the main Economics sheet (Figure UNC-21).
If an additional forecast is generated in the main program, the economic analysis results will be
automatically updated when the Economics model is re-opened.
Once the economic runs are completed the resulting economic forecast information can be saved
to a csv file by ‘Save Econ Results’ and imported into PE² Essentials Chart (Figure UNC-27). A
separate file for each hydraulic fracture forecast is saved.
Figure UNC-27: Economics Model – Output Results
UNC.9 History Match Tool

The History Match Tool (Figure UNC-28) is a very powerful tool that enables the engineer to
evaluate the net effectiveness of the frac program (as well as other parameters). Refer to Section
UNC.10 for an example using this tool. Note the history match in Figure UNC-28 was generated
using the ‘History Match Model’ in the PE Tools database and Excel history file
(History_Fracs_NoWater.xlsx) is included in the “PE Essentials\Example Input Files\ Example
Input Files\Excel Files” directory.
Figure UNC-28: History Match Tool
The base model and frac parameters used in history matching are entered on the main screen.
The historical production data is imported through the ‘Import History’ button. History data can
be imported from the PE Tools database or an Excel Spreadsheet (Figure UNC-29). The Data
Import page will specify the units for the historical data. Refer to the info button on the Data
Import sheet for information on the file formats. If the gas model is being used, gas rate will be
plotted otherwise oil rate will be plotted (Figure UNC-28) after the data is loaded.
Figure UNC-29: import History Data
A history match of a number of parameters can be performed (Figure UNC-30). The history match
routine will report the optimum value for the chosen parameter based on the well’s production
history, which can then be compared to the expected value. It should be noted that the ‘Fracture
Half Length’ is always entered in feet.
When a history match parameter is chosen, the value from the model is shown as ‘Model Value’
on the main page (Figure UNC-28).
Figure UNC-30: History Match Parameters

If matching on ‘Number of Fractures’, ‘Fracture Perm’ or ‘Fracture Half Length’, the effectiveness
of the frac program can be evaluated by comparing the history match results to the expected
values.
It is possible to force a history match run on a specific value by entering the same value for the
minimum and maximum values.
The Numerical Model (reservoir simulator) can be used for history matching but it will increase
the time required to find the optimum parameter. The Analytical Model could be used to perform
a preliminary match, and then the numerical Model used to finalize the history match.
Note that the History Match Model uses BHP as the matching constraint. If only THP is available
for a gas well, the ‘THP-BHP Gas Well’ model can be used to generate the BHP and export the
history file for use in the History Match Model.
Following the history match run, the generated optimum history match parameter can be saved
to the model through “Save HM Parameter” which will transfer the matched parameter
(“Optimum Value”) to the original model. If the parameter is not saved, it will be discarded during
the next operation. Multiple parameters can be sequentially matched and saved but only one
parameter can be matched at a time.
The updated model, including any saved history match parameters, can be used for forecasting.
The updated model should be saved to retain the history match parameters.
None of the intermediate history match runs are saved. To save the history match results, re-run
the history match with the ‘Optimum Value’ set as the maximum and minimum values then save
the results by ‘Export HM Results to CSV’. This will save the history match results to a CSV file.
This file can be imported into PE Essentials Chart for plotting (Figure UNC-31).
Figure UNC-31: History Match / Historical Data Comparison

UNC.10 Unconventional Forecast Example

A very dry unconventional gas reservoir was tested by a horizontal well completed without fracs.
The production data (HistoryData_Example_Well With BHP.xlsx in “Book Examples\Example
Unconventional Forecast\Unconventional Forecast\History” directory) is presented in Table
UNC-1 and Figure UNC-32. The question is: what is the optimum number of fractures for a future
horizontal well in this reservoir.
Table UNC-1: Example Flow Data: Dry Gas Well Example
Figure UNC-32: Example Data

Step 1: Determine gas properties using the ‘Generate Gas Components’ in the PVT section of the
Unconventional Forecast Model.
The gas produced from the well had an average gas gravity of 0.6, CO2 mol% of 0.3, N2 mol% of
1.5 and there was minimal condensate production from the well.
The gas components were generated by clicking the ‘Generate Gas Components’. These
components were then modified to match the gas specific gravity of ~0.6 and CGR<1 (Figure UNC-
33).
Figure UNC-33: Gas Components – Matched Gas Gravity, Example Well
Step 2: Create Reservoir Model using estimated/known parameters (Figure UNC-34).
Figure UNC-34: Estimated Reservoir Model – Example Well

Step 3: Create Wellbore Model using known parameters (Figure UNC-35).
Figure UNC-35: Wellbore Model – Example Well
Step 4: Build the wellbore model and convert flowing THP to BHP using the THP-BHP Gas Well
Tool (Figure UNC-36 and Figure UNC-37). THP data is in the ‘THP to BHP Unconventional Example
Data.xlsx‘ located in the “PE Essentials\ Book Examples\Example Unconventional Forecast\THP-
BHP” directory.
Figure UNC-36: THP-BHP Model – Example Well

Figure UNC-37: Convert THP to BHP – Example Well
After conversion of THP to BHP, export the data to a CSV file. Import the BHP data into the history
file for the well (refer to ‘HistoryData_Example_Well With BHP.xlsx’ in the “PE Essentials\ Book
Examples\Example Unconventional Forecast\THP-BHP” directory).
Step 5: Select ‘History Match’ and import the history data from the Excel file (Figure UNC.38).
Perform a history match of the data to determine the permeability of the reservoir (Figure UNC-
39) and ‘Save HM Parameter’ which was 0.002md.
Figure UNC-38: Permeability History Match – Example Well

Figure UNC-39: Permeability History Match – Example Well
It should be noted that a history match could also be generated based on the number of hydraulic
fractures in the well (Figure UNC-40). This well was not hydraulically fractured so this history
match is rejected, and the permeability was matched instead (Figure UNC-39).
Figure UNC-40: Number of fractures History Match – Example Well

Step 6: Generate fracture parameters with the ‘Hydraulic Fracture Design Model’ (Figure UNC-
41).
Figure UNC-41: Hydraulic Fracture Design – Example Well
Save the model to the PE Tools database for import into the Unconventional Forecast model.
Step 7: Import the fracture design parameters into the Frac Properties. (Figure UNC-42)
Figure UNC-42: Hydraulic Fracture Parameters – Example Well

Step 8: Generate a forecast using ~100 acre well drainage and a range of hydraulic fractures
(Figures UNC-43 and UNC-44).
Figure UNC-43: Reservoir Parameters – 100 Acre Drainage, Example Well
Figure UNC-44: Unconventional Forecast – Example Well

Step 9: Enter the Economics Model and enter Capital Costs and Operating Costs (Figure UNC-45).
Figure UNC-45: Capital and Operating Costs – Example Well
Step 10: Run the Economics assuming a gas price escalation of 5%/year (Figure UNC-46).
Figure UNC-46: Economics Results – Example Well
The result is that at $2.75/mscf, the optimum number of hydraulic fractures would be 13.
Basic Reservoir Simulation Tool

The PE² Essentials ‘Basic Reservoir Simulator’ (Figure SIM-1) is a single well numerical simulator
based on the DOE BOAST (Black Oil Applied Simulation Tool) public domain simulator. A number
of versions of the DOE simulator exist and the PE² Essentials implementation of the simulator
incorporates a number of options from the different versions. An excellent reference on the
practical aspects of reservoir simulation is the book by Mike Carlson – Practical Reservoir
Simulation published by PennWell in 2003.
Figure SIM-1: PE² Essentials – Basic Reservoir Simulator Tool
The PE² Essentials Basic Reservoir Simulator can model recovery using a vertical or horizontal well
in a gas or oil reservoir. For a gas reservoir, only a dry, volumetric gas reservoir is modeled (no
retrograde condensate, no aquifer). For an oil reservoir, all phases can be modeled and there is
an option to model a dipping reservoir to allow both up dip gas-oil and down dip water-oil
contacts. Since the simulator is an IMPES solution, caution should be used if modeling gas coning
or water cusping.
The Unconventional Forecasting tool includes a stripped-down version of this simulator

specifically designed to model a single stage of a horizontal hydraulically fractured well.
The Basic Reservoir Simulator includes an option to export the reservoir model (‘Export to Sim’)
to a data file that can be run by an industry standard simulator. The DOE BOAST III executable
(max grid size: 30 x 28 x 7), manual and example data file are included in the ‘PE
Essentials/Public/BOAST III’ directory.
During a run of the PE² Essentials Basic reservoir Simulator, a file with the extension ‘.dvxg’ is
stored in the “PE Essentials\Simulator Run Files” directory. This file contains the dynamic grid
260 Basic Reservoir Simulation Tool
information for the forecast. The file name includes the date and the entered ‘Run Info’ as well
as a randomly generated number (eg - 201766_Gas, Horizontal,1-frac_757.dvxg). The number
included on the initial run will be incremented by 1 for re-runs. The dynamic grid data can be
imported and examined using ‘Basic Plot’. Note that previously stored ‘.dvgx’ files can be
imported into the simulator and examined without running the model by loading the model and
then clicking ‘Basic Plot’.
Note that checking the ‘Save Simulation Run Data’ will save and update a run file as the run
progresses in the “PE Essentials\Simulator Run Files” directory with a similar file naming
convention as described above for the dvxg file but with a ‘.csv’ extension. To compare forecasts
from different runs, the ‘Save Simulation Run Data’ option can be used and the forecast data
imported into PE Essentials Chart and compared with other runs.
There are a number of model input files included in the PE Essentials Tools Database included in
the “Example Input Files\PEE Tools Database” directory as described below.
- Gas,NonComLayers:
Layered dry gas reservoir section with non-communicating layers.
Layer communication set to zero by placing ‘shale’ streaks between each layer.
- Gas,Horizontal:
Dry gas reservoir section with a horizontal well in the x-direction.
The horizontal well is placed in i=2 to 14, j=10 and layer 2.
- Gas,Horizontal,1-frac:
Dry gas reservoir section with a hydraulically fractured horizontal well at 2-14,10,2.
Model of one frac assuming a 500ft frac stage and a 1000ft well spacing.
Frac has an xf of 200ft and is placed at i=8 by modifying the perms in i=8 and j=5 to 15.
The equivalent fracture permeability of 5 md for this grid system was calculated from:
kf = FCD Perm xf / dx(8), where FCD=5, Perm=0.1, xf=200 and dx(8)=2.
Maximum time step size for this model was limited to 0.5 days.
- Oil,Dipping,GaussianPermPoro:
Model of a dipping oil reservoir section with a gas cap and a large active aquifer.
Aquifer is modeled with large outer blocks.
Aquifer strength is modeled by increasing permeability in the aquifer blocks.
- Oil,Dipping,GaussianPermPoro,Metric:
Model of a dipping oil reservoir section with a gas cap and a large active aquifer.
Aquifer is modeled with large outer blocks.
Aquifer strength is modeled by increasing permeability in the aquifer blocks.
Note that the ‘Export to Sim’ button will generate a data input file that can be used with an
industry standard simulator.
SIM.1 Grid Construction

The ‘Grid’ button will open the grid construction sheet (Figure SIM-2).
Figure SIM-2: Basic Reservoir Simulator – Grid Construction
The grid entry form will increase in size as the grid dimensions are entered (Figure SIM-3) until
the final grid is entered.
Figure SIM-3: Grid Construction – Entering Grid Block Sizes
After entering the average layer properties, it is possible to apply variability to the model’s
porosity and permeability through a Gaussian distribution routine (Figure SIM-4).
Figure SIM-4: Grid Construction – Variable Properties

By checking the appropriate box, a Gaussian distribution routine will modify the entered value
based on the input +/-% range (Figure SIM-5).
Figure SIM-5: Grid Construction – Property Variation
To generate the same ‘random’ distribution, a seed value should be entered.
If a dipping reservoir (Figure SIM-6) is being modeled, then then the ‘Dipping Reservoir’ box
should be checked and a dipping angle, α, is entered (insert on Figure SIM-6).
Figure SIM-6: Grid Construction – Dipping Reservoir
Only vertical faults can be modeled, and they are assumed to be vertical planes extending from
the top to the bottom of the reservoir. Faults are modeled as zero transmissibility at the fault
location. Up to 10 faults can be specified in the reservoir model (Figure SIM-7).
Figure SIM-7: Grid Construction – Faults
Fault location is determined by the grid block number in the x- and y-direction, and the relative
position of the fault in the block. As an example, for a sealing fault in grid 2,4 as indicated in
Figure SIM-7, the fault is represented by two segments located at (2,4). The right segment is
labeled ‘X’ because the fault is on the east side of the grid block center and will have a constant
x-grid value and variable y-grid values. The northern segment is labeled ‘Y’ since the fault is
located on the north side of the grid block and will have a constant y-grid value and variable x-
grid values.
To model a slanted fault, set it up as two vertical faults, one x-fault at I=2 and a y-fault at J=4, as
shown in Figure SIM-8.
Figure SIM-8: Grid Construction – Modeling a Slanted Fault

Shale barriers can be entered and are modeled as zero z transmissibility (Figure SIM-9).
Figure SIM-9: Grid Construction – Shale Barriers
Shale barriers can be used to restrict communication between layers. The shale layers have zero
thickness and are modeled as zero transmissibility across the layer. Shale layers can be entered
over any portion of the layer.
There is an option to enter up to 10 modifications to the x and y permeability values (Figure SIM-
10). X and Y permeability values can be entered simultaneously by specifying ‘X/Y’.
Figure SIM-10: Grid Construction – Permeability Modifications

As an example, this option could be used to model an aquifer in a reservoir. Figure SIM-11 shows
the modelling of a strong aquifer by setting x-perm and y-perm to 1000md.
`
Figure SIM-11: Grid Construction – Modelling an Active Aquifer Using Permeability
Shale and tight streaks within layers can be modeled by inactivating grid cells in the model.
Inactivating a complete layer can be implemented when commingled separate intervals, with no
inter-layer crossflow, are being produced. Inactivating a cell sets the porosity and permeabilities
for that cell to 1x10-10.
There is an option to enter up to 10 regions containing inactive cells (Figure SIM-12).
Figure SIM-12: Grid Construction – Specifying Inactive Regions/Cells
After the grid has been fully defined, the grid values can be saved to a ‘csv’ file through ‘Export
Grid Properties’. This grid file can be imported into an external plotting program, but it is not
formatted for input into PE Essentials Chart.
SIM.2 PVT Properties

The PVT input properties are dependent on whether the ‘Dry Gas Model’ (Figure SIM-13) or the
‘G/O/W Model’ (Figure SIM-14) is being used.
Figure SIM-13: PVT Properties – Dry Gas Model
Figure SIM-14: PVT Properties – Gas/Oil/Water Model

The Basic reservoir Simulator uses the PVT BASE correlations as described for the PVT tool. PVT
properties can also be imported from the PE Tools database.
The oil model includes an option to calibrate the oil properties to lab-derived or EOS-derived
properties. This is implemented through the ‘Match PVT Parameters’ button (Figure SIM-14 and
SIM-15).
Figure SIM-15: PVT Properties – Match Oil Properties
The matching parameters can be obtained from a laboratory PVT report or can be generated by
the PE² Essentials Basic EOS PVT program. The pressure and temperature point entered for the
match is normally the reservoir temperature and the bubble point pressure of the oil.
The match routine will apply a constant correction to all the values calculated by the internal
routines. The adjusted oil parameters are used in the simulator. Refer to the PVT tool for more
information on matching oil PVT properties.
SIM.3 Relative Permeability

The relative permeability inputs are dependent on whether the ‘Dry Gas Model’ (Figure SIM-16)
or the ‘G/O/W Model’ (Figure SIM-17) is being used.
The letter k represents the absolute permeability of the reservoir (in md) and ko, kg and kw
represent the effective permeability to oil, gas and water. The fluid saturations, So, Sg and Sw must
also be specified to fully define the conditions for the value of effective permeability. Studies
have shown that a reservoir’s effective permeability in terms of the reservoir fluid is a function
of the saturation of that fluid and the wetting characteristics of the reservoir.
Figure SIM-16: Relative Permeability – Gas Model
Figure SIM-17: Relative Permeability – Gas/Oil/Water Model

Since there are many possible values for saturation, effective permeability is normally reported
as relative permeability: kro, krg and krw.
kro = ko/k
krg = kg/k
krw = kw/k
Effective permeability ranges from zero to k so relative permeabilities range from zero to one.
0 < kro , krg , krw < 1.0
When all three phases are present in the reservoir, the sum of the relative permeabilities is
variable and less than or equal to one: kro + krg + krw < 1.0.
Oil-Water System (Kro/Krw and Pcow)
For an oil-water sytem the relative permeability analytical equations use exponents as follows.
kro = krow ((1 - Sw - Sorw)/(1 - Swi - Sorw))No

krw = krwe ((Sw - Swi)/(1 - Swi - Sorw))Nw
Pcow = Pc(Swi) ((1 - Sw - Sorw)/(1 - Swi - Sorw))Npo
Where No is the exponent for oil, Nw is the exponent for water and Npo is the oil capilary
pressure exponent, krow is the kro at Swi, Swi is the initial water saturation, Sw is the desired water
saturation, Sorw is the residual oil saturation, krwe is the krw at Sorw, Pcow is the oil-water capilary
pressure and Pc(Swi) is the capilary pressure at Swi.
Gas-Oil System (Krog/Krg and Pcog)
For gas-oil sytem the relative permeability analytical equations are as follows.
krg = krg(Sorg) ((Sg - Sgc)/(1 - Swi - Sorg - Sgc))Ng

krog = kroge ((1 - Sg - Swi - Sorg)/(1 – Swi - Sorg))Nog
Pcog = Pc(Sgc) ((Sg - Sgc)/(1 - Swi - Sorg - Sgc))Npg
Where Ng is the exponent for gas, Nog is the exponent for oil in gas and Npg is the gas capilary
pressure exponent, krg(Sorg) is the krg at Sorg, Sorg is is the residual oil saturation in gas, Sgc is the
critical gas saturation, kroge is the kro at Sorg, Swi is the initial water saturation, Sg is the desired gas
saturation, Pcog is the gas-oil capilary pressure and Pcog(Sgc) is the capilary pressure at Sgc.
Normalized water saturation is calculated as Swn = (Sw – Swi )/(1 – Swi - Sorw)
The following can be used (with caution) as a rule of thumb.
Oil Wet:
No >= 6
Nw < 3
Krow >= 0.5
Intermediate Wet:
6 < No >= 3
5 < Nw >= 3
Water Wet:
No < 3
Nw >= 5
Krow < 0.5
There is an option to default the parameters.
SIM.4 Reservoir Parameters

The reservoir input parameters are dependent on whether the ‘Dry Gas Model’ (Figure SIM-18)
or the ‘G/O/W Model’ (Figure SIM-19) is being used.
Figure SIM-18: Reservoir Parameters – Gas Model

Figure SIM-19: Reservoir Parameters – Gas/Oil/Water Model
The reservoir inputs are similar to the inputs for the Unconventional Forecast Model. Refer to
the Unconventional Forecast tool for more information.
SIM.5 Wellbore Parameters

The wellbore input parameters are dependent on whether a vertical well (Figure SIM-20) or
horizontal well (Figure SIM-21) is being modeled.
Figure SIM-20: Wellbore Parameters – Vertical Well

Figure SIM-21: Wellbore Parameters – Horizontal Well
The main difference between the well inputs is in how the location of the well is entered into the
model. For a vertical well the top and bottom completion layer are specified along with the I,J
grid location for the wellhead. For a horizontal well, the starting location of the lateral (I, J) is
specified, the layer of the completion (K) and the ending block (I) for the lateral are specified.
The available tubing correlations are dependent on the fluid model being used; only the Modified
Hagedorn-Brown correlation is available for an oil reservoir.
SIM.6 Well Schedule

The ‘Schedule’ button is used to input the scheduling parameters for the well. It can also be used
to enter historical data if the simulator is being used to generate a history match prior to
forecasting (Refer to Section SIM.9 for an example of importing a historical schedule). Figure SIM-
22 shows the schedule input screens. It is possible to enter a schedule manually (maximum of 10
events) or import a schedule from either a CSV file or an Excel spreadsheet.
Importing a schedule file will overwrite any previous schedule data as well as any manually
entered schedule events. Up to ten schedule events can be entered manually (there is no limit
on the number of imported schedule events).
Note: To model the shut in of a well, the minimum rate on the main screen should be set to -1.
This ensures the simulation is not stopped because of a minimum rate caused by the shut in.
Figure SIM-22: Importing a Well Schedule
The first time is always ‘0’ since it represents the start of the simulation. All other items can be
entered for any time but subsequent events must be increasing in time. If more than 10 items
are required, they must be entered using the ‘Import Schedule’ option. It should be noted that
for a gas reservoir, ‘Rate’ is the gas rate in mscf/d or 10 3sm3/d. For an oil reservoir, rate is bopd
or m3/d. BHP is entered in psi or kPa. In all cases time is entered in years. Refer to the info button
for more information on file formats.
Completion layers in a vertical well, or 'I' blocks in a horizontal well, can be opened and closed
during the simulation using the ‘Open/Close Block’ input box. This allows recompletion of
sections of the well. To open or close a grid block, it must be included in the original completion
description of the well (Section SIM.5).
To shut in a layer, or I-block, enter the layer or I-block number as a negative value. For instance,
'-3' will shut-in layer 3 in a vertical well. To reopen the layer, enter a positive value ('3' will re-
open layer 3). Once shut-in, a layer or I-block, will remain shut in until it is specifically re-opened
in the schedule. Multiple schedule items can occur at the same time by entering the same time
for the activity.
When using a historical schedule, the BHP values in the schedule file should be set to 15 psi or
101 kPa. This value is specified since the rate is used as the known parameter and BHP is
calculated by the simulator. The BHP value in a schedule file is the minimum flowing pressure
and will be used to modify the rate.
SIM.7 Simulator Options

The ‘Simulator Options’ (Figure SIM-23) allow the execution parameters of the simulator to be
modified.
Figure SIM-23: Simulator Options
If a number of runs are being made, it may be worthwhile to generate a couple short runs to
determine the optimum parameters to be used for that specific model. The Simulator Options
are saved with the model, so they do not have to be re-entered.
‘Run Control Parameters’ (Figure SIM-24) control how the run progresses and is terminated.
Figure SIM-24: Simulator Options – Run Control Parameters
The following describes each of the parameters.

Nmax: Maximum number of time steps allowed before run is terminated
Factor1: Factor for increasing time step size under automatic time step control
Factor1 = 1.0 for a fixed time step size of ∆T
Factor2: Factor for decreasing time step size under automatic time step control
Factor2 = 1.0 for a fixed time step size of ∆T
WORmax: Limiting maximum field water-oil ratio, in bbl/bbl (m3/m3)

simulation will be terminated if total producing WOR exceeds WORmax
GORmax: Limiting maximum field gas-oil ratio, in scf/bbl (scm/m3)
simulation will be terminated if total producing GOR exceeds GORmax
Pb Repressure: Bubble Point repressurisation algorithm
0 = no repressurisation; 1 = repressurisation will be performed
avgPmin: Limiting minimum average field pressure, psia/kPaa
simulation terminated if average reservoir pressure falls below avgPmin
avgPmax: Limiting maximum average field pressure, psia/kPaa
simulation terminated if average reservoir pressure exceeds avgPmax
∆T: Initial time-step size (days)
∆Tmin: Minimum time-step size (days)
∆Tmax: Maximum time-step size (days)
The Run Control Parameter that can significantly impact the run time is ‘∆Tmax’.
‘Solution Control Parameters’ (Figure SIM-25) controls the solution tolerance for the iteration.
Iterations will be performed until all tolerances have been met.
Figure SIM-25: Simulator Options – Solution Control Parameters
There are no specific optimum values for these parameters; the user should experiment with
several different values, especially LSORx vs LSORz, to see which ones will yield the optimal result
for a particular problem.
Max Iterations: Maximum number of iterations for LSOR convergence

∆Ptolerance: Maximum acceptable pressure change for LSOR iteration convergence.
OMEGA: LSOR solution acceleration parameter.
The initial value for OMEGA must be in the range 1.0 < OMEGA < 2.0.
∆OMEGA: Parameter for determining when to change OMEGA.
∆OMEGA = 0.0 the initial value is used for the entire simulation
∆Smax: Maximum saturation change (fraction) permitted in any grid-block.

∆Pmax: Maximum pressure change permitted in any grid-block.
The Solution Control Parameters that impact the run time are the ‘∆Ptolerance’ and ‘OMEGA’.
‘Output Grid Parameters’ (Figure SIM-26) control the timing for the generation of an output file
of the grid parameters.
Figure SIM-26: Simulator Options – Output Grid Parameters
All the parameters default to zero and have to be set if run time grid files are required. The
parameters are described below.
Map Output: Enable saving of solution grid data based on the output interval
Enter '1' to enable, '0' to disable output maps
Pressure Map: Output map of map of grid block pressures
File name: PE_Essentials_ResSim_PrMAP.csv
Sw Map: Output map of map of grid block water saturations
File name: PE_Essentials_ResSim_SwMAP.csv
So Map: Output map of map of grid block oil saturations
File name: PE_Essentials_ResSim_SoMAP.csv
Sg Map: Output map of map of grid block gas saturations
File name: PE_Essentials_ResSim_SgMAP.csv
Pb Map: Output map of map of grid block saturation pressures
File name: PE_Essentials_ResSim_PbMAP.csv
Output maps will be generated at the specified interval. For an output interval of 365 days, maps
will be output every 365 days (+/-). For an output interval of 0 days, maps will be output after
initialization and at the end of the simulation.
One of the issues with an IMPES solution is when the flow through a cell during a time step is
more than 100% of the mobile phase existing in that cell. This occurrence tends to destabilize the
solution of the equations since the material balance of the system is impacted. In a simulation
model, this normally occurs in the wellbore grid blocks. To remedy this situation, a technique
termed ‘Stabilized IMPES’ is implemented in the PE Essentials Basic Reservoir Simulator.
Stabilized IMPES ensures that the throughput in any cell at any time step is limited to less than
100%. If throughput is >100%, the time step size is reduced and the time step is recalculated. This
continues until the throughput in the cell is less than the defined throughput value. This value
has been set at 75% for this simulator.
The result of implementing a Stabilized IMPES routine is that the initial time step, and subsequent
rate changes, can cause incremental time step sizes to be dramatically reduced. The simulator
time step size will recover as flow stabilizes. To account for Stabilized IMPES, the minimum step
size, ∆Tmin (Figure SIM-24) should be set to a very small value.
Note that all the run files and grid files generated by this tool are stored in the ‘Simulator Run
Files’ directory. As a result, an old run can be imported into the plotting tool by loading model
but not running it. Then open the plot module and load the run file from the ‘Simulator Run Files’
directory.
SIM.8 Quick Plot

The ‘Quick Plot’ button will generate plots of the production forecast, grid parameters and
forecast grid parameters. The following examples are from a run of the
Oil,Dipping,GaussianPermPoro model.
SIM.8.1 Production Forecast Plots

Following a run, the production forecast plotting options are shown in Figure SIM-27.
Figure SIM-27: Simulator Options – Production Plot

The production parameters available for plotting are dependent on whether a gas model or an
oil model is being used. The data can also be plotted on a log scale and in terms of cumulative
volume (Figure SIM-28).
Figure SIM-28: Simulator Options – Production Plot Options
SIM.8.2 Static Grid Plots
The grid parameters can be plotted as shown in Figure SIM-29. When plotting the grid, the
location of the wellbore is presented in red and the faults are presented in blue.
Figure SIM-29: Simulator Options – Static Grid Plot

Figure SIM.30 is an x-y plot of layer 2 showing the location of the horizontal well in the
Gas,Horizontal model.
Figure SIM-30: Simulator Options – Static Grid Plot: Horizontal Well
Figure SIM-31 shows the depth to the top of the grid blocks in the x-z plane. This model is a
‘Dipping Reservoir’ model with gas/oil and oil/water contacts.
Figure SIM-31: Simulator Options – Static Grid Plot: Layer Depths

Figure SIM-32 shows the Gaussian modified kx permeability along the J=6 plane. Note that this
model included a strong aquifer in the large outer grid block (k=1000md).
Figure SIM-32: Simulator Options – Static Grid Plot: X-Permeability
Figure SIM-33 presents the Gaussian modified porosity along the I=8 plane.
Figure SIM-33: Simulator Options – Static Grid Plot: Porosity
All the available plotting options for the static grid properties allow the model to be QC’d.
SIM.8.2 Dynamic Grid Plots
The dynamic grid parameters at a specific forecast time can be plotted as shown in Figure SIM-
34. Either an X cross section or a Y cross section can be plotted.
Figure SIM-34: Simulator Options – Dynamic Grid Plot: Pressure
Figures SIM-35 and SIM-36 show water encroachment into the lower well completion.
Figure SIM-35: Simulator Options – Oil Saturation, Lower Well Completion Layer
Figure SIM-36: Simulator Options – Water Saturation, Lower Well Completion Layer
‘Save Graph’ will save the active plot to a ‘png’ file. The ‘Update’ button can be used to
dynamically update the production plot as the run progresses.
SIM.9 Example – History Matching a Horizontal Oil Well Test

This example was generated as a demonstration of importing and using a schedule file to match
the historical data.
A horizontal test well was drilled into a tight oil reservoir to test the productivity of the reservoir
and obtain parameters for future drilling and completion programs. The well was completed
without cement using a pre-perforated liner with external casing packers to isolate and test
specific intervals.
The well was produced for a total of 203 days and the production data is included in the
‘Horizontal Oil Well Test.xlsx’ file located in the “PE Essentials\ Book Examples\Example Basic
Reservoir Simulator” directory. The Excel file includes the schedule for input into the simulator.
A CSV file containing the production history was generated from the Excel file for plotting
purposes (‘Horizontal Oil Well Test History Data.csv’).
The Horizontal Oil Well Test model located in the ‘PEE Tools Database Book Examples’ database
file located in the “Book Examples\PEE Tools Database Examples“ directory includes the model
construction parameters. The best guess estimate of permeability for this tight oil reservoir was
0.1 md, as shown on figure SIM-37.
Figure SIM-37: Grid Properties – Horizontal Test Well
No other modifications were made to the grid. The remaining model parameters can be viewed
by clicking the appropriate buttons. The wellbore information is presented in Figure SIM-38.
Figure SIM-38: Wellbore Properties – Horizontal Test Well

Figure SIM-38 shows that the horizontal well was completed in layer 2 and is located from grid
block (3, 8) to grid block (12, 8). In fact, the well was actually completed in two intervals,
equivalent to grid blocks (3, 8) to (4, 8) and grid blocks (9, 8) to (12, 8). The schedule is used to
shut-in the intervening grid blocks, (5, 8) to (8, 8).
Figure SIM-39 shows the import of the schedule file from the Excel spreadsheet. In total, there
were 179 schedule events imported for this well. If a forecast is also required, the forecast
parameters can be manually entered by changing the ‘Number of Well Schedule Events’ to a
number between 2 and 10 on the ‘Well Schedule’ page.
Figure SIM-39: Wellbore Properties – Horizontal Test Well
Figure SIM-40 shows the data in the Excel file that was imported in Figure SIM-39.
Figure SIM-40: Well Schedule – Excel File

A forecast was generated for the model and the results were exported to a CSV file by clicking
‘Save Results’ and then compared to the BHP history data using PE² Essentials Chart. A second
run was made using a permeability of 0.15 md which appears to yield a closer match to the
historical BHP (Figure SIM-41).
Figure SIM-41: Pressure History Match – Horizontal Test Well
For future analysis, a permeability of 0.15md would then be used.

286 Streamtube Waterflood Simulation Tool
Streamtube Waterflood Simulation Tool

The PE² Essentials ‘StreamTube WaterFlood’ tool is a simplified streamtube simulation model
based on the Leighton and Higgins model (Leighton, A., J. and Higgins, R., V., Improved Method
to Predict Multiphase Waterflood Performance for Constant Rates or Pressures, United States
Department of the Interior, Bureau of Mines Report of Investigation #8055, 1975). An excellent
reference on streamtube simulation is the book by Datt-Gupta and King – Streamline Simulation:
Theory and Practice published by SPE in 2007.
A streamtube is a region bounded by two streamlines (Figure SWF-1: Reference – Figure 6 from
http://petrowiki.org/Scaleup_to_full_field_miscible_flood_behavior).
Figure SWF-1: Streamlines and Streamtubes
In Figure SWF-1, the dashed lines outline the area that is being affected by the injector-producer
pair. The solid lines are streamlines and represent the tangent of the vx-vy velocity field at a given
point at a snapshot in time in the reservoir. They could also (very loosely) be considered to
represent the ‘time of flight’ for a particle along the path of the streamline – so the shortest time
(highest velocity) to go from the injector to the producer follows the straightest streamline. The
streamline is a map of the instantaneous velocity field in the reservoir and does not necessarily
represent the physical movement of the particles.
The filled area on Figure SWF-1 is the area between two streamlines and is termed a streamtube.
All fluid movement remains within the streamtube since the streamlines are tangential to the
fluid velocity. The wider the streamtube the slower the flow and the narrower the streamtube,
the faster the flow (flow is cross-sectional area times velocity). As indicated on Figure SWF-1, the
fastest flow will occur in a direct line from the injector to the producer – narrowest streamtube.
Streamline simulation provides an alternative to the grid-based techniques used in numerical

reservoir simulation. Streamlines represent a snapshot of the instantaneous flow field and
thereby produce data such as drainage regions associated with injector-producer well pairs and
flow rate allocation between injector-producer pairs that can be quickly determined and are not
easily determined by other numerical reservoir simulation techniques.
The PE² Essentials ‘Stream Tube WaterFlood’ tool is used to simulate a pattern waterflood at a
constant water injection rate or at a constant pressure (Figure SWF-2).
Figure SWF-2: PE² Essentials - StreamTube WaterFlood Tool
Four streamtube models are included in this model (Figure SWF-3):

▪ 5 Spot: 1-to-1 ratio of producer to injector
▪ Line Drive: 1-to-1 ratio of producer to injector
▪ Staggered Line Drive (4 Spot): 1-to-3 ratio of producer to injector
Figure SWF-3: StreamTube WaterFlood Model - Patterns

Note that this model is similar to the Miscible/Immiscible CO2 WAG WaterFlood Simulator in that
they both simulate a waterflood using a streamtube simulator, but this model is limited to
waterflood simulations only.
SWF.1 Reservoir Model

Entering the reservoir model is comprised of inputting ‘Reservoir Parameters’ and ‘Corey
Function’ (Figure SWF-4) to generate relative permeabilities.
Figure SWF-4: StreamTube WaterFlood Model - Reservoir Parameters
Most of the reservoir parameters and the Corey function parameters are straightforward. The
‘Current Gas Saturation’ and the ‘Current Water Saturation’ enable initialization of the reservoir
at a depleted condition.
The ‘Layer Parameters’ allow some degree of heterogeneity to be modeled.
SWF.2 Pattern Waterflood Model

Choosing the waterflood model to use for the simulation is comprised of three options (Figure
SWF-5).
Figure SWF-5: StreamTube WaterFlood Model - Waterflood Parameters

One of the four available patterns is chosen and whether this is to be a constant pressure
waterflood or a constant rate waterflood. The distance between the injector and producer is
entered and, depending on the option chosen, constant pressure drop between the injector and
the producer is entered or the constant injection rate. Achieving the maximum water cut value
will stop the simulation.
After running the forecast, the results are presented in a table and as plots (Figure SWF-6).
Figure SWF-6: StreamTube WaterFlood Model - Results
‘Export Results to CSV’ will save the forecast to a ‘csv’ file for plotting and comparison purposes
(PE² Essentials Chart). The waterflood forecast can also be saved to the PE Tools database.
SWF.3 Example - Choosing a Waterflood Pattern

As an example, consider Figure SWF-7 which shows producing well locations in the Weyman –
Glasscock County of Texas.
The concept to be investigated is that a pattern waterflood is to be implemented to increase the

ultimate oil recovery from the area. Some of the producers will be converted to injectors for the
re-development of the area.
A common practice is to keep the best wells as producers and convert the poor wells to water
injectors, which normally defines the pseudo waterflood pattern that will be implemented. The
reality is that the best producers make the best injectors and that more thought should go into
selecting a pattern to implement.
Figure SWF-7: Well Locations in Weyman - Glasscock County of Texas
The StreamTube WaterFlood Model can be used to perform preliminary evaluation of the
recovery resulting from utilizing different waterflood patterns.
The model presented in Figure SWF-8 was used to evaluate the recovery resulting from different
waterflood patterns. In order to make an equivalent comparison, the ‘Distance From Injector to
Producer’ was modified for each pattern in order to set up all patterns to flood a 14 acre area.
The models are saved in the ‘PEE Tools Database Book Examples.PEEdb’ located in the “PE
Essentials 2019\Book Examples\PEE Tools Database Examples” directory.
The results for each forecast were exported and compared with PE² Essentials Chart (Figures
SWF-9).
The recovery factors for each pattern are as follows:

• 5 spot: 50.9%
• 7 spot: 53.3%
• Line Drive: 54.4%
• Staggered LD: 53.0%
Figure SWF-8: WaterFlood Model for Comparison
Figure SWF-9: WaterFlood Pattern Comparisons – 14 Acre Patterns
From this example, with all other things being equal, a line drive pattern will yield the highest oil
recovery. At this point more analyses, including economics, should be run.
292 Misc/Immis CO2 WAG Simulation Tool
Misc/Immis CO2 WAG Simulation Tool

The PE² Essentials ‘Misc/Immisc CO2 WAG WF’ tool is a streamtube simulation model based on
the DOE “CO2 Miscible Flood Predictive Model” (1986). This public domain simulation model was
developed for the DOE to estimate economically recoverable oil from reservoirs amenable to CO2
miscible flood processes. The DOE program can be used for both secondary (mobile oil) and
tertiary (residual oil) floods, and for either continuous CO2 injection or water-alternating-gas
(WAG) processes. Texaco modified the program in 1994 and re-named it CO2-Prophet. The
CO2PM executable, manual and example data files are included in the ‘PE
Essentials/Public/CO2Miscible’ directory.
Rather than re-build the program in the PE² Essentials development environment, this tool will
build a data file for the CO2PM program and then runs the executable. The results are then
imported into the Misc/Immisc CO2 WAG WF tool for evaluation (Figure CO2-1).
Figure CO2-1: PE² Essentials – Misc/Immisc CO2, WAG, WaterFlood StreamTube Tool
For more information on miscibility and miscibility pressures, refer to EOR / Heavy Oil Tool.
Six streamtube models are included in this model (Figure CO2-2):

▪ Inverted 9 Spot: 1-to-3 ratio of producer to injector
▪ Line Drive: 1-to-1 ratio of producer to injector
Figure CO2-2: Misc/Immisc CO2 WAG WF Tool - Patterns
This tool can simulate straight waterfloods as well as miscible/immiscble gas/CO₂ and Water-
Alternating-Gas (WAG) pattern floods. Up to four periods of injection can be simulated
sequentially in the PE² Essentials implementation of the DOE model. For example, a sequence of
a waterflood, followed by CO₂ injection, followed by a CO₂ WAG and followed by a final
waterflood could be simulated.
For a straight waterflood, information for only one period is entered. Alternatively, up to four
waterflood periods at different injection rates could also be modeled.
Since this is an in-depth streamtube simulation model, data over and above the input required
for the PE² Essentials StreamTube WaterFlood tool has to be entered.
‘Export Results to CSV’ will save the forecast to a ‘csv’ file for plotting and comparison purposes
(PE² Essentials Chart). The waterflood forecast can also be saved to the PE Tools database.
CO2.1 Reservoir Model

Entering the reservoir model is comprised of inputting ‘Reservoir Parameters’ (Figure CO2-3). In
addition to the normal reservoir parameters, additional parameters are required for the model.
Figure CO2-3: Misc/Immisc CO2 WAG WF Tool - Reservoir Parameters
For more information on the input parameters, refer to the ‘CO2-Manual.pdf‘ and the
‘CO2ProfitManual.pdf’ located in the ‘PE Essentials/Public/CO2Miscible’ directory.
It should be noted that the program does not check the reasonableness of the input data; it is up
to the user to enter realistic data. The program will inform the user if the results generated by
using the input data are unrealistic (normally causes a divide-by-zero error in the simulator).
The ‘Max Water Inj Rate’ and ‘Max Gas Inj Rate’ are calculated after the reservoir parameters
have been input and are presented on the input sheet (Figure CO2-3).
CO2.1.1 Minimum Miscibility Pressure

The ‘Minimum Miscibility Pressure’ is used to determine whether the flood will be completely
miscible (reservoir pressure is greater than the MMP), partially miscible (reservoir pressure is
greater than 0.75 MMP but less than MMP), or totally immiscible (reservoir pressure is less than
0.75 MMP). MMP must be set below the reservoir pressure for the flood to be completely
miscible. MMP can be estimated using the PE² Essentials EOR/Heavy Oil Tool.
CO2.1.2 Degree of Mixing

The mixing parameter approach (‘Degree of Mixing’), as proposed by Todd and Longstaff, is used
for simulation of a miscible CO₂ process. Mixing parameter models simulate the mixing and
viscous fingering which occurs in miscible displacements by adjusting solvent and oil viscosities.
These adjustments alter the fractional flow of solvent and oil.
The mixing parameter, ω, is used to adjust the viscosities of the solvent and the oil. Omega
determines the effective viscosity of the solvent and oil. If the mixing parameter is set to 0.0,
then there is no mixing and the solvent and oil viscosities are equal to their individual immiscible
values. If the mixing parameter is set to 1.0, then there is complete mixing, and the oil and solvent
viscosities are made equal.
CO2.1.3 Miscible Phase Relative Permeability Option

In the model, there are three phases which can flow: water, oil, and gas/solvent (the terms gas
and solvent are used interchangeably). The only gas phase that is permitted in the model is the
solvent phase. A separate hydrocarbon gas does not exist in the model. The relative permeability
equations which are used for simulating both miscible and immiscible flow are constructed from
two-phase flow equations.
For immiscible flow (reservoir pressure is less than 0.75 MMP), the water relative permeability
(krw) is a function only of Sw. The gas (or solvent) relative permeability (krg, krs) in immiscible flow
is a function of the gas (i.e., solvent) saturation only. The three-phase oil relative permeability
(kro3) is determined using the modified Stone method (refer to PE2 Essentials Gas/Oil/Wat PVT /
Rel Perm Tool for information on relative permeability).
kro3 = (a – krg – krw) / kro

a = (kro/krow + krw)(krog/krow + krg)
For miscible flow (reservoir pressure greater than MMP) there are actually only two phases,
water and a miscible phase composed of solvent and oil. The water relative permeability is the
same as it is in immiscible flow and remains a function of only the water saturation. However,
the miscible phase relative permeability (krm) must be computed since it is not measured.
There is no definitive way to compute or handle the miscible phase relative permeability. There
are three options for calculation of krm. The option to calculate the miscible phase relative
permeability, krm, can be selected as:
- Option=0: Use linear variation between solvent/gas and oil relative permeability
- Option=1: Use average of oil and solvent/gas relative permeability
- Option=2: Use oil relative permeability
The third option (Option=2), which sets krm equal to kro, the oil phase relative permeability, is the
standard formulation in most mixing parameter models. However, the first option which makes
krm a saturation weighted average is physically more realistic. In addition, the saturation
weighted formulation produces results closest to those of a compositional simulator when krg
parameters are used for the solvent relative permeability, krs.
CO2.1.4 Dykstra-Parsons Coefficient

The Dykstra-Parsons coefficient is a common measure of reservoir heterogeneity. There is a large
variation in reservoir permeability when there is a large Dykstra-Parsons coefficient. For
reservoirs with large variations in reservoir permeability, the recovery will be greatest in the
highest permeability layer.
The Dykstra-Parsons coefficient normally varies from 0.1 to 0.9 and can have a large impact on
recovery. The Dykstra-Parsons coefficient is used to calculate the permeability variation between
the layers in the model. This variation, in turn, determines the relative injectivity of fluids in each
layer. This calculation of layer permeabilities is done internally in the program and the results are
presented on the main screen.
CO2.2 Relative Permeability

The relative permeability parameters are entered in the ‘Rel Perm Data’ Section (Figure CO2-4).
Figure CO2-4: Misc/Immisc CO2 WAG WF Tool - Relative Permeability

CO2.3 Pattern Waterflood Model

Choosing the waterflood pattern to use for the simulation is comprised of two options (Figure
CO2-5).
Figure CO2-5: Misc/Immisc CO2 WAG WF Tool - Waterflood Parameters
The waterflood pattern is chosen and the area of the pattern is entered. This will set up the
distance from the injector to the producer.
The ‘WAG Ratio’ is the amount of time water is injected, relative to the time gas is injected. For
water injection, the number is 1.0; for CO₂ injection, the number is 0.0; for WAG injection, the
number is greater than 0.0 but less than 1.0. For example, if the number is 0.6, then 60% of the
time water is injected and 40% of the time CO₂ is injected at the specified rates.
The hydrocarbon pore volume parameter (HCPV) determines the cumulative volume injected for
that period. Up to four periods are entered and the model is executed.
CO2.4 Example: Waterflood Model Comparison

A comparison between the two streamline waterflood models was made using similar data in
each model. The models were run out to a 98% water cut. For the Misc/Immisc CO 2 WAG WF
Model, this was done by injecting 1.25 HCPV. Figure CO2-6 shows the Misc/Immisc CO2 WAG WF
Model results and Figure CO2-7 shows the StreamTube WaterFlood Model results. The models
are saved in the ‘PEE Tools Database Book Examples.PEEdb’ located in the “PE Essentials
2019\Book Examples\PEE Tools Database Examples” directory.
The Misc/Immisc CO2 WAG WF Model, in general, shows a similar waterflood development as
the StreamTube WaterFlood Model. In terms of recovery, they are similar with the Misc/Immisc
CO2 WAG WF Model indicating an ultimate recovery of 55.0% compared to 54.6% for the
StreamTube WaterFlood Model.
Figure CO2-6: Misc/Immisc CO2 WAG WF Tool Results - Comparison Run
Figure CO2-7 StreamTube WaterFlood Tool Results - Comparison Run

Gas Material Balance Analysis Tool

The general material balance (MB) equation can be applied to all reservoir types and was first
developed by Schilthuis in 1936. It is a tank model equation that accounts for production from
an initial point to any point in time as follows:
(oil + gas + water) produced = oil expansion + gas expansion +

oil zone and gas zone pore volume expansion +
connate water expansion + water influx +
water injected + gas injected
For a gas reservoir, the material balance equation simplifies to Equation GMB-1
Gp Bg = GIIP (Bg – Bgi) (GMB-1)
Expanding the term for Bg and simplifying, Equation 5-26 becomes Equation GMB-2.
Gp (Pi/Zi) = GIIP (Pi/Zi – P/Z) (GMB-2)
Where: Gp is gas production, Pi is the initial pressure, Zi is the Z-factor at Pi, GIIP is the initial gas
in place, P is the current pressure and Z is the Z-factor at P.
The material balance equation is zero-dimensional, meaning that it is based on a tank model and
does not take into account the geometry of the reservoir, the reservoir drainage area, the
position of the well or orientation of the well.
A material balance tank is equivalent to a single grid block in a numerical reservoir simulation
model. In a numerical simulator, the conditions in a large number of communicating material
balance tanks (grid blocks) are solved simultaneously.
Most reservoirs are made up of compartments that are separated by faults that may be closed
or open (partially or fully). If the faults are closed, then there is no communication between the
tanks and they can be modelled as isolated material balance tanks. If the faults are totally open,
then the whole reservoir can be modelled as one material balance tank.
However, if the faults separating different compartments are semi-permeable, a transfer of fluid
from one compartment to the other (governed by the pressure difference between the
compartments) will occur.
300 Gas Material Balance Analysis Tool
GMB.1 Multi-Tank Gas MB Model

The PE² Essentials ‘Gas Material Balance’ tool was designed to model a fractured unconventional
gas reservoir. It creates a multi-tank material balance model that includes transmissibility
between the tanks in the form of a connection factor (Figure GMB-1). Note: Although the tool
was designed to forecast multi-tank production, a forecast for a single tank can be generated by
setting the connection factor to zero.
Figure GMB-1: PE² Essentials Multi-Tank Gas Material Balance Tool
The Gas Material Balance tool uses two interconnected tanks to generate the initial high decline,
evident in fractured low perm / unconventional reservoirs, followed by the lower long-term
depletion period. In general terms, the primary tank would be considered the near-frac reservoir
area and the secondary tank would be the reservoir that feeds the fractured system.
Once calibrated, the multi-tank model could be used to generate type curves for a specific area
based on reservoir parameters rather than a straight averaging of production data. This would
allow variability to be applied to specific well forecasts to generate a range of forecasts for
economic evaluation purposes.
The gas material balance model includes the option of using either the Hall correlation (refer to
PVT tool) or the Newman correlation (Equation GMB-3) to model rock compressibility.
Cr = 9.732x10-5 / (1 + 55.867 * Porosity) (GMB-3)

GMB.2 Water Drive Gas MB Model

The PE² Essentials ‘Gas Material Balance’ tool can also model a gas reservoir containing an aquifer
(Figure GMB-2). The aquifer is modeled as a Fetkovich finite aquifer.
Figure GMB-2: PE² Essentials Gas/Aquifer Material Balance Tool
Refer to the oil material balance tool for complete information on the Fetkovitch aquifer.
When running the gas/aquifer MB model, the response of no-aquifer is also plotted. This is useful
when plotting historical data, to evaluate the strength of the aquifer.
With the current version of the tools, it is not possible to forecast production using the
gas/aquifer MB model.
GMB.3 MB Tank Parameters

Dry gas material balance tanks do not require a large amount of input data in order to generate
a forecast. There are four input sections for the multi-tank model (Figure GMB-3).
Figure GMB-3: Gas Material Balance Tool – Input Parameters
The ‘Gas Properties’ input parameters are straight forward. There is no compositional analysis
performed on the gas in this model.
The ‘Primary Tank Parameters’ sets up the properties of the initial high decline tank. This is the
tank where all the well production comes from. For a single tank model, only the parameters for
this tank have to be entered.
The ‘Secondary Tank Parameters’ assigns the GIIP of the external reservoir to this tank and sets
the ‘Inter-Tank Communication Factor’ which determines the amount of support supplied to the
primary tank. Setting the ‘Inter-Tank Communication Factor’ to zero will stop all communication
between the tanks.
The ‘MatBal Forecast Parameters’ specifies the final depletion pressure for the primary tank. If
zero is entered here, the model will default to 100psi (690 kPa).
Based on the parameters listed in Figure GMB-3, the material balance depletion profile is shown
in Figure GMB-4 and Figure GMB-5.
Figure GMB-4: Gas Material Balance Tool – MB Depletion Plot

Figure GMB-5: Gas Material Balance Tool – MB Depletion Table
The data presented in Figure GMB-5 are as follows: ‘Cum Gas’ is the total production from the
multi-tank model; ‘P_Tank1’ and ‘PTank2’ are the pressures in Tank1 and Tank2, respectively;
and ‘Gp_Tank1’ and ‘Gp_Tank2’ are the production volumes contributed by each tank to the total
production.
If a match to historical production is required, the history data is imported, and the tank
parameters are modified until a match is achieved (refer to the example in Section GMB.6).
Historical production data can be imported from a CSV file or an Excel spreadsheet (Figure GMB-
6).
Figure GMB-6: Gas Material Balance Tool – Importing Historical Data

GMB.4 Material Balance Tank Forecast

Once the depletion characteristics of the multi-tank model are finalized, it is possible to add a
well to the primary tank and generate a production forecast (Figure GMB-7).
Figure GMB-7: Gas Material Balance Tool – Production Forecast
To generate a forecast, check the ‘Include Wellbore Calculations’ box (Figure GMB-1) and enter
the well parameters before running the material balance forecast (Figure GMB-8).
Figure GMB-8: Gas Material Balance Tool – Wellbore Parameters
The input wellbore parameters are straightforward. The ‘Tubing Correlations’ are similar to the
correlations presented in THP-BHP Tool.
The production forecast can be saved to PE Tools database for import into other tools.
GMB.5 Gas Material Balance Example

Table GMB-1 lists the historical production data for a vertical well in tight gas sand reservoir.
Inc Prod Pressure Inc Prod Pressure
Days Days
mmscf psi mmscf psi
1 3.418165 4000 1001 174.64 2869.38
91 311.053 3723.56 1092 170.86 2842.63
182 276.727 3497.2 1183 167.05 2815.82
273 248.165 3309.06 1274 163.25 2789.23
364 224.175 3150.54 1365 159.47 2763.03
455 203.83 3015.32 1456 155.76 2737.34
546 190.2 2983.97 1547 152.12 2712.26
637 187.68 2966.33 1638 148.56 2687.83
728 184.95 2944.98 1729 145.09 2664.09
819 181.78 2921.14 1820 141.72 2641.05
910 178.3 2895.71
Table GMB-1: Production History
The estimated reservoir parameters for Tank 1 are entered into the Material Balance tool (Figure
GMB-9) and the historical production data is imported into the model (Figure GMB-10). The Gas
MB models are contained in the PE Tools database located in the “PE Essentials 2019\Book
Examples\PEE Tools Database Examples” directory and the production data resides in the
‘Multiltank_Example.xlsx’ file in the “PE Essentials\Book Examples\Example Gas Multi-Tank
MBal” directory.
Figure GMB-9: Estimated Tank1 Parameters

Figure GMB-10: Production History
The first step to evaluating this well is to determine the characteristics of the Tank1. This is done
by setting the connection factor to zero and matching on the early data (Figure GMB-11).
Figure GMB-11: Tank 1 Match
The value of 5.3 Bscf will be a maximum value for this tank. When the second tank is added, the
volume in this tank will be reduced because of pressure support from Tank2.
The second step is to add the second tank and match the production history (Figure GMB-12).
The final match occurred with Tank1 containing 3 Bscf and Tank2 containing 38 Bscf.
Figure GMB-12: Multi Tank Match
Prior to generating a forecast, the wellbore parameters are entered and modified until the
historical production forecast matches the production history. The final reservoir parameters for
this example are presented in Figure GMB-13. After the production history is matched a
production forecast is then generated (Figure GMB-14).
Figure GMB-13: Multi Tank Forecast Well Parameters

Figure GMB-14: Multi Tank Gas Rate History Match and Forecast
The material balance forecast can be saved to the PE Tools database for use in other tools.
Oil Material Balance Analysis Tool

PE² Essentials ‘Oil Material Balance’ analysis and forecasting tool (Figure OMB-1) may be more
appropriate for use with conventional reservoirs than with unconventional reservoirs. It can be
used with unconventional reservoirs, but caution should be exercised when reporting and using
the results.
Figure OMB-1: PE² Essentials Oil Material Balance Analysis
One of the “anomalous” responses that has been observed when producing unconventional oil
reservoirs, is that the PVT properties of the oil may not follow the conventional correlations that
are used in PE² Essentials. Research is underway on this topic.
OMB.1 General Material Balance

Nearly all hydrocarbon reservoirs contain water-bearing rocks called aquifers. The aquifers may
be substantially larger than the oil or gas reservoirs so that they can appear to be infinite in size
There are numerous textbooks that cover material balance. The main references for the following
information is Mian, M.A.; Petroleum Engineering Handbook for the Practicing Engineer, Volume
1, PennWell, 1992 and Smith, C.R., Tracy, G.W., and Farrar, R.L.; Applied Reservoir Engineering,
Volume 2, OGCI, 1992.
The general material balance equation relates the original oil, gas, and water volumes in the
reservoir to production volumes and current pressure conditions and fluid properties. The
310 Oil Material Balance Analysis Tool
assumption of a tank behaviour means that the reservoir is considered to have the same pressure
and fluid properties at all locations in the reservoir. Consider Figure OMB-2.
Figure OMB-2: Reservoir Material Balance
The simplest way to visualize material balance is that if the measured surface volume of oil, gas
and water were returned to the reservoir at the reduced pressure, it must fit exactly into the
volume of the total fluid expansion plus any fluid influx.
The general material balance for an oil reservoir can be expressed as follows:
Net Reservoir Withdrawal = Expansion of Oil

+ Original Dissolved Gas
+ Expansion of Gas Cap
The general oil material balance equation is presented as Equation OMB.1.
N p Bo + (G ps + G pc − N p Rs ) B g − Gi B g + (W p − Wi ) Bw − We (OMB.1)
N=
( Bo − Boi ) + ( Rsi − Rs ) B g + mBoi ( B g − B gi ) / B gi + Boi (1 + m)(S w c w + c f )P /(1 − S w )
Where:
N p Bo = oil production
G pc Bg = gas cap production
(G ps − N p Rs ) Bg = liberated solution gas production
Gi Bg = gas injection
W p Bw = water production
Wi Bw = water injection
We = water influx from the aquifer

N ( Bo − Boi ) = expansion of initial oil
N ( Rsi − Rs ) Bg = volume occupied by liberated gas
 Bg − Bgi 
mNBoi   = gas cap expansion
 B 
 gi 
NBoi (1 + m)
S wcw P = water expansion
1 − Sw
NBoi (1 + m)
c f P = rock expansion
1 − Sw
P = Pi − Pr = pressure drop from initial pressure
Most of the terms in the material balance equation are either measured or derived from fluid
properties (PVT). The most significant issue is the determination of the aquifer influx term, W e.
Aquifer models ae normally used to predict We.
OMB.2 Aquifer Models

As reservoir fluids are produced and reservoir pressure declines, a pressure differential develops
between the aquifer and the reservoir causing water to encroach into the hydrocarbon bearing
zone. It is this water encroachment that leads to the water influx term, We, in the material
balance equation.
Aquifers are commonly classified by their degree of pressure maintenance as an active water
drive; a partial water drive; or a limited water drive. The term active water drive refers to an
aquifer in which the rate of water influx equals the reservoir total production rate. Active water
drive reservoirs are normally evident by their slow reservoir pressure decline.
Since wells are seldom (intentionally) drilled into an aquifer, there is uncertainty concerning the
porosity, permeability, thickness, geometry and extent of the aquifer. Commonly, these
properties are estimated using the reservoir properties.
Several models have been developed for estimating water influx that are based on assumptions
that describe the characteristics of the aquifer. The most common aquifer models are:
• Steady State Aquifer Model (Schilthuis)
• Finite Aquifer Model (Fetkovich)
• Unsteady State Infinite Aquifer Model (Van Everdingen and Hurst)
The PE² Essentials Oil Material Balance forecasting tool includes all three of these models. To
disable aquifer calculations and run a depletion drive model, check ‘Disable Aquifer’.
OMB.2.1 Schilthuis Steady State Aquifer Model

The simplest method for characterizing water influx was presented by Schilthuis in 1936 (Figure
OMB-3). This model is often tried first since the calculations are considerably less involved than
with either of the other two methods and can be performed by hand.
Figure OMB-3: Schilthuis Aquifer Model
The limitation of the Schilthuis model is the assumption that the aquifer is very large and highly
permeable. It is assumed that the permeability is so high that the pressure gradient across the
aquifer itself is negligible. In addition, the aquifer is assumed to be so large that the pressure
within the aquifer never declines, i.e., the initial pressure, Pi always exists within the aquifer.
In practical terms, for this model to be valid, the reservoir/aquifer system must have a relatively
high permeability: 50 md or more; the aquifer must be at least 10 to 20 times as large as the
reservoir and preferably at least 100 times as large.
Aquifer water rate is based on Darcy’s law for steady state fluid flow:
q w = C ( Pi − Pr ) (OMB.2)
Where: qw is the aquifer water rate in bbls/d, C is the water influx constant bbl/d/psi, P i is the
initial pressure in psi, and Pr is the current reservoir pressure in psi.
Equation OMB.2 is integrated to generate the cumulative water influx, W e, for the material
balance equation.
t
We = Cs  ( Pi − Pr )dt
0
n
(We ) n = C s  [( Pi − 0.5( Pr ( j −1) + Pr ( j ) )]t j (OMB.3)
j =1
Where: (We)n is cumulative water influx at time tn in mbbls, Cs is the Schilthuis aquifer constant
in mbbls/(month-psi), and time change, Δt, is in months.
For practical purposes, as a starting point when trying to match the aquifer performance, Cs can
be estimated as follows:
0.0002152k a ha / 360

Cs = (OMB.4)
 w ln(ra / re )
Where: Cs is the Schilthuis aquifer constant in mbbl/(month-psi), ka is the aquifer permeability in

md, ha is the thickness of the aquifer in feet, θ is the encroachment angle in degrees, µw is the
aquifer water viscosity in cp, ra is the radius of the aquifer in feet, and re is the radius of the
reservoir in feet.
As an example, for the aquifer parameters below, calculate We at 2976.6 psi and 3 months.
Pi = 3000psi ka = 500md ha = 20ft µw = 0.8cp ra = 20,000ft re = 1000ft θ = 360
From the aquifer parameters, Cs = 0.8979 mbbl/(month-psi). So, influx is calculated as follows:
We = 0.8979 [3000 - 0.5(3000 + 2976.6)] 3 = 31.5 mbbl
OMB.2.2 Fetkovich Aquifer Model

In 1971, Fetkovich developed a method for predicting the water influx of a finite aquifer (Figure
OMB-4). Fetkovich began with the premise that the productivity index concept is adequate for
describing water influx from a finite aquifer into a hydrocarbon reservoir. In other words, the
water influx rate is directly proportional to the pressure drop between the average aquifer
pressure and the pressure at the oil/water contact.
Figure OMB-4: Fetkovich Aquifer Model
The Fetkovich model uses two basic equations, the productivity index equation for the aquifer
and an aquifer material balance equation for a constant compressibility. These equations are as
follows:
q w = J ( Pa − Pf ) (OMB.6)
We = Va ce ( Pi − Pa ) / 360 (OMB.7)
Where: qw is water influx rate from the aquifer in mbbl/month, J is the aquifer productivity index
in mbbl/month-psi, Pa is the average aquifer pressure in psi, Pf is the pressure at the
reservoir/aquifer boundary in psi, We is the cumulative water influx in mbbls, Va is the initial
volume of water in the aquifer in mbbls, ce is the effective aquifer compressibility (cw + cf) in 1/psi,
cw is the water compressibility in 1/psi, cf is the formation compressibility in 1/psi, Pi is the initial
pressure in the aquifer in psi, and θ is the encroachment angle in degrees.
After manipulations and integrations (not presented here) the final equations for We is:
n n
(We ) n = Va ce B[ Pa ( j −1) − 0.5 ( Pf ( j −1) + Pf ( j ) )] / 360
j =1 j =1
 Jt 
− 
B = 1− e  Va ce 
0.0002152k a ha / 360

J= (OMB.8)
 w [ln(ra / re ) − 0.75]
Where: (We)n is cumulative water influx at time tn in bbls, B is the Fetkovich aquifer constant, Pa(j-
1) is the average aquifer pressure at the end of time step (j-1), Pf(j) is the average pressure at the
oil/water contact at time step j, Δt is time change in months, J is the aquifer productivity index
for a finite aquifer with a no flow outer boundary in mbbl/(month-psi), ka is the aquifer
permeability in md, ha is the thickness of the aquifer in feet, µw is the aquifer water viscosity in
cp, ra is the radius of the aquifer in feet, re is the radius of the reservoir in feet.
As an example, for the aquifer parameters below, calculate We at 2976.6 psi and 3 months.
Pi=3000psi ka=500md ha=20ft µw=0.8cp ra=100,000ft re=1000ft Ce=7x10-6 Va=2,195 mmbbls
From the aquifer parameters, J = 0.6978 mbbl/(month-psi), B = 0.1274 and Vace = 15.365
mbbls/psi. Water influx is as follows:
We = (15.365) (0.1274) [3000 - 0.5(3000 + 2976.6)] = 22.9 mbbl
The water influx, We, for the Fetkovich aquifer is similar to the infinite Schilthuis aquifer since the
ra/re ratio is 100, which is equivalent to an infinite aquifer.
It should be noted that the Fetkovich model can be used for non-circular aquifers through the
θ/360 term and modifying the J calculation as follows:
a) Finite aquifer, no flow outer boundary

0.0002152k a ha  / 360 0.0001028k a h a w
Radial: J = Linear: J =
 w [ln(ra / re ) − 0.75] w L
b) Finite aquifer, constant pressure outer boundary

0.0002152k a ha  / 360 0.00003426k a h a w
Radial: J = Linear: J =
 w ln(ra / re ) wL
Where: J is the productivity index of the aquifer in mbbls/month-psi, θ is the encroachment angle
in degrees, w is the width of the linear aquifer in feet, and L is the length of the linear aquifer in
feet.
OMB.2.3 Van Everdingen and Hurst Unsteady State Infinite Aquifer Model
In 1949, Van Everdingen and Hurst (VEH) proposed the most complete formulation of an aquifer
model (Figure OMB-5). They solved the diffusivity equation for the aquifer-reservoir system by
applying the Laplace transformation to the equations representing aquifer flow. Van Everdingen
and Hurst assumed that the aquifer is characterized by uniform thickness, constant permeability,
uniform porosity, constant rock compressibility and constant water compressibility.
Figure OMB-5: Van Everdingen and Hurst Aquifer Model
Van Everdingen and Hurst expressed the mathematical relationship for calculating water influx
in the form of a dimensionless parameter that is called cumulative influx function Q(td). This
dimensionless parameter was expressed as a function of the dimensionless time tD and
dimensionless radius rD, and as a result the solution to the diffusivity equation was generalized
and is applicable to any aquifer where the flow of water into the reservoir is essentially radial.
The Van Everdingen and Hurst solution was presented in tabulated and graphical form which
made it difficult to use. In 1988, Klins, M.,A.; Bouchard, A.,J.; and Cable, C.,L. presented a paper
(A Polynomial Approach to the Van Everdingen-Hurst Dimensionless Variables for Water
Encroachment, SPE 15433) that included equations that could be used to implement the VEH
solution. The equations for an infinite aquifer presented in Appendix F of the SPE paper have
been implemented in the Van Everdingen and Hurst aquifer model in PE² Essentials.
The Van Everdingen and Hurst equation for water influx is as follows:
We = C v PQ (t d ) (OMB.9)
Cv = 0.001119 h c r  / 360
a a e e
2
(OMB.10)
t d = At (OMB.11)
0.1924k a
A= (OMB.12)
 w ce re2
(OMB.13)
Pj = 0.5( Pj − 2 − Pj )
Where: We is cumulative water influx in mbbls, Cv is the Van Everdingen and Hurst aquifer
transmission constant in mbbls/psi, ΔPj is the constant pressure drop across the aquifer for time
step j (note ΔP1=0.5(Pi-P1)) in psi, Q(td) is the cumulative influx function, td is dimensionless time,
фa is the aquifer porosity in decimal, ha is the aquifer thickness in feet, ce is the effective aquifer
compressibility (cw + cf) in 1/psi, cw is the water compressibility in 1/psi, cf is the formation
compressibility in 1/psi, re is the radius of the reservoir in feet, θ is the encroachment angle in
degrees, ka is the aquifer permeability in md, µw is the aquifer water viscosity in cp, t is time in
months, and A is a time constant used to convert t to td.
From SPE 15433, Q(td) can be calculated as follows:
a) td <= 0.01
2 td
Q(t d ) = (OMB.14)

b) 0.01 <= td < 200
b0 (t d )b7 + b1 (t d ) + b2 (t d )b8 + b3 (t d )b9
Q(t d ) = (OMB.15)
b4 (t d )b7 + b5 (t d ) + b6
c) td > 200
Q(t d ) = 10(b10 +b11 ln( t d ) +b12 [ln( t d )]
b13
)
(OMB.16)
Where: bo = 1.129552, b1 = 1.160436, b2 = 0.2642821, b3 = 0.01131791, b4 = 0.5900113, b5 =

0.04589742, b6 = 1.00, b7 = 0.5002034, b8 = 1.500, b9 = 1.979139, b10 = 4.3989, b11 = 0.43693, b12
= -4.16078, and b13 = 0.09
For the aquifer parameters below, calculate We at 2976.6 psi and 3 months.
Pi = 3000psi ka = 500md ha = 20ft µw = 0.8cp re = 1000ft Ce = 7 x 10-6 фa = 0.2 θ = 360
From the aquifer parameters, Cv = 0.03133 mbbl/psi, A = 85.8928 /month and td = 257.7. Since
td > 200, equation OMB.16 is used to calculate Q(td) = 93.2866. Water influx is as follows:
We = (0.03133) (93.2866) 0.5(3000 – 2976.6) = 34.2 mbbl
For a finite aquifer, the SPE paper uses Bessel functions and hyperbolic cosecants to calculate
Q(td). The finite aquifer model has not been implemented in this version of PE² Essentials tools.
The Fetkovich model should be used for finite aquifers.
OMB.3 Depletion Drive Option

If the ‘Depletion Drive’ box is checked, the oil material balance tool will run a model using a
depletion / solution gas formulation (Figure OMB-6). There is no gas cap or aquifer when using
this option. Recovery above and below the bubble point pressure, as well as the formation of a
secondary gas cap, is modeled.
Figure OMB-6: Depletion Drive Model
If a match to historical production is required, the history data is imported with ‘Import History’
and the tank parameters are modified until a match is achieved (refer to the example in Section
OMB.6).
The data generated by the PE2 Essentials Oil Material Balance tool can be saved to the PE Tools
database for import into other tools, including the PE² Essentials Volumetric (MB) Analysis tool
for comparison to field data.
OMB.4 Height versus Pore Volume

Material balance analysis assumes that the system is a dimensionless tank – this would be similar
to building a 1-block reservoir simulation model. This assumption does not allow evaluation of
fluid contact depth or movement of the fluid contacts.
In a reservoir containing an aquifer, the water from the aquifer will encroach into the oil zone,
increasing the water saturation. In classical material balance calculations, the water saturation in
the tank will increase uniformly (no variation of Sw in the reservoir). However, if the increase in
water saturation is related to a pore volume fraction, then the movement of the OWC can be
calculated and more relevant water cut forecasts can be generated.
The PE2 Essentials Oil Material Balance tool subdivides the tank into 10 intervals (Figure OMB-7).
Figure OMB-7: Height vs Pore Volume
The calculation of water-oil ratios (WOR), becomes a two-step process. The conventional
material balance WOR is calculated as follows:
WORest = (krw / kro) * (μo / μw)
The WORest is then corrected based on the location of the oil-water contact as follows:
WOR = WORest * [Height Water / (Total Height – Height Water)] * (Bo / Bw)
The [Height Water / (Total Height – Height Water)] term represents the movement of the water
contact in the reservoir.
OMB.5 Aquifer Model Comparison

As a check on the different aquifer models, a comparison was made by inputting parameters for
an infinite aquifer. Figure OMB-8 presents the comparison showing that all aquifer models yield
similar results for a similar infinite aquifer.
Figure OMB-8: Aquifer Model Comparison
It is possible to use the tool to estimate the value for the aquifer constants before generating a
run by entering the aquifer properties. This is accessed by clicking the ‘Est Const’ button on the
main screen.
Alternatively, after manually changing the aquifer constants to generate a match, the ‘Est Const’
option can be used to determine the equivalent aquifer parameters that will yield the final
matched aquifer constants (refer to Section OMB.6).
OMB.6 Oil Material Balance Example – History Match

For history matching, reservoir pressure/volume history is imported through the “Import History”
button. The aquifer properties can then be modified to yield a good match of the historical data
which will then be used to forecast depletion of the reservoir.
Figure OMB-9 shows the importing of the example data presented by Smith et al on page 12-71;
included as ‘MB Data.xlsx’ in the “Book Examples\Example Oil Mat Bal” directory. The actual Oil
MBal models are included in the ‘PEE Tools Database Book Examples’ database included in the
“Book Examples\PEE Tools Database Examples” directory.
Figure OMB-9: Example Data from Smith et al
This example was evaluated using the Fetkovich aquifer model and matched with the following
aquifer input parameters (Figure OMB-10).
Figure OMB-10: Example Data from Smith et al – Matched Aquifer Parameters

A run was initially made using basic aquifer constants, then iterations were made by varying the
constants until an acceptable match to the historical data was achieved.
As a rule of thumb, if early history is not matched, modify the ‘J’ constant for the Fetkovich aquifer
model or the ‘A’ constant for the Van Everdingen and Hurst aquifer model. Late time pressure
support is matched by varying the transmission constants for either aquifer model after the early
time trend is matched.
The final history matched model is included in the ‘PEE Tools Database Book Examples’ database
and yields the history match presented in Figure OMB-11. Note that, for the match, initial oil in
place was assumed to be the value reported by Smith et al and the aquifer properties were
modified to obtain the match.
Figure OMB-11: Example Data from Smith et al – History Match
At the end of the history data, the reservoir pressure was 2568 psi. The We at this time was 11.6
mmbbls (versus 10.3 mmbbls as reported in Smith et al).
A closer match to the Smith et al aquifer estimate could be achieved by modifying the relative
permeability parameters. Note that the Smith et al example was not generating a forecast, it was
just attempting to “back out” oil in place and aquifer performance. The PE² Essentials Oil Material
Balance tool is a forecasting tool so all the reservoir effects are taken into account.
The material balance forecast will be saved with the model in the PE Tools database (‘Save to PE
Tools dBase’) for use with other tools.
Decline Curve Analysis
DCA.1 Introduction
Oil and gas wells usually reach a maximum rate shortly after completion after which they start
declining in production. A production “decline curve” indicates the amount of oil and gas
produced per unit of time for several consecutive periods. If the flowing conditions remain
constant, the resulting decline curve may be consistent and, if projected into the future, will yield
information as to the future production from the well.
Decline curve analysis (DCA) is a graphical procedure used for analyzing declining production
trends and forecasting future performance of oil and gas wells. Fitting a line through the plot of
a well’s performance history and assuming the trend will continue into the future forms the basis
of DCA. The caveat is that in the absence of stabilized production trends, DCA cannot be expected
to give reliable results. Since DCA is a means of predicting future well production based on past
production history, it is also a technique that can be used to identify well production problems.
Decline curves are the most common means used to forecast oil and gas production since: they
use data which is easy to obtain; the decline curves are easy to plot; they yield results on a time
basis; and they are relatively easy to analyze.
For more in-depth information on decline curve analysis and the models/equations available to
perform the analysis, refer to the Appendices. Appendix DCA1 describes a number of the
concepts associated with DCA including decline factors; the decline equations; and
considerations to keep in mind when performing DCA. Appendix DCA2 includes the formulations
of each of the decline models used in this tool. Appendix DCA3 presents the mathematical
definitions of DCA parameters.
DCA.2 Decline Curve Analysis Tool

PE² Essentials Decline Curve Analysis (DCA) is a tool that can be used to generate decline
parameters for a well and generate a forecast (Figure DCA-1).
An Excellent reference for this type of analysis is Poston, S., W. and Poe Jr., B. D., Analysis of
Production Decline Curves, SPE, 2008.
The ‘Decline Curve Analysis’ tool will generate the best-fit solution for the decline curve based
on the decline model chosen. Optionally, the parameters generated by the PE² Essentials ‘Monte
Carlo DC Forecast’ tool can be imported for use in the tool. Following decline curve analysis, a
production forecast can be generated.
324 Decline Curve Analysis
Figure DCA-1: PE² Essentials – Decline Curve Analysis Tool
The active Database being used for DCA is listed in the upper right area of the DCA tool screen. It
should be noted that the NormDCA (Section DCA.7) data is not part of the DCA database but is
stored separately in the Tools Database. Both the DCA database and the NormDCA data can be
stored in the same PEE Tools Database or different databases.
The active well is chosen from the dropdown menu on the upper left of the DCA tool screen.
The PE² Essentials Decline Curve Analysis tool includes enhanced decline curve analysis called
eDCA (Figure DCA-2).
With eDCA, it is possible to fit the production data to the following DCA models:
• Arps Hyperbolic Decline
• Stretched Exponential Decline
• Duong Decline
• Logistic Growth Decline
• Power Law Exponential Decline
• LeBlanc-Okouma Power Law Decline
Figure DCA-2: PE² Essentials – Enhanced Decline Curve Analysis (eDCA)
When using eDCA (Section DCA.4), the Arps equivalent parameters for the modeled well of
interest can be transferred to the DCA and DCA(Arps) Forecast tabs by clicking the ‘Transfer
Params to DCA / Forecast’ button. This will allow a forecast to be generated for the well as well
as to generate Monte Carlo DCA forecasts.
It should be noted that in order to generate a valid equivalent Arps model for any eDCA model,
other than Arps, that is valid for the production forecast, a forecast ‘End Time (yrs)’ should be
entered and either the ‘Generate DCA Model’ or the ‘Run Model/Forecast’ button clicked. When
using the Arps eDCA model, this forecast step is not required since the Arps and the equivalent
Arps models are the same.
The selected eDCA model will be forecasted to the entered end time and the combined history
and forecast will be used to generate the equivalent Arps model that will be valid for the entire
forecast period.
Refer to Section DCA.4 for information on using eDCA.

The ‘DCA’ tab (Figure DCA-3), is used to modify the Arps equivalent parameters or to generate a
different fit to the Arps model if so desired.
This tab is also used to generate the DCA analysis of the other phases (Section DCA.5) for use in
DCA forecasting.
Figure DCA-3: PE² Essentials – Arps Decline Curve Analysis Parameters
If the Arps model parameters are modified on this tab, the ‘Save to eDCA / Forecast’ button is
clicked to transfer the results to the other tabs. This is not necessary for the analysis of the other
phases which are automatically saved for forecasting.
It is possible to perform all the analysis on this tab, rather than eDCA, to generate the Arps DCA
model parameters.
Refer to Section DCA.5 for information on generating DCA parameters with this tab.
The ‘DCA(Arps) Forecast’ tab (Figure DCA-4), is used to generate a production forecast based with
the Arps equivalent parameters.
If the DCA forecast parameters for the other phases were generated on the ‘DCA’ tab, then this
forecast will also be included in the forecast. It is possible to set the secondary phase ratios to
constants for the forecast.
Figure DCA-4: PE² Essentials – DCA(Arps) Production Forecast
Refer to Section DCA.6 for more in-depth information on generating DCA forecasts with this tab.
Normalized DCA (Figure DCA-5) is used to generate an analysis and production forecast for
choked, constant rate wells.
The base reference for normalized decline curve analysis is Anderson, S., Anderson, D., Edwards,
K., Epp, K., Stalgorova, K., Pressure Normalized Decline Curve Analysis for Rate-Controlled Wells,
SPE 162923, 2012.
Figure DCA-5: PE² Essentials – Norm DCA Analysis and Production Forecast
Refer to Section DCA.7 for more in-depth information on generating NormDCA analysis and
forecasts.
DCA.3 Production Data Import

The PE² Essentials Decline Curve Analysis tool includes a DCA database that can include multiple
gas wells, oil wells and fields at the same time (Figure DCA-6).
Figure DCA-6: Decline Curve Analysis – DCA Database

The DCA database is independent of the PE Tools database and can be stored either independent
of the PE Tools database or in the PE Tools database.
To import data into the DCA database the ‘Import Well Data’ button on the ‘Import DCA Data’
tab is clicked (Figure DCA-7). The available wells in the current PE Tools database will be listed
and can be imported into the DCA tool (Figure DCA-8).
Figure DCA-7: Decline Curve Analysis – Well Data Import
Figure DCA-8: Decline Curve Analysis – Data Import Options
Well data is imported from the PE Tools database. This ensures that only edited analyzable data
is imported into the DCA tool. Note that if a well is already available in the DCA database, that
well will be disabled and cannot be re-entered. To re-enter a DCA well, delete it from the DCA
database (not the PE Tools database) using the ‘Delete Well’ button on the ‘Import DCA Data’
tab.
To permanently save the DCA data, click the ‘Save DCA Database’ menu button. This will present
an option to save the DCA database to the PE Tools database or to a standalone DCA database
(DVX) file.
To add additional production data to the well, select the well from the dropdown menu and click
the ‘Update Well Data’ button. This will update the production data from the PE Tools database
– make sure the correct PE Tools database is open.
To duplicate a well (for what-if analyses) or to delete a well from the DCA database, select the
well from the dropdown menu and click the ‘Duplicate Well’ or the ‘Delete Well’ button. A
confirmation screen will be displayed before the action occurs.
Note that any changes to the DCA database are not permanent until the ‘Save DCA Database’
button is clicked to update the database.
After the well data is entered, or a well is selected, the production data is listed and plotted for
review purposes (Figure DCA-9).
Figure DCA-9: Decline Curve Analysis – Input Data Review
The tool includes the capability of plotting each of the input parameters as well as a number of
calculated parameters such as calculated annual decline factor.
DCA.4 Enhanced DCA (eDCA)

Figure DCA-2 shows the eDCA tab which is used to perform the DCA analysis. eDCA includes 6
different decline models (Figure DCA-10). The included models are as follows:
• Arps Hyperbolic Decline

• Stretched Exponential Decline
• Duong Decline
• Logistic Growth Decline
• Power Law Exponential Decline
• LeBlanc-Okouma Power Law Decline
Figure DCA-10: PE² Essentials – eDCA Models
The reference documents and information for each model are listed in Appendix DCA2.
For the analysis of well performance, diagnostic plots are used to identify characteristic features
exhibited by production data. Diagnostic plots help to identify flow regimes (e.g., bi-linear or
linear flow, compound linear flow, etc.) and compare data to a well and reservoir model.
Generation of diagnostic plots is performed in the PE² Essentials PDA tool.
DCA.4.1 Model Fitting in eDCA

Since eDCA includes numerical derivatives of noisy production data, it may be necessary to apply
a “smoothing factor” to the derivate formulation. Figure DCA-11 shows the options for the
derivative calculation and smoothing.
Figure DCA-11: PE² Essentials – eDCA: Derivative Options

Derivatives can be calculated using cumulative production if the rate data is too noisy. This will
yield a very smooth derivative but the forecast cumulative may be an issue.
In the eDCA tool, an automatic model fitting routine is performed by multi-component linear
regression using the Levenberg–Marquardt algorithm (refer to the Wikipedia website:
https://en.wikipedia.org/wiki/Levenberg%E2%80%93Marquardt_algorithm for information about this
algorithm). The regression interval can be restricted by entering the start time and end time of
the period (Figure DCA-12).
Figure DCA-12: PE² Essentials – eDCA: Regression Interval
Caution should be used when restricting the regression interval. It is possible that the regression
will indicate exponential decline if the early time is not included in the regression.
To fit an eDCA model, the model is chosen (Figure DCA-10) and the ‘Generate DCA Model’ button
is clicked (Figure DCA-13) and the best-fit for the model will be generated.
Figure DCA-13: PE² Essentials – eDCA Model Generation
If the DCA database contains more than one well, the ‘Fit All Wells to Model’ button can be
clicked and all wells in the database will be fitted to the chosen model.
DCA.4.2 Equivalent Arps Parameters

If an eDCA model other than Arps is selected, the equivalent Arps parameters that will yield a
similar result as the eDCA model will be generated for the model (Figure DCA-13).
To generate a DCA Arps-based forecast from the eDCA model, the forecast parameters are
entered (Figure DCA-14) and the ‘Run Model/Forecast’ button is clicked. This button is also
clicked if the eDCA model parameters are modified and a new Arps equivalent model is to be
generated.
Figure DCA-14: PE² Essentials – eDCA: Generating a Forecast
The equivalent Arps parameters will approximate the match and the forecast. If required, a value
for Dlim (Figure DCA-13) can be entered to improve the late time match of the equivalent Arps
forecast.
After completing the eDCA analysis, the equivalent Arps parameters for the selected eDCA model
are transferred to the ‘DCA’ tab and the ‘DCA(Arps) Forecast’ tab by clicking the ‘Transfer Equiv
Arps Params to DCA’ button.
The eDCA generated forecast can be saved to the PE Tools database by clicking the appropriate
button on the main menu. The eDCA model parameters can be exported to a csv file by clicking
the ‘Report eDCA Results’ button on the main menu. A single well or multiple well data can be
exported to the CSV file (Figure DCA-15).
Figure DCA-15: PE² Essentials – Export eDCA Results

DCA.5 DCA Parameters

The PE² Essentials Decline Curve Analysis tool can be used as an Arps model production decline
curve and forecasting tool on the ‘Arps’ tab (Figure DCA-16). This tab is also where the secondary
phase forecasting models are generated.
Figure DCA-16: PE² Essentials – Arps DCA Analysis
As a starting point, two points can be entered on the plot and an exponential DCA model will be
generated and is shown in the ‘Manual Analysis’ box. To transfer the analysis to the Arps model,
click the ‘Use Results’ button. The type of Arps model can then be modified by varying the
parameters until an acceptable match is achieved. To modify the line, just enter two more points
and click ‘Use Results’ when the line is satisfactory, and the original line will be replaced.
The ‘Manual Analysis’ option only generates parameters for an exponential decline.
Alternatively, a new analysis option has been added to the DCA tab and is accessed by selecting
‘1/D vs Time’ on the drop-down menu. (Figure DCA-17) – also refer to Section A1.3.2.2.
Figure DCA-17: PE² Essentials – Estimating Hyperbolic Parameters
This option allows an estimate of hyperbolic decline and b parameters to be generated. The initial
rate is not generated and must be entered manually (Figure DCA-18).
Figure DCA-18: PE² Essentials – Estimating Initial Rate

After the ‘Calculate’ button is clicked, the optimum model is generated and plotted within the x-
Range specified in the ‘Model X-Plot Range’. The results are also summarized in the ‘Decline
Curve Analysis Results’ box.
This tab is also where the secondary phase forecast models, water and GOR/CGR, are generated
for forecasting purposes (Figure DCA-19).
Figure DCA-19: PE² Essentials – Secondary Phase Analysis
Secondary phase models are based on a linear relationship with Cum Oil/Gas. Water and
GOR/CGR forecast models are not required to generate a production forecast, but if available,
water and oil/condensate rates will be included in the DCA production forecast.
There are two options to generate the secondary phase forecast models. Two points can be
placed on the graph (and moved around by clicking and dragging the point) and the resulting
analysis transferred to the DCA forecast model by clicking the ‘Use Results’ button.
Alternatively, a linear regression can be used to generate the forecast model for the secondary
phase or used to refine the manually generated forecast model. Regression will generate the
forecast model for the ‘Model X-Plot Range’ and save it to the DCA forecasting model.
DCA.6 DCA Production Forecasting

After entering the ‘Forecast Parameters’ (Figure DCA-20) on the ‘DCA(ARPS) Forecast’ tab, a
forecast can be generated by clicking the ‘Run Forecast’ button (Figure DCA-21).
Figure DCA-20: Decline Curve Analysis – Production Forecast Parameters
Figure DCA-21: Decline Curve Analysis – Production Forecast

The transferred eDCA/DCA results are used and the forecast begins at the end of the production
history. To change the ‘Forecast Decline Parameters’, use the eDCA or DCA tabs. Note that all
forecasts are generated using the equivalent Arps parameters.
To generate a forecast, the default is to start the forecast from the final rate but this can be
changed to any starting rate by entering a value in the ‘Start Forecast Rate’ (Figure DCA-18). To
reset the rate to the last historical rate, set the ‘Start Forecast Rate’ value to zero prior to clicking
the ‘Run Forecast’ button.
The 'Min Eff Decline Factor' is used with hyperbolic forecasts. Once the minimum decline factor
is reached, the forecast will revert to an exponential decline forecast at the specified minimum
decline factor for the remainder of the forecast period.
The 'Water Forecasting Option' is available if a forecast model has been generated for the water
data on the ‘DCA’ tab. The water forecast options available to be used for forecasting will depend
on which secondary phase forecasting models are available.
The 'Minimum/Cutoff Rate' and 'Maximum Water Cut' ('Maximum WGR' for a gas well)
parameters are used to limit the duration of the forecast. The forecast time will be unlimited
unless these parameters limit the forecast, or the time period is limited by entering ‘Max Forecast
Years’.
The ‘Forecast Options’ section (Figure DCA-22) modifies the secondary phase forecast. To
forecast a secondary phase at a constant value, or zero, check the appropriate box and enter the
required value for ‘Starting Value’. This value will be used for the entire secondary phase forecast.
Figure DCA-22: Decline Curve Analysis – Production Forecast Options
If the secondary phase forecast model has been generated, it is possible to modify the starting
value for the forecast. The default value for a secondary phase is the final historical value. For
example, the value of 78.58499 in Figure DCA-20 is the final historical water cut for this well but
this could be changed to any other value to start the forecast.
Note that for an oil well, water cut is always entered for the water phase starting value regardless
of what water forecasting model is used.
As an alternative to using the Arps parameters generated by the DCA tool, decline parameters
generated by the PE² Essentials Monte Carlo DC Forecast tool and stored in the PE Tools database
can be imported for forecasting (Figure DCA-23).
Figure DCA-23: Decline Curve Analysis – Optional Forecast Decline Parameters
Once Monte Carlo decline parameters are loaded, the specific parameters to be used for the
forecast are chosen and the forecast is run.
To save alternative forecasts, the well should be duplicated for each forecast. To run and store a
P10, P50 and P90 forecast, three versions of the well are required to be saved to the database.
The PE² Essentials Monte Carlo DC Forecast tool generates and saves decline parameters for the
P01 to P99 values. As a result, any of the Pxx parameters can be used to generate a forecast by
selecting the ‘User P’ option and entering the specific Pxx value to use.
The forecast results can be saved to the PE Tools database for import into other PE² Essentials
tools by using the ‘Save Forecast to PE Tools dB’ or ‘Save History/Forecast to PE Tools dB’ button.
Saving the history and the forecast data allows full cycle economics to be run.
After the forecast has been saved to the PE Tools database, the PE² Essentials Production Analysis
Tool can be used to view and export the data to a csv file.
DCA.7 Normalized DCA

The DCA tool includes a normalized DCA technique that is used to perform decline curve analysis
for choked, constant rate wells (Figure DCA-24). It is accessed from the ‘NormDCA’ tab.
The NormDCA well data is not part of the DCA database but is stored separately in the PE Tools
Database because of the different data requirements for NormDCA. Normalized DCA requires
pressure (THP, CHP or BHP) and PVT data which is not required for conventional eDCA/DCA.
Since NormDCA requires a forecast of flowing pressure (THP, CHP or BHP) only gas wells can be
analyzed and forecasted by the PE² Essentials DCA tool. To analyze and forecast oil wells, water
and gas forecasting would have to be included to forecast flowing pressures.
Figure DCA-24: PE² Essentials – Normalized Decline Curve Analysis (NormDCA) Tool
For a description of the NormDCA technique refer to Appendix A2.8.
Norm DCA wells can only be loaded from the ‘Norm DCA’ tab. Clicking the ‘Import Well Data’
button will read the current PE Tools database and present a list of gas wells in the database that
include pressure, which is required for NormDCA analysis. This will also pre-load any Norm DCA
wells that are available in the database.
Selecting and loading a well will automatically create and add the NormDCA well to the PE Tools
database. After analyzing the well, the ‘Update Well in dBase’ button should be clicked to update
and save the analysis in the database. Note – you must update the well in the database before
changing to another well if you want to save the analysis.
NOTE – to copy a well to a new well in the same PE Tools database, change the name of the well
before clicking ‘Update Well in dBase’. A new NormDCA well will be created with the new well
name.
After NormDCA wells have been created and saved to a PE Tools database, ‘Load Norm DCA
Wells’ will load the well list for subsequent runs. To add additional wells to the well list click the
‘Import Well Data’ button to list the wells that are available for NormDCA analysis. Wells from
other PE Tools databases can be imported this way after the database has been opened.
To copy NormDCA wells to another database, load the NormDCA wells, open the target PE Tools
database and copy each well to the target database using the ‘Update Well in dBase’ button.
NormDCA analysis is performed by placing two points on the q/Δp or q/ΔΨ versus Cum plot to
generate a straight. The line can be moved by dragging one of the end points of the line. ‘Use
Results’ will transfer the data to the appropriate ‘DCA Analysis’ box for use (Figure DCA-25 and
Figure DCA-26).
Figure DCA-25: NormDCA – Normalized Rate Plot
Figure DCA-26: NormDCA – NormDCA Parameters
Note, it is only possible to place points on the plot when the proper parameters and scales are
selected for the graph.
The second part of the analysis is to set up the pressure forecasting model. Choose whether to
use THP, CHP or BHP in ‘Normalization Parameters’ (Figure DCA-27) and enter the value for Pi.
The ‘Pressure >’ option enables removing low pressures from the plot.
Figure DCA-27: NormDCA – Normalization Parameters
Select time for the x-axis and place two points on the plot to generate a straight line on the
Pressure versus time plot (Figure DCA-28). Clicking the ‘Use Results’ button will transfer the data
to the appropriate ‘DCA Analysis’ boxes.
Figure DCA-28: NormDCA – Pressure Plot
As an option, ‘Min Rate’, ‘Min Pressure’ and ‘Forecast Years’ can be entered to limit the forecast
(Figure DCA-29). Entering zeros for these parameters will run the forecast out to a maximum of
50 years.
Figure DCA-29: NormDCA – Forecast Parameters
Note that ‘Min Pressure’ should be entered and represents the pressure at which the well is no
longer choked and the rates start to declining.
A forecast is generated by clicking the ‘Generate Forecast’ button. The monthly or yearly forecast
results, as well as the historical data, can be viewed by selecting the relevant button. The
NormDCA well forecast can be saved to the PE Tools database for subsequent use.
For the ‘Y-Axis’ options (Figure DCA-30), checking/unchecking the ‘Show Forecast’ box will cycle
between the forecast and the analysis plot. Note – only the analyzed plots will include the analysis
straight line. For example, if CHP analysis is selected, the analysis line will only be plotted on the
CHP vs Years plot.
Figure DCA-30: NormDCA – y-Axis Options
Following the NormDCA forecast the equivalent Arps parameters representing the forecast can
be generated by clicking the ‘Equiv Arps’ button (Figure DCA-31).
Figure DCA-31: NormDCA – Equivalent Arps Parameters for NormDCA

DCA.8 DCA Example – Marcellus

Marcellus production data (Figure DCA-32 and Table DCA-1) was downloaded from the Internet:
http://shale.typepad.com/marcellusshale/production-curve.
Figure DCA-32: Marcellus Production Curves
Table DCA-1: Marcellus Production Data
The data in Table DCA-1 is available in the ‘Marcellus 7Bcf Data.xlsx’ file located in the “PE
Essentials\Book Examples\Example Marcellus\DCA” directory. The Excel data was imported into
a PE Essentials Production Database (PEE Database Book Examples.DVXdb) which is located
stored in the “Book Examples\PEE Production Database” directory using the Production Database
Tool.
The data was then imported into PDA and stored in a PE Tools database (PEE Tools Database
Book Examples.PEEdb) located in the “Book Examples\PEE Tools Database Book Examples”
directory (Figures DCA-33 and DCA-34).
Figure DCA-33: Marcellus Production Data Imported to Production Database
Figure DCA-34: Marcellus DCA Example Production Data

The data was then imported into the DCA tool by clicking ‘Import Well Data’ (Figure DCA-32) and
saved to a standalone DCA database ‘PE_Essentials_DCA_DataBase_Marcellus.dvx’ in the “Book
Examples\Example Marcellus \DCA” directory.
The log(rate) versus cumulative plot indicated a straight line so ‘Super Hyperbolic’ was chosen
for the regression analysis. Regression was started after 0.1 years to filter out the early data. The
results are shown in Figure DCA-35.
Figure DCA-35: Marcellus DCA Analysis
Note that the ‘Start Forecast Rate’ of 1558 mscf/d is the rate indicated by the decline curve at
the end of producing time. It is possible to change the rate for the start of the forecast to a more
appropriate value.
This example is continued in Example included with the PE2 Essentials Mont Carlo DC Forecast
tool.
DCA Appendix 1– DCA Theory & Practice
A1.1 Introduction
There are numerous phases in the development of an oil or gas field and “estimated ultimate
recovery” (EUR) calculation techniques are different for each phase (Figure A1.1).
Figure A1.1 – Production Forecasting based on Field Development Phase
During the exploration phase, there is no specific information available about the reservoir so
analogue field information is used to generate EUR. After discovery and during the delineation
phase, in-place volumes and reservoir parameters are determined and models are used to
generate production forecasts and determine EUR. Models can be as simple as well type-curves
or analytical models, to a full reservoir simulation model. During the production life of the field,
these models are updated/calibrated and continue to be used to update production forecasting
and EUR.
All production has an initial transient flow period followed by a boundary-dominated flow period.
During the transient flow period, the pressure at the flow boundary remains constant at the initial
reservoir pressure and the boundary moves away from the well into the reservoir. This period of
a well’s flow is characterized by very high decline rates. When the flow boundary reaches the
actual reservoir boundary, which could be the flow boundary of another well, the pressure at the
boundary starts to decline and the well exhibits boundary-dominated flow. It is within this flow
period that traditional decline curve analysis is performed.
Analysis of a decline curve is possible after the declining production trend has been established
and flowing pressure is relatively constant. This typically occurs anytime after >35% of the
ultimate recovery has been produced. Decline curve analysis is the preferred tool of reserves
evaluators/auditors.
A1.2 Decline Curve Analysis Theory

Production decline analysis is an empirical technique that extrapolates trends in the production
data from oil and gas wells. The purpose of decline curve analysis is to generate a forecast of
future production rates and to estimate the ultimate recoverable volumes (EUR).
Oil and gas wells will initially reach a maximum rate after which they start declining in production.
A production “decline curve” indicates the amount of oil and gas produced per unit of time for
several consecutive periods. If the flowing conditions (pressure) remain constant, the resulting
decline curve may be consistent and, if projected into the future, will yield information as to the
future production from the well.
DCA is a graphical procedure used for analyzing declining production trends and forecasting
future performance of oil and gas wells. Fitting a line through the plot of a well’s performance
history and assuming the trend will continue into the future forms the basis of DCA. The caveat
is that in the absence of stabilized production trends, DCA cannot be expected to give reliable
results. Since DCA is a means of predicting future well production based on past production
history, it is also a technique that can be used to identify well production problems.
All decline curve analysis models begin with the concept of the instantaneous decline rate (𝐷),
which is called the nominal decline, and is defined as the fractional change in rate (𝑑𝑞) per unit
time (𝑑𝑡) and is presented as Equation A1-1. Note to convert 𝐷 from annual to monthly, 𝐷𝑚 =
𝐷/12 and for daily, 𝐷𝑑 = 𝐷/365.
1 𝑑𝑞 𝑑(ln(𝑞))
𝐷= − = − (A1-1)
𝑞 𝑑𝑡 𝑑𝑡
When production is plotted as flow rate versus time, the nominal decline is equal to the slope at
𝑑𝑞
a point in time ( 𝑑𝑡 ) divided by the rate (𝑞) at that point (Figure A1-2).
Figure A1-2: Nominal Decline
This calculation method for the decline factor is termed the tangent method.
Excellent references for DCA are as follows:

• Poston, S., W. and Poe Jr., B. D.; Analysis of Production Decline Curves, SPE,
2008(1)
• Poston, S., W.; Laprea-Bigott, Marcelo and Poe Jr., B. D., Analysis of Oil and
Gas Production Performance, SPE, 2019(2)
• Monograph 4, Estimating Ultimate Recovery of Developed Wells in Low-
Permeability Reservoirs, SPEE, August 2016(3)
• LeBlanc, Don; Petroleum Engineering & Economics Essentials: Tools and
Techniques to Evaluate Unconventional (and Conventional) Wells and
Reservoirs, https://www.eastexpetroleum.com/PE_Essentials, 2020(4)
There are several models available for use in decline curve analysis:
o Arps
o Stretched Exponential Decline
o Duong Decline
o Logistic Growth Decline
o Power Law Exponential Decline
o LeBlanc-Okouma Power Law Decline
Only the Arps and Stretched Exponential Decline (SEDM) models are presented herein.
Information on the other models is available in the LeBlanc(4) reference.
A1.2.1 Arps General Decline Model

Historically there are three types of production decline trends that have been presented:
exponential, hyperbolic and harmonic. With the development of unconventional and shale
reservoirs, additional decline trends have been observed: transient, super-hyperbolic.
The decline equations defining these trends are not necessarily grounded in fundamental theory
but are based on empirical observations of production decline, although there is some theoretical
basis for the decline processes. It can be demonstrated that under certain conditions, such as
constant well flowing pressure, the fluid flow equation under boundary dominated flow is
equivalent to an exponential decline.
Arps introduced a constant b-factor to account for the observation that the nominal decline
factor may change over time (Equation A1-2).
1
𝑑(𝐷)
𝑏= = 𝑐𝑜𝑛𝑠𝑡𝑎𝑛𝑡 (A1-2)
𝑑𝑡
Integrating Equation A1-2 from 0 to t and defining the initial decline rate at t = 0 as 𝐷𝑖 , then:
𝐷𝑖
𝐷= (A1-3)
1 + 𝑏𝐷𝑖 𝑡
Using Equation A1-1, Equation A1-3 can be rewritten as Equation A1-4

𝑑(ln(𝑞)) 𝐷𝑖
− = (A1-4)
𝑑𝑡 1 + 𝑏𝐷𝑖 𝑡
Integrating Equation A1-4 from 0 to t yields the Arps general decline equation (Equation A1-5).
This is a hyperbolic equation, valid for b>0 and is plotted in Figure A1-3.
𝑞 = 𝑞𝑖 (1 + 𝑏𝐷𝑖 𝑡)−1⁄𝑏 (A1-5)
Where:
𝑞 = Instantaneous Rate at time 𝑡 [Volume / Unit Time]
𝑞𝑖 = Initial Instantaneous Rate [Volume / Unit Time]
𝐷𝑖 = Initial Nominal Decline Factor [Fraction / Unit Time]
𝑏 = Hyperbolic Exponent factor (Valid for all 𝑏 <> 1)
𝑡 = Time in units consistent with units in the decline and rate parameters
Figure A1-3: Arp’s Hyperbolic Decline Equation
In general terms, the b-factor has the following characteristics:

• b = 0: Exponential decline (linear: log rate versus time)
• 0 < b < 1: Hyperbolic decline (non-linear: log rate versus time)
• b = 1: Harmonic decline (linear: log rate versus cum)
• b > 1: Super harmonic / Transient decline (tight sands and shale)
• b = 2: Ideal Linear flow (fractured well)
• b = 4: Ideal Bilinear flow (horizontal fractured well)
A1.2.1.1 Exponential Decline (b=0)

When b = 0, Equation A1-3 indicates that 𝐷 is a constant (𝐷 = 𝐷𝑖 ) and Equation A1-1 can be
integrated from 0 to t to yield the exponential rate decline equation (Equation A1-6).
𝑞 = 𝑞𝑖 𝑒 −𝐷𝑡 (A1-6)
Where:
𝑞𝑖 = Initial Instantaneous Rate (𝑡 = 0) [Volume / Unit Time]
𝐷 = Nominal Decline Factor [Fraction / Unit Time]
𝑡 = Time in units consistent with the decline and rate
The exponential decline equation (Equation A1-6) can be integrated with respect to time. This
results in the cumulative volume produced since time 0 (Equation A1-7). Note at time 0, 𝑞 =𝑞𝑖 .
𝑞𝑖 − 𝑞
𝑄𝑝 = (A1-7)
𝐷
Where:
𝑄𝑝 = Volume produced to rate 𝑞 from time zero
The signature for exponential rate decline is a straight line on a log(rate) versus time plot and a
rate versus cumulative production plot.
Exponential decline has been observed for the following conditions.

• Oil reservoirs above the bubble point (undersaturated oil)
• Oil reservoirs with gravity drainage from down dip wells
• Oil reservoirs with solution gas drive and unfavourable kg/ko
• Gas reservoirs with high reservoir pressure (liquid-like compressibility)
• Gas reservoirs with wells experiencing liquid-loading problems
A1.2.1.2 Effective Decline Factor
Although nominal decline is used in all the decline equations, effective decline is a more intuitive
decline measurement. Unlike nominal decline, it can be read directly from tabular data. It is
generated by the slope of a straight line from time 0 to a point 1-year later (Figure A1-4).
Figure A1-4: Effective Decline
The effective decline equation is defined by equation A1-8 and is also called the secant method.
It is commonly used when manually forecasting from tabular data.
𝑞𝑖 − 𝑞
𝐷𝑒 = (A1-8)
𝑞𝑖
Where:
𝐷𝑒 = Effective Decline Factor [Fraction / Unit Time]
𝑞𝑖 = Instantaneous Rate at time 0 [Volume / Unit Time]
𝑞 = Instantaneous Rate at time one year [Volume / Unit Time]
The most common time period used to evaluate 𝐷𝑒 is one year. If no time period is stated, a
yearly period is implied. Effective decline (𝐷𝑒 ) in units of volume per year can be converted to
effective monthly decline (𝐷𝑒𝑚 ) or effective daily decline (𝐷𝑒𝑑 ) using Equation A1-9 or A1-10.
𝐷𝑒𝑚 = 1 − (1 − 𝐷𝑒 )1/12 (A1-9)
𝐷𝑒𝑑 = 1 − (1 − 𝐷𝑒 )1/365 (A1-10)
Some confusion exists regarding these decline factors, but both the nominal (tangent) and
effective (secant) decline factors are theoretically correct. The main difference between the
decline factors is that the nominal decline is a continuous function, used in the equations relating
to decline curve analysis, and the effective decline factor is a stepwise function that is easy to
calculate and use with tabular historical data.
For exponential decline, it is simple to convert from a nominal decline factor to an effective
decline factor and vice versa using Equations A1-11 and A1-12.
𝐷𝑒 = 1 − 𝑒 −𝐷 (A1-11)
𝐷 = −𝑙𝑛(1 − 𝐷𝑒 ) (A1-12)
As an example, an annual effective decline factor of 0.37 will give a nominal decline factor
(Equation A1-12) of 0.46. Converting the nominal decline factor to a daily decline is 0.00126
(𝐷/365). To calculate the daily effective decline factor, Equation A1-10 is used to give 0.00127.
A1.2.1.3 Hyperbolic Decline (0<b<1)

Commonly accepted 𝑏 values for single porosity reservoirs with good drive energy usually range
between 0 and 1. Reservoirs with low permeability, multi-porosity, fracture stimulations or poor
reservoir drive energy (gravity drainage) may demonstrate 𝑏 factors greater than 1 but they
seldom go above 2.
Hyperbolic decline is the most commonly observed decline trend. It occurs when 0 < b < 1 in the
hyperbolic decline equation (Equation A1-5). Equation A1-5 has three constants: the initial
production rate (𝑞𝑖 ), the initial decline rate (𝐷𝑖 ), and the hyperbolic exponent (𝑏).
The hyperbolic decline equation can be integrated to yield the cumulative production (Equation
A1-13).
𝑞 = 𝑞𝑖 (1 + 𝑏𝐷𝑖 𝑡)−1⁄𝑏 (A1-5)
𝑞𝑖𝑏
𝑄𝑝 = (𝑞1−𝑏 − 𝑞1−𝑏 ) (A1-13)
(1 − 𝑏)𝐷𝑖 𝑖
Where:
𝐷𝑖 = Initial Nominal Decline [Fraction / Unit Time]
𝑏 = Hyperbolic Exponent
The decline factor is not a constant but changes with time. The hyperbolic exponent is the rate
of change of the decline rate with respect to time. This means that ‘b’ is actually the second
derivative of production rate with respect to time. The nominal decline factor at any time, 𝐷(𝑡),
can be determined from the hyperbolic exponent as shown in Equation A1-3.
𝐷𝑖 (A1-3)
𝐷(𝑡) =
A1.2.1.4 Modified Hyperbolic Decline

Extrapolation of curves exhibiting hyperbolic behaviour over long periods of time may result in
unrealistically high reserves because of the continuously decline D. To avoid the problem of
overestimating production, it is common practice to convert the hyperbolic decline into an
exponential decline at some future time.
For example, assume that the decline rate starts at 30% and decreases through time in a
hyperbolic trend. When it reaches a specified value, 10% for example, the hyperbolic decline
(Equation A1-5) can be changed to an exponential decline (Equation A1-7) and the forecast
continued using the exponential decline rate of 10%. In other words, the forecast is converted to
an exponential forecast when 𝐷(𝑡) = 𝐷𝑙𝑖𝑚 .
The rate at which the conversion occurs (𝑞𝑙𝑖𝑚 ) can be calculated from Equation A1-14 and the
corresponding time (𝑡𝑙𝑖𝑚 ) can be calculated from Equation A1-15.
𝐷𝑙𝑖𝑚 1⁄𝑏
𝑞𝑙𝑖𝑚 = 𝑞𝑖 ( ) (A1-14)
𝐷𝑖
𝑞
(𝑞 𝑖 )𝑏 − 1
𝑡𝑙𝑖𝑚 = 𝑙𝑖𝑚 (A1-15)
𝑏𝐷𝑖
The rate (𝑞) for the modified hyperbolic decline is calculated with Equations A1-5 and A1-16.
Cumulative production is calculated with Equations A1-13 and A1-17:
𝐹𝑜𝑟 𝐷 > 𝐷𝑙𝑖𝑚 : 𝑞 = 𝑞𝑖 (1 + 𝑏𝐷𝑖 𝑡)−1⁄𝑏 (A1-5)
𝐹𝑜𝑟 𝐷 ≤ 𝐷𝑙𝑖𝑚 : 𝑞 = 𝑞𝑙𝑖𝑚 𝑒−(𝐷𝑙𝑖𝑚 (𝑡−𝑡𝑙𝑖𝑚 )) (A1-16)
𝑞𝑖𝑏
𝐹𝑜𝑟 𝐷 > 𝐷𝑙𝑖𝑚 : 𝑄𝑝 = (𝑞𝑖1−𝑏 − 𝑞1−𝑏 ) (A1-13)
(1−𝑏)𝐷𝑖
𝑞𝑖𝑏 𝑞𝑙𝑖𝑚 − 𝑞
𝐹𝑜𝑟 𝐷 ≤ 𝐷𝑙𝑖𝑚 : 𝑄𝑝 = (𝑞𝑖1−𝑏 − 𝑞𝑙𝑖𝑚
1−𝑏
)+ (A1-17)
(1 − 𝑏)𝐷𝑖 𝐷𝑙𝑖𝑚
Where:
𝑞𝑙𝑖𝑚 = Instantaneous Rate at 𝑡 = 𝑡𝑙𝑖𝑚 [Volume / Unit Time]
𝑡𝑙𝑖𝑚 = Time when 𝐷 = 𝐷𝑙𝑖𝑚 [Fraction / Unit Time]
𝐷𝑙𝑖𝑚 = Limiting Nominal Decline / Exponential Decline [Fraction / Unit Time]
A1.2.1.5 Harmonic Decline (b=1)
When b=1, this is called harmonic decline. The existence of a harmonic decline may be indicating
that production is still in the transient flow period. It may also be evident in reservoirs with
extremely large and active aquifers or in very efficient water floods. The signature of harmonic
decline is a straight line on a log(rate) versus cumulative plot or a log(rate) vs log(time) plot.
The harmonic decline equations use Equation A1-5 with 𝑏 =1 and are given as Equations A1-18
and A1-19.
𝑞𝑖
𝑞= (A1-18)
(1 + 𝐷𝑖 𝑡)
𝑞𝑖 𝑞𝑖
𝑄𝑝 = 𝑙𝑛( ) (A1-19)
𝐷𝑖 𝑞
Where:
A1.2.1.6 Modified Harmonic Decline
Since nominal decline is not constant, unconstrained harmonic curves will severely overestimate
future production. To avoid the problem of overestimating production, it is common practice to
convert the harmonic decline into an exponential decline at some future time. The nominal
decline factor at time, 𝐷(𝑡), can be determined from Equation A1-3, with 𝑏 = 1.
𝐷𝑖
𝐷(𝑡) = (A1-20)
1 + 𝐷𝑖 𝑡
For example, assume that the decline rate starts at 30% and decreases through time in a
harmonic trend. When it reaches a specified value, 10% for example, the harmonic decline
(Equation A1-18) can be changed to an exponential decline (Equation A1-7) and the forecast
continued using the exponential decline rate of 10%. In other words, the forecast is converted to
an exponential forecast when 𝐷(𝑡) = 𝐷𝑙𝑖𝑚 .
The rate at which the conversion occurs (𝑞𝑙𝑖𝑚 ) can be calculated from Equation A1-21 and the
corresponding time (𝑡𝑙𝑖𝑚 ) can be calculated from Equation A1-22.
𝐷𝑙𝑖𝑚
𝑞𝑙𝑖𝑚 = 𝑞𝑖 ( ) (A1-21)
𝐷𝑖
𝑞
(𝑞 𝑖 ) − 1
𝑙𝑖𝑚 (A1-22)
𝑡𝑙𝑖𝑚 =
𝐷𝑖
The rate for the modified harmonic decline is calculated with Equations A1-18 and A1-16.
Cumulative production is calculated as follows (Equations A1-19 and A1-23):
𝑞𝑖
𝐹𝑜𝑟 𝐷 > 𝐷𝑙𝑖𝑚 : 𝑞 = (A1-18)
(1 + 𝐷𝑖 𝑡)
𝐹𝑜𝑟 𝐷 ≤ 𝐷𝑙𝑖𝑚 : 𝑞 = 𝑞𝑙𝑖𝑚 𝑒 −(𝐷𝑙𝑖𝑚(𝑡−𝑡𝑙𝑖𝑚)) (A1-16)

𝑞 𝑞
𝐹𝑜𝑟 𝐷 > 𝐷𝑙𝑖𝑚 : 𝑄𝑝 = 𝐷𝑖 𝑙𝑛( 𝑞𝑖 ) (A1-19)
𝑖
𝑞𝑖 𝑞𝑖 𝑞𝑙𝑖𝑚 − 𝑞
𝐹𝑜𝑟 𝐷 ≤ 𝐷𝑙𝑖𝑚 : 𝑄𝑝 = 𝑙𝑛( )+ (A1-23)
𝐷𝑖 𝑞𝑙𝑖𝑚 𝐷𝑙𝑖𝑚
Where:
𝑞𝑙𝑖𝑚 = Instantaneous Rate at 𝑡 = 𝑡𝑙𝑖𝑚 [Volume / Unit Time]
𝑡𝑙𝑖𝑚 = Time when 𝐷 = 𝐷𝑙𝑖𝑚 [Fraction / Unit Time]
𝐷𝑙𝑖𝑚 = Limiting Nominal Decline / Exponential Decline [Fraction / Unit Time]
A1.2.1.7 Super-Harmonic Decline (b>1)
Hyperbolic curve fits with a b-factor greater than 1 usually imply production is being influenced
by long-term transient behavior. Common reasons for 𝑏 > 1 are as follows:
• The interpretation is wrong and a value where 𝑏 < 1 will fit the data
• The reservoir is still in transient flow (tight reservoir, shale reservoir)
• The reservoir is layered
• Fractured well
As an example, 𝑏 = 2 corresponds to transient linear flow and is commonly found when

analysing hydraulically fractured, unconventional reservoirs. Caution must be exercised with
these cases and decline factor limits should be built into the forecast to capture the eventual
transition from transient to boundary dominated (exponential) flow.
Super harmonic decline is calculated using the hyperbolic equation (Equations A1-5 and A1-13).
𝑞 = 𝑞𝑖 (1 + 𝑏𝐷𝑖 𝑡)−1⁄𝑏 (A1-5)
𝑞𝑖𝑏
𝑄𝑝 = (𝑞1−𝑏 − 𝑞1−𝑏 ) (A1-13)
(1 − 𝑏)𝐷𝑖 𝑖
Where:

The modifications presented in Section A1.2.1.4 must also be applied to the super harmonic
production forecasts.
A1.3 Decline Curve Analysis Practice

DCA is most reliable for wells producing at high drawdown with a relatively constant flowing
pressures, so that the production rate decline follows the decline in reservoir pressure. Because
of this link to reservoir pressure, application of DCA techniques was historically restricted to the
boundary dominated flow period. However, development of tight and unconventional reservoirs
has extended its usage to the transient flow period, necessitating the development of alternative
techniques that attempt to match the transient and boundary-dominated flow. The decline curve
analysis presented by Arps is predominantly applicable to boundary dominated flow (depletion
period), whereas SEPD, for example, focuses on the early period of production (transient flow)
and as a result is specifically applicable to tight/shale reservoirs.
When performing DCA, always consider the type of producing mechanism and use the
appropriate analysis. In addition to gas, oil and liquid rates, take note of changes in pressure,
water cut and GOR, as appropriate. Be aware of changes to skin (positive or negative) and take
downtime into account.
A1.3.1 Identifying Flow Regimes

To use the Arps DCA techniques, the existence of a boundary dominated flow (BDF) period needs
to be confirmed. If BDF cannot be identified in a tight/shale well, then the SEDM could be used.
It should be noted that pseudo-steady state flow and steady state flow are not the same as BDF.
A1.3.1.1 Transient, Pseudo Steady State, Steady State, Boundary Flow
Pressure transient analysis is based on the assumption that the well flows at a constant rate
where decline curve analysis assumes that the well flows at a constant pressure.
Several terms are used when describing flow regimes in a well (Figure A1-5):
• Transient Flow — Pressure migrates outward from the well without
encountering any boundaries; no pressure depletion occurs
• Linear / Bilinear Flow – A special form of transient flow that occurs in
tight/shale or horizontal hydraulically fractured wells when well pressure
is maintained constant; no pressure depletion occurs
• Steady State Flow — Pressure has reached all of the boundaries but the
static pressure at the boundary does not decline, termed a “constant
pressure boundary”, which can occur for waterfloods or infinite aquifers;
no pressure depletion occurs
• Pseudo-Steady State Flow — Pressure has reached all of the boundaries
and the static pressure is declining at the boundary and correspondingly
at the well since production rate is held constant; pressure depletes
• Boundary-Dominated Flow — Pressure has reached all the boundaries
and the static pressure is declining at the boundary, but not at the well
since flowing pressure is maintained constant and rate declines; pressure
depletion occurs
Figure A1-5: Representation of Flow Regimes
Pseudo steady state is normally associated with well testing or wells operating under a rate
constraint. The well flow rate is maintained constant and the flowing pressure declines.
Boundary dominated flow is normally associated with long term production and decline curve
analysis where a well is constrained by surface pressures (separators, pipelines, etc). In this case,
the flowing pressure is approximately constant, and the rate declines.
In reality, although well head pressure may be constant, the bottom hole pressure may not be
constant since tubing pressure drop will change as flow rate declines and flowing fluid properties
change. For gas wells, this is not significant for lower flowing pressures but fluid properties
(water-oil ratio, GOR) will impact pressure drop in naturally flowing oil wells.
When pressure is moderately changing, the concept of “Material Balance Time” (𝑡𝑚𝑏 =
𝑐𝑢𝑚/𝑟𝑎𝑡𝑒) was developed to provide a correction of the actual production to an equivalent
constant pressure production. Plotting production data using 𝑡𝑚𝑏 allows solutions with both
declining rates and pressures to appear similar to the equivalent constant pressure solution,
similar to the superposition time function in pressure transient analysis but applied to boundary
dominated flow in DCA.
A1.3.1.2 Boundary Dominated Flow

Although the Arps equations can handle a number of flow regimes, it was developed to handle
boundary dominated flow (BDF). BDF is identified by plotting the data as log(𝑞) 𝑣𝑠 log(𝑡) and
confirming that a section with a straight line having a slope of 1 (unit slope) exists (Figure A1-6).
To assist in flow regime identification, lines with ¼ slope and ½ slope can be used to identify
bilinear and linear flow periods, respectively.
Figure A1-6: Example: Identification of Boundary Dominated Flow
Boundary dominated flow (unit slope) was evident in the Eagle Ford example after approximately
150 days (0.41 years) of production.
A1.3.2 Determining Arps DCA Model and Arps DCA Parameters
A1.3.2.1 Determining Arps DCA Model

Production data can be plotted in different ways to identify a representative decline model. If a
plot of log(𝑞) 𝑣𝑠 𝑡 and a plot of 𝑞 𝑣𝑠 𝑄𝑝 (cum production) exhibit a straight line (Figure A1-7),
then the decline trend is exponential and the exponential decline model should be used.
Figure A1-7: Identification of Exponential Decline

If a plot of log(𝑞) 𝑣𝑠 log (𝑡) and a plot of log (𝑞) 𝑣𝑠 𝑄𝑝 exhibit a straight line (Figure A1-8), then
the decline trend is harmonic and the harmonic decline model should be used.
Figure A1-8: Identification of Harmonic Decline
If no straight line is observed on these plots, the hyperbolic, super harmonic or the SEDM decline
model may be applicable.
A1.3.2.2 Determining Arps DCA Parameters
All decline curve analysis models are based on the concept of the instantaneous nominal decline
rate (𝐷), as presented as Equation A1-1 and the corresponding Equation A1-3.
1 𝑑𝑞
𝐷= − (A1-1)
𝑞 𝑑𝑡
𝐷𝑖
𝐷= (A1-3)
Equation A1-3 can be rearranged into Equation A1-24 which is the equation of a straight line
having a slope of 𝑏 and an intercept of 1/𝐷𝑖 . Figure A1-9 is a plot of Equation 3.2-1.
1 1
= 𝑏𝑡 + (A1-24)
𝐷 𝐷𝑖
Figure A1-9: Plot of Equation A1-24

Figure A1-10 presents the plot generated using Eagle Ford example data. The slope (𝑏) of the
plot is 0.91 and the intercept (1/𝐷𝑖 ) is 0.22 for an initial annual nominal decline (𝐷𝑖 ) of 4.
Figure A1-10: Example
Figure A1-11 presents the resulting Arps hyperbolic analysis. The blue decline curve in Figure A1-
11 was generated assuming an initial rate (𝑞𝑖 ) of 295 bopd.
Figure A1-11: Example DCA
1
If the slope on the 𝐷 𝑣𝑠 𝑡 plot is equal to or greater than 1, then analysis using the SEDM could
be used rather than using Arps harmonic or super harmonic DCA (refer to Section A1.2.1.7).
A1.3.3 Multi-Segment DCA
The multi-segment DCA method uses DCA model segments, and complete analysis can be any
combination of decline curve analysis models. For instance, the multi-segment DCA could be used
to capture distinct flow regimes, including transient flow (b > 1, SEDM), harmonic decline (b = 1),
hyperbolic decline (0 < b < 1), and exponential decline (b = 0). Figures A1-12 and A1-13 present a
conceptual three-segment DCA.
The multi-segment method is well-suited for unconventional reservoirs that exhibit multiple flow
regimes and could be used as an alternative to the Stretched Exponential decline model.
`
Figure A1-12: Conceptual Multi-Segment DCA
Figure A1-13: Multi-Segment DCA Forecast
A1.3.4 General Considerations for Arps b-factor

As a rule of thumb, a general guide for the ranges of b-factor for specific reservoir production
characteristics are as follows:
• Undersaturated Oil Reservoir
o Exponential: b=0
• Solution Gas Drive / Gas Cap Reservoirs
o Increasing GOR: b = 0.0 to 0.1
o Poor Mobility Ratio: b = 0.1 to 0.3
o Moderate GOR: b = 0.3 to 0.6
o High API: b = 0.6 to 0.8
• Waterflood
o Constant Pressure: b >= 0.8
• Gas Reservoir
o High Pressure: b=0
o Low-to-Mod Pressure: b = 0.3 to 0.5
o Increasing Pwf: b trends towards 0.0
• Unconventional
o Shale: b > 1 (use SEDM)
When determining the decline period, use the most representative period in history, preferably
BDF, that would also be representative the future. Calculate the decline trend during the
representative period and determine the starting point (rate and time) for the forecast. Finally,
consider the constraints under which the forecast is to be generated.
A1.3.5 Water Drive / WaterFlood Considerations

Typically, water drive oil reservoirs (either natural drive or waterflood) can be difficult to analyze
by decline analysis techniques because drawdown increases as water breakthrough occurs due
to total fluid mobility declines. Also, the total reservoir fluid production rate may be increased
near and after breakthrough in order to maintain required oil production. Finally, pressure
balancing can cause drainage pattern changes and variable deliverability.
Most successful waterfloods are implemented in good continuity reservoirs with moderate to
high permeabilities and, as a result, well interference may occur. Because of well interference
effects, individual well decline analysis should be used with caution. It may be better to use an
aggregate analysis of a complementary group of interfering wells.
Field cases as well as analytical / simulation modelling generally supports hyperbolic/harmonic

decline for late stage waterflood behavior, in other words the value of b normally is between 0.8
and 1.0. This does not mean that exponential decline or super harmonic decline cannot occur in
waterflood reservoirs, but whenever it is observed, it is common that non-reservoir factors are
influencing the performance.
The waterflood decline analysis period should meet the following criteria:
• Undersaturated oil reservoir, no free gas
• Water Cut should be greater than 50%
• Voidage replacement ratio should be close to one
• Well count should be relatively constant
• Injection and total fluid production rates should be relatively constant
• The reservoir pressure should be relatively constant
• Producing well pressures should be constant
• The GOR should be relatively constant
• The volume of water injected should be greater than 25% of the hydrocarbon
pore volume
The main DCA plots for water drive and waterflood reservoirs are as follows:
• Log(Oil Rate) vs Cum Oil
• Log(Oil Cut) vs Cum Oil
• Log(WOR) vs Cum Oil
• Log(Cum Liquid) vs Cum Oil
A1.3.6 Reserves
A major use of decline curve analysis is to generate an estimate of reserves. Even for assets where
history matched simulation models are available, a cross check with DCA is normally made to give
increased confidence in the numbers.
Financial institutions tend to accept DCA estimates over other, more technical methods. The
ultimate recovery numbers become more important than the profiles. Application of constraints
in the production system, operating costs, capital costs and well behavior all need to be taken
into account to generate reliable reserve estimates.
There are many, equally valid ways to use DCA to determine low/most likely/high values, the
following technique is suggested:
• If the hyperbolic constant, b, is between 0.3 and 0.7, use it as most likely
• For the low case, use a 0.5b value with a min = 0.0
• For the high case, use a 1.5b value with a max = 0.9
Forecasts generated based on decline analysis (whether production profiles or reserves) should
be fundamentally grounded in a good understanding of the factors that control production. No
‘One size fits all‘ principal applies when it comes to application of DCA, specifically, using an
exponential decline for water drive, solution gas drive and gravity drainage systems is neither
technically, nor empirically, justified.
Oil reservoirs producing with high water cut or high GOR need to be analyzed using ratio plots -
Log(WOR) vs Cum Oil, Log(GOR) vs Cum Oil, Watercut vs Cum Oil - in addition to conventional
plots to ensure there is no over-estimation of volumes based on rate plots alone. Care needs to
be taken to understand the minimum criteria for application of these plots. For example
Log(WOR) vs Cum Oil should only be used if WOR is equal to or higher than 1 (water cut is equal
to, or higher than, 50%).
A1.3.7 Non-Ideal Behaviour Which May Require Special Consideration

General items to look out for:
• Reconciled/allocated data
• Invented data
• Splits – especially water
• Changing skin factors / stimulation
• Anything not yet observed – water/gas breakthrough, coning, cusping, etc
• Piston-like displacement – use Buckley Leverett or simulation
• Constant Pwf, be aware of changes in choke settings, lift gas, etc
• Transient behaviour
• Downtime: Unscheduled and scheduled downtime will distort decline trends
It is recommended to filter out downtime data before fitting a decline curve through the
production data, and then applying a ‘downtime factor’ to the resulting profile. You can also use
recorded uptime or producing hours to adjust the production data prior to analysis. Be aware of
transient effects after prolonged downtime - these effects should be filtered out.
A1.3.8 Noisy Data
Available historic data tends to be in a monthly or yearly format. Today, data may be available in
a daily format. Daily data can result in very noisy data (Figure A1-14). Note that the GW-01 well
is an example Granite Wash well.
Figure A1-14: Daily vs Monthly Data for GW-01 Well.
The noisy data may be resolved by using monthly data as shown in Figure A1-14. The issue with
this is that the number of data points available for analysis is significantly reduced unless there is
significant historical data.
As an alternative, applying a moving average to the daily data could be used to “smooth” the
data. Figure A1-15 presents 3-Day, 5-Day and 8-Day moving averages of the GW-01 data.
Figure A1-15: Smoothing Daily Data by Moving Average - GW-01 Well.
When smoothing using a moving average, the minimum averaging that yields acceptable data
should be applied. For the different averaging presented in Figure A1-15, the 5-Day moving
average appears sufficient to analyse the data.
A1.3.9 Use of De-Superposition
Full field profiles can be difficult to match for a number of reasons:

• Incremental development
• Different depletion processes in different parts of the field
• Wells or field segments coming on and off production
It is straightforward to split out various wells and/or field segments (de-superposition) and
perform decline analysis on each one separately. In most cases, a well-by-well analysis will give a
better result than a single full-field analysis.
The requirement for de-superposition is dependent on whether or not boundary dominated flow
has been achieved. For wells under boundary dominated flow, the well drainage volumes in a
bounded reservoir are proportional to the rates of withdrawal from each drainage volume. In
other words, q/Qp is a constant for each well and for the total reservoir. For boundary dominated
flow, analysis by summation of wells or analysis at an overall reservoir level should give identical
results. This assumes uniform reservoir properties – faults and large changes in permeability or
different production mechanisms may result in different q/Qp ratios for different groups of wells.
DCA Appendix 2 – Model Formulations

This appendix provides a summary of the time-rate relations used in the PE² Essentials eDCA tool,
as well as formulations for the various diagnostic functions (i.e., q(t), D(t), b(t), β(t), q/Gp(t) and
ζ(t)) which are incorporated into the tools.
A2.1 Arps Hyperbolic Model

Reference: Arps, J.J. 1945. Analysis of Decline Curves. Trans. AIME 160: 228-247.
The hyperbolic rate decline relation termed the Arps decline model is given as Equation A1.5:
qi
q(t ) = (A1.5)
1/b
(1 + bDi t )
Where: q(t) is the rate at time t, qi is the rate at time 0, b is the hyperbolic exponent, D i is the
initial nominal decline factor and t is cumulative production time.
If the Arps hyperbolic rate decline relation is used when long term transient flow regimes are
evident (i.e. b>>1.0), then extrapolation of the Arps hyperbolic relation will almost always lead
to significant overestimations of EUR and future performance.
Since the Arps hyperbolic relation can often model the early-time flow behavior, the industry has
adopted a protocol to "constrain" the ultimate extrapolation by including a terminal exponential
decline trend — hence, the "modified hyperbolic" designation. The modified hyperbolic model
(Appendix A2.2) is enabled in eDCA by entering a value for D lim in the Arps model (Figure A2-1).
Figure A2-1: PE² Essentials – eDCA: Arps Model
After the Dlim is reached the Arps equation reverts to the exponential decline equation (Equation
A1-6).
It should be noted that the decline factors are presented as annual values and as a fraction.
The defining equations for the Arps Model are as follows:

qi
q(t ) = (A2.1)
(1 + bDi t )1 / b
Di
D(t ) = (A2.2)
(1 + bDi t )
b(t ) = b (A2.3)
Di t
 (t ) = (A2.4)
(1 + bDi t )
q (t ) (1 − b) Di
= (A2.5)
Gp (t ) (1 + bDi t ) (1 / b) − (1 + bDi t )
A2.2 Modified Hyperbolic Model

Reference: Robertson, S. Generalized Hyperbolic Equation; USMS SPE 18731, 1988.
This model has two trends — an initial trend that is hyperbolic (i.e., for t<texp), and a final trend
that is exponential (i.e., for t> texp) — where texp is the time of change from hyperbolic to
exponential.
For the hyperbolic function, the modified hyperbolic model follows the Arps Hyperbolic model,
Equations A2.1 to A2.5). For times greater than texp, Di = Dlim (constant terminal decline) and the
model becomes exponential.
The defining equations for the Modified Hyperbolic Model (for t> texp) are as follows:
q(t ) = qi exp[ − Di t ] (A2.6)
D(t ) = Di (A2.7)
b(t ) = 0 (A2.8)
 (t ) = Di t (A2.9)
q (t ) exp[ − Di t ]
= Di
Gp (t ) (1 − exp[ − Di t ])
(A2.10)
A2.3 Stretched Exponential Model (SEPD)
Reference: Valkó, P.P. 2009. Assigning Value to Stimulation in the Barnett Shale: A Simultaneous
Analysis of 7000 Plus Production Histories and Well Completion Records. Paper SPE 119369
presented at the SPE Hydraulic Fracturing Technology Conference, College Station, TX, 19-21
January.
The stretched exponential rate decline relation presented by Valkó is given as Equation A2.11
(Figure A2.2):
q(t ) = qˆ i exp[ −(t /  ) n ] (A2.11)
Where: q(t) is the rate at time t, qi is the rate at time 0, τ and n are model terms and t is cumulative
production time.
Figure A2-2: PE² Essentials – eDCA: Stretched Exponential Decline Model
The basis for the stretched exponential decline model (SEDM) is elementary physics. The
discharge of fluid from a tank against a fixed back-pressure results in an exponential decline of
flow rate over time. By considering a gas reservoir as a collection of connected tanks (cells)
discharging against different back-pressures and with different resistances (‘time constants’), this
leads to the interpretation of stretched exponential decay. The SEPD model was generated by a
sum (integral) of pure exponential decays with a distribution of time constants in the form of
Gamma functions. As a result, it is not simple to implement this model.
An advantage of the SEPD model is that production under variable bottom hole pressures can be
modeled. Forecasts can also be generated assuming changing bottom hole pressures.
The defining equations for the Stretched Exponential Model are as follows:
q(t ) = qˆ i exp[ −(t /  ) n ] (A2.11)
D(t ) = n −n t n −1 (A2.12)
1 − n n −n
b(t ) =  t (A2.13)
n
 (t ) = n −n t n (A2.14)
  t n 
exp −   
q(t ) n    
= (A2.15)
G p (t )    n 
  1  −   1 ,  t   
  n   n    
  
A2.4 Power Law Exponential Model (PLE)
Reference: Ilk, D., Perego, A.D., Rushing, J.A., and Blasingame, T.A. 2008. Exponential vs.
Hyperbolic Decline in Tight Gas Sands — Understanding the Origin and Implications for Reserve
Estimates Using Arps' Decline Curves. Paper SPE 116731 presented at the SPE Annual Technical
Conference and Exhibition, Denver, CO, 21-24 September; and, Ilk, D., Rushing, J.A., and
Blasingame, T.A. 2009. Decline Curve Analysis for HP/HT Gas Wells: Theory and Applications.
Paper SPE 125031 presented at the SPE Annual Technical Conference and Exhibition, New
Orleans, LA, 04-07 October.
The Power Law Exponential decline relation is given as Equation A2.16 (Figure A2-3):
q (t ) = qˆ i exp[ − Dˆ i t n − D t ] (A2.16)
Where: q(t) is the rate at time t, qi is the rate at time 0, Di is the initial decline rate, D ∞ is the
decline rate at infinite time, n is power exponent and t is cumulative production time.
Figure A2-3: PE² Essentials – eDCA: Power Law Exponential Decline Model
The Power Law Exponential decline model (PLE) is analogous to the SEDM but was derived
independently and is simpler to use. To account for the non-convergence to zero of a pure power
equation, the D∞ term was incorporated to account for the boundary dominated flow regime.
The PLE model is very flexible and can be used to match transient, transition and boundary
dominated flow data.
The defining equations for the Power Law Model are as follows:
q (t ) = q ˆ t n − D t]
î exp[ − D (A2.16)
i 
D(t ) = D + Dˆ i n t n −1 (A2.17)
Dˆ i (1 − n) n t n
b(t ) = (A2.18)
( D t + Dˆ n t n ) 2
 i
 (t ) = D t + Dˆ i n t n (A2.19)
No q(t)/Gp(t) formulation is available for the Power Law Exponential relation since no closed form
relation exists for the cumulative production function, Gp(t). This is because of the complexity
introduced into Equation A2.16 by the D term.
A2.5 Duong Model
Reference:Duong, A.N. 2011. Rate-Decline Analysis for Fracture-Dominated Shale Reservoirs SPE
Reservoir Evaluation and Engineering 14 (3): 377-387.
The Duong decline relation is given as Equation A2.20 (Figure A2-4):
−m  a (1− m )  (A2.20)
q(t ) = qi t exp  [t − 1] + q
 (1 − m ) 
Where: q(t) is the rate at time t, qi is the rate at time 0, a and m are model terms, q∞ is the rate
at infinite time and t is cumulative production time.
The Duong decline model is based on an extended linear/bilinear flow regime. It is derived from
transient behavior of unconventional-fractured reservoirs. The model is extracted from the
straight-line behavior of q/Gp vs. time on a Log-Log plot.
Figure A2-4: PE² Essentials – eDCA: Duong Decline Model
For the Duong model, rate does not start out at a maximum, as for the other models, but quickly
reaches a maximum within days. If q∞ is negative then rate will theoretically become negative,
there may be a requirement to limit the rate function using a minimum rate or a maximum time
constraint before the onset of a negative rate.
The defining equations for the Duong Model are as follows:
 a 
q(t ) = q1 t − m exp  [t (1−m ) − 1] + q (A2.20)
 (1 − m) 
D(t ) = m t −1 − a t − m (A2.21)
m t m [t m − at ]
b(t ) = (A2.22)
[a t − m t m ]2
 (t ) = m − at (1−m) (A2.23)
q(t )
= a t −m (A2.24)
G p (t )
A2.6 Logistic Growth Model (LGM)
Reference: Clark, A.J., Lake, L.W., and Patzek, T.W. 2011. Production Forecasting with Logistic
Growth Models. Paper SPE 144790 presented at the SPE Annual Technical Conference and
Exhibition, Denver, CO, 30 October-02 November.
The Logistic Growth decline relation is given as Equation A2.25 (Figure A2-5):
( n −1)
K nat (A2.25)
q (t ) =
[a + t ]
n 2
Where: q(t) is the rate at time t, qi is the rate at time 0, K is the carrying capacity, n is the
hyperbolic exponent, a is a constant and t is cumulative production time.
Figure A2-5: PE² Essentials – eDCA: Logistic Growth Decline Model
The Logistic Growth decline model (LGM) was adopted from population growth models. It is a
modified form of hyperbolic logistic growth models. The LGM implies a growth equation which,
in this case, represents the growth of cumulative oil or gas production.
The defining equations for the Logistic Growth Decline Model are as follows:
K tn
Gp (t ) = (A2.25)
[a + t ]
n
dGp (t ) aK n t ( n −1)
q (t ) = = (A2.26)
[a + t ] 2
n
dt
a(1 − n) + (1 + n) t n
D(t ) = (A2.27)
t (a + t n )
a 2 (1 − n LGM ) − 2a(n 2 − 1)t n + (n + 1)t 2 n

b(t ) = (A2.28)
[a (1 − n) + (n + 1)t n ]2
a (1 − n) + (1 + n) t n
 (t ) = (A2.29)
(a + t n )
q(t ) an
= (A2.30)
Gp (t ) t (a + t n )
A2.7 LeBlanc-Okouma Power Law Model (LOPM)
Reference: LeBlanc, D., Okouma, V. 2018. New Rate-Decline model for unconventional reservoirs,
World Oil, March 2018.
The LeBlanc-Okouma Power Law Exponential decline relation is given as Equation A2.31 (Figure
A2-6):
q(t ) = qi t − exp( − t ) (A2.31)
Where: q(t) is the rate at time t, qi is the rate at time 0, α is the power exponent, ξ is term to
account for boundary dominated flow and t is cumulative production time.
Figure A2-6: PE² Essentials – eDCA: LeBlanc-Okouma Power Law Decline Model
The LeBlanc-Okouma Power Law decline model (LOPL) is an extension of the PEE Empirical model,
presented as Equation A2.35, accounting for the boundary dominated flow regime by
incorporating the e(-ξt) term. The LOPL is analogous to the SEDM and PLE models but was derived
independently.
If ξ=0, the LOPL is equivalent to the PEE Empirical model. In this case, the model will overpredict
production unless a Dlim is used, which is incorporated into the DCA tool.
The LOPL decline model is easily applied and with the inclusion of a late time term, ξt, in the
power-law equation it is able to model transient, transition, and boundary-dominated flow. The
LOPL Model has been shown to yield more consistent estimates of short-to-long term production
in unconventional reservoirs, compared to other time-rate models, while still maintaining the
capability of modeling conventional reservoirs.
The defining equations for the LeBlanc-Okouma Power Law Decline Model are as follows:
q(t ) = qi t − e t (A2.31)
 + t
D(t ) = (A2.32)
t

b(t ) =
( + t )2
(A2.33)
 (t ) =  + t (A2.34)
𝑞(𝑡) = 𝑞𝑖 𝑡 −𝛼 (A2.35)
No q(t)/Gp(t) formulation is available for the LeBlanc-Okouma Power Law model since no closed
form relation exists for the cumulative production function, Gp(t). This is because of the
complexity introduced into equation A2.31 by the ξ term.
A2.8 Normalized DCA

During the early stages of production, the well may be produced at a choked rate (constant rate,
variable pressure). For this situation, the flowing pressure (𝑝𝑤𝑓 ) continues to decline. After a
minimum pressure is reached, the pressure will be maintained constant and the rate will decline.
It is possible to analyze the constant rate, declining pressure flow period using a technique called
normalized decline curve analysis. Conventional decline curve analysis (DCA) is performed on the
constant pressure, declining rate flow period.
Reference: Anderson, S., Anderson, D., Edwards, K., Epp, K., Stalgorova, K., Pressure Normalized
Decline Curve Analysis for Rate-Controlled Wells, SPE 162923, 2012.
Normalized decline analysis is based on the following definition of pressure normalized rate (𝑞𝑁 )
in terms of initial pressure (𝑝𝑖) and flowing pressure (𝑝𝑤𝑓 ):
𝑞
𝑞𝑁 = (A2-32)
𝑝𝑖 − 𝑝𝑤𝑓
And, in terms of pseudo pressure, 𝛹:

𝑞
𝑞𝑁 = (A2-33)
𝛹𝑖 − 𝛹𝑤𝑓
Many high-pressure wells are rate restricted during early production due to operating facility and
pipeline constraints. These high-pressure wells may display a harmonic decline of normalized rate
with time which results in a linear relationship between normalized rate and cumulative
production when plotted on a semi-log scale (Figure A2-7).
Figure A2-7: Normalized Rate vs Cumulative Production
The equation of the straight line in Figure A2-7 is given as Equation A2-34.
log(𝑞𝑁 ) = 𝑚1 𝑄𝑝 + C1 (A2-34)
Where:
𝑄𝑝 = Cumulative Volume
𝑞𝑁 = Normalized Flow Rate [Rate / Pressure]
𝑚1 = Slope of the 𝑞𝑁 𝑣𝑠 𝑄𝑝 plot [1 / Time-Pressure]
𝐶1 = Intercept (𝑄𝑝 = 0) of the 𝑞𝑁 𝑣𝑠 𝑄𝑝 plot [Rate / Pressure]
Extrapolating the straight line to an abandonment 𝑞𝑁 / 𝛥𝑃 value will yield the EUR for the well.
Using Equation A2-34, a rate can be estimated for any 𝑄𝑝 if 𝛥𝑃 is known. In order to forecast
rate, the flowing pressure (𝑝𝑤𝑓 ) as a function of time must also be known.
Future flowing pressure is dependent on future operating conditions. The general assumption for
rate-restricted wells is that pressure will decline logarithmically with time until a minimum value
is reached (Figure A2-8).
Figure A2-8: Flowing Pressure vs Time
The equation of the declining straight line in Figure A2-8, is presented as Equation 2.3-4.
log(𝑝𝑤𝑓 ) = m2 𝑡 + C2 (A2-35)
Where:
𝑝𝑤𝑓 = Flowing Pressure – CHP, THP or BHP
𝑡 = Time in units consistent with forecast
𝑚2 = Slope of the 𝑝𝑤𝑓 𝑣𝑠 𝑡 plot [Unit Pressure / Unit Time]
𝐶2 = Intercept (𝑡= 0) of the 𝑝𝑤𝑓 𝑣𝑠 𝑡 plot [Unit Pressure]
Caution must be used when attempting to forecast pressure in oil wells that have varying water
and gas rates.
The procedure to generate a production forecast (Figure A2-9) using Normalized DCA, is to use
the following equations and start by estimating a rate 𝑞𝑒𝑠𝑡 and an incremental time 𝛥𝑡.
𝑄𝑝 (𝑡) = 𝑄𝑝 (𝑡 − 1) + 𝑞𝑒𝑠𝑡 (𝑡)𝛥𝑡

𝑞𝑁 (𝑡) = 10^[𝑚1 𝑄𝑝 (t) + C1]
𝑝𝑤𝑓 (t) = 10^[m2 𝑡 + C2]
𝑞(𝑡) = 𝑞𝑁 (𝑡)[(𝑝𝑖 − 𝑝𝑤𝑓 (𝑡))]
Iterations are then performed until the estimated rate 𝑞𝑒𝑠𝑡 (𝑡) and the calculated rate 𝑞(𝑡) are
within an acceptable difference for each time step.
Figure A2-9: Normalized Decline Curve Forecast
DCA Appendix 3 – Miscellaneous Definitions

For rate-time decline analysis, the governing equations are based on the loss ratio concept
(Equation A3.1) introduced by Arps. The base definitions are as follows:
378 Monte Carlo DC Production Forecast
Definition of the loss ratio:

1 q(t )
− (A3.1)
D(t ) dq(t ) / dt
Definition of decline parameter:

1 dq(t )
D(t )  − (A3.2)
q(t ) dt
Derivative of the loss ratio:

d  1  d  q(t ) 
b(t )    −  (A3.3)
dt  D(t )  dt  dq(t ) / dt 
Beta derivative – relates rate and derivative function:
1 dq(t )
 (t )  t  t D(t ) (A3.4)
q(t ) dt
ζ (Zeta)-Derivative — time derivative of q(t)/GP(t) ), inverse material balance time:
d  q (t ) 
 (t )    (A3.5)
dt  G P (t ) 
Definition of reciprocal material balance time:
q(t )
(A3.6)
G p (t )
Equation A3.2 is the definition of the decline parameter, D(t); Equation A3.3 is the derivative of
the loss-ratio, b(t); Equation A3.4 is the Beta function which relates rate and a derivative function;
and Equation A3.5 is the Zeta function which is the derivative of q/Gp or q/Np, Equation A3.6.
All these functions can be plotted in the PE² Essentials PDA tool.
Monte Carlo DC Production Forecast

The PE² Essentials ‘Monte Carlo DC Forecast’ tool (Figure MCD-1) was developed as an extension
to the PE² Essentials Decline Curve Analysis tool to include probabilistic DCA.
Figure MCD-1: PE² Essentials Monte Carlo DC Forecast Tool
This Monte Carlo simulation model can be used to generate probabilistic EUR values based on a
probabilistic simulation of decline curve parameters. The model evaluates all the parameters that
go into the decline curve equations and will produce equivalent production forecasts for specific
realizations.
The tool can directly import information from a separate DCA database in order to generate a
probabilistic production profile for a well that has been previously analysed (Refer to Section
MCD.2). If the DCA database includes GOR/WGR/WOR/WCut analysis information, then these
parameters will also be forecasted.
MCD.1 Monte Carlo DCA

There are two options for running a Monte Carlo simulation: starting from initial conditions (t=0)
or from the end of production history. To run the forecast from the end of production history,
the is imported from a separate DCA database file.
The ‘Probabilistic Parameters’ are entered into the model (Figure MCD-2). The P50 numbers can
be based on the results of the initial analysis performed using the PE² Essentials Decline Curve
Analysis tool.
Figure MCD-2: Monte Carlo DC Forecast Tool – Probabilistic Parameters
As is the case for all PE² Essentials Monte Carlo simulation models, it is possible to generate a
deterministic forecast by making P90=P50=P10. This essentially disables the Monte Carlo
simulator.
Following Monte Carlo simulation of the EUR, the equivalent P90/P50/P10 and EV deterministic
production forecasts are generated based on the decline curve realizations (Figure MCD-3 and
MCD-4).
Figure MCD-3: Monte Carlo DC Forecast Tool – Probabilistic Results

Figure MCD-4: Monte Carlo DC Forecast Tool – Probabilistic Results
A 'Seed' value is included so the same probabilistic result can be re-generated. By entering a
specific seed value, the same random Gaussian distribution will be used for the simulation.
Entering '-1' for the seed will generate random seed numbers so the results will be different for
each simulation run.
The default number of Monte Carlo simulations is 10,000. This gives a relatively smooth
probabilistic distribution curve. For slow computers, 1,000 simulations may be acceptable.
During the simulation, a deterministic model is extracted for the P1, P10, P50, P90, P99 and EV
realizations (Figure MCD-5). These equivalent models can be used to represent the probabilistic
models in deterministic realizations. It should be noted that these values are just one realization
for the deterministic values - different values will be generated for each run if Seed=-1.
Figure MCD-5: Monte Carlo DC Forecast Tool – Deterministic Results

'Save Pxx Results to PE Tools dB' will save the EUR results for the P1 to P99 results as well as the
deterministic decline curve parameters for P1/P10/P50/P90/P99 and EV realizations (Figure
MCD-5). These parameters can then be imported into the PE² Essentials Decline Curve Analysis
tool to generate forecasts (refer to Section MCD.2).
It is also possible to save the history+forecast and the incremental forecast (when run from the
end of the history) to the PE Tools database for use in other PE² Essentials tools.
MCD.2 Monte Carlo DC Forecast Example – Marcellus

This is a continuation of the example presented in Section DCA.9 of the DCA document. The
Parameters from the DCA database file were imported into the Monte Carlo DC Forecast tool
(‘Load DCA dbase’) and P90 and P10 values were entered (Figure MCD-6). The DCA database file
is ‘PE_Essentials_DCA_DataBase_Marcellus.dvx’, located in the “Book Examples\Example
Marcellus Example\DCA” directory.
Importing the DCA database file data will also load history.
Figure MCD-6: Monte Carlo DC Forecast Tool – Marcellus Analysis
A constant ‘Seed’ was entered so the forecast could be re-generated and ‘Run Monte Carlo DC
Simulation’ yielded the deterministic and probabilistic results shown in Figure MCD-7.
Figure MCD-7: Monte Carlo DC Forecast Tool – Deterministic Results
The results were saved using 'Save Pxx Results to PE Tools dB' and imported into the PE² Essentials
Decline Curve Analysis tool. The different deterministic realizations were used to generate
forecasts with the DCA tool which were then plotted with PE² Essentials Chart (Figures MCD-8
and MCD-9) and used for economic analysis.
Figure MCD-8: Deterministic Values Imported into DCA Tool
Figure MCD-9: PE² Essentials Chart – Marcellus Forecast Results

384 PE IOR/EOR/Heavy Oil Essentials
PE IOR/EOR/Heavy Oil Essentials

The EOR/IOR/Heavy Oil Essentials section contains the following programs:
• IOR/EOR Screening Tool

• IOR/EOR/Heavy Oil Tools
IOR / EOR Screening Tool

The PE² Essentials ‘IOR/EOR Screening Tool’ performs a screening for IOR and EOR processes
based on reservoir properties (Figure IOR-1).
Figure IOR-1: PE² Essentials IOR / EOR Screening Tool
All the screening parameters used in this tool were obtained from published SPE papers, which
are too numerous to list here.
IOR.1 Screening Parameters

There are a number of sources that include screening parameters for IOR and EOR processes.
Unfortunately, there are variations in the published parameters. Figure IOR-2 presents the
criteria incorporated into the PE² Essentials IOR/EOR Screening Tool to ensure a consistent
approach.
Figure IOR-2(a): IOR/EOR Screening Parameters – Oilfield
Figure IOR-2(b): IOR/EOR Screening Parameters - Metric
The PE² Essentials IOR/EOR Screening Tool parameters are accessed through the ‘IOR/EOR
Screening Criteria’ (Figure IOR-3).
Figure IOR-3: IOR/EOR Screening Criteria

386 IOR / EOR Screening Tool
IOR.2 IOR / EOR Screening

In order to determine the applicability of a specific process, the fluid properties (Figure IOR-4)
and reservoir parameters have to be entered (Figure IOR-5).
Figure IOR-4: IOR/EOR Screening Tool – Fluid Parameters
Figure IOR-5: IOR/EOR Screening Tool – Reservoir Parameters
Not all parameters have to be entered. For example, if a miscible flood is not contemplated, then
the ‘Min Miscibility Pressure’ does not have to be entered.
After input of the appropriate parameters, the results of the screening are presented (Figure IOR-
6) along with applicability of the process (Yes/No) and the ‘Failed Criteria’ for each process.
Figure IOR-6: IOR/EOR Screening Tool – Results
To determine why a process failed, select the process on the screening guide (Figure IOR-7).
Figure IOR-7: IOR/EOR Screening Tool – Criteria
The ‘Failed Criteria’ numbers are referenced by the number on the screening guide.
388 EOR / Heavy Oil Tool
EOR / Heavy Oil Tool

A number of general-purpose tools are included in the PE² Essentials ‘EOR/Heavy Oil Tools’ as
shown in Figure EOR-1.
Figure EOR-1: PE² Essentials EOR / Heavy Oil Tool
The EOR / Heavy Oil Tools are comprised of the following:

▪ Minimum Miscibility Pressure
▪ Reservoir Thermal Properties
▪ Hot/Cold Water Injection
▪ SteamFlood Residual Oil Saturation
▪ PCP Pump Rates
▪ Sucker Rod Pump Rates
The only input data required is the input for the specific tool that is being used.
EOR.1 Minimum Miscibility Pressure

The ‘Minimum Miscibility Pressure’ tool (Figure EOR-2) calculates MMP for gas injection, nitrogen
injection and carbon dioxide injection.
Figure EOR-2: IOR/EOR/Heavy Oil Tools – Minimum Miscibility Pressure
There are a number of references available for minimum miscibility pressure (MMP) correlations.
The following is a list of the correlations and references, along with their range of validity, that
are used in the Minimum Miscibility Pressure model.
Glasø Correlation (1985)
Glasø, O.S.: Generalized minimum miscibility pressure correlation, Society of Petroleum

Engineers journal, vol. 25, no. 6, pp. 927–934, December 1985. Valid for:
- Oil molecular weight of C2 through C6 from 34 to 54 g/mol
- For N₂ injection, methane concentrations in the oil from 30% to 60%
- For N₂ injection, molecular Weight of C7+ in oil less than 200 g/mol
- Gas injection MMP from 1100 to 7000 psi
- CO₂ injection MMP from 900 to 4400 psi
For hydrocarbon gas injection, the following equations apply.
MMP34 = 6329 - 25.41y - (46.745 - 0.185y)z + [1.127y5.258exp(319.8zy-1.703)] (10-12) T (EOR-1)

MMP44 = 5503 – 19.238y - (80.913 – 0.273y)z + [1.7y3.73exp(13.567zy-1.058)] (10-9) T (EOR-2)
MMP54 = 7437 – 25.703y - (73.515 – 0.214y)z + [4.92y5.52exp(21.706zy-1.109)] (10-14) T (EOR-3)
x = MW of C2 to C6 in injected gas
y = (2.622 C7SG0.846)6.588
z = Mol%C1 in injected gas
MMPx, MMP for MW of x, is calculated as an interpolation between MMP34, MMP44 and MMP54
Where: MMP is minimum miscibility pressure in psi and T is temperature in ˚F.
For N2 gas injection the following equations apply.
If molecular weight of C7+ , MWC7+, in the oil is greater than or equal to 160:
MMP = 6364 - 12.09MWC7+ + [1.127MWC7+5.258 exp(23025MWC7+-1.703) - 20.8] (10-12) T (EOR-4)
If molecular weight of C7+ , MWC7+, in the oil is greater than or equal to 160:
MMP = 7695.1 - 12.09MWC7+ + [1.127MWC7+5.258 exp(23025MWC7+-1.703) – 39.77] (10-12) T
(EOR-5)
If the total mole% of C2 to C6 (x’) in the oil is less than 28%, the following equation applies:
MMP = 93640 - 12.09MWC7+ + [1.127MWC7+5.258 exp(23025MWC7+-1.703) – 20.8](10-12)T – 99.3x’
(EOR-6)
Where: MMP is minimum miscibility pressure in psi, MWC7+ is the molecular weight of C7+ in the
oil, x’ is the mole% of C2 to C6 in the oil and T is temperature in ˚F.
For CO2 gas injection the following equations apply.
If the total mole% of C2 to C6 (x’) in the oil is greater than or equal to 18%:
MMP = 810 – 3.404MWC7+ + [1.7MWC7+3.73 exp(786.8MWC7+-1.058)] (10-9) T (EOR-7)
If the total mole% of C2 to C6 (x’) in the oil is less than 18%:

MMP = 2947.9 – 3.404MWC7+ + [1.7MWC7+3.73 exp(786.8MWC7+-1.058)] (10-9) T – 99.3x’ (EOR-8)
oil, x’ is the mole% of C2 to C6 in the oil and T is temperature in ˚F.
Maklavani Correlation (2010)
Maklavani, A.M.; Vatani, A.; Moradi, B.; Tangsirifard J.: New Minimum Miscibility Pressure
(MMP) Correlation for Hydrocarbon Miscible Injections, Brazilian Journal of Petroleum and Gas,
vol. 4, no. 1, pp. 11-19, 2010. Valid for:
- Temperature from 130˚F to 300˚F
- Methane concentration 6% to 55%
- Sum of C2-C6 concentrations from 1% to 63%
- Molecular Weight of C7+ from 120 to 302 g/mol
- Injection gas C2+ concentrations from 0% to 48%
- Injection gas C2+ molecular weight from 0 to 72 g/mol
For hydrocarbon gas injection, the following equation applies.
MMP = 145.04 (43.664 - 4.542α + 0.689α2 - 0.132β) (EOR-9)

α = (XC2-C61.72785 XC10.1)/(T0.5 MWC7+)
β = YC2+(1.064+0.00686*MWC2+)
oil, XC2-C6 is the mole% of C2 to C6, CO2 and H2S in the oil, XC1 is the mole% of C1 in the oil, , YC2+ is
the molecular weight of C2+ in the injected gas, MWC2+ is the molecular weight of C2+ in the
injected gas and T is temperature in ˚F.
Firoozabadi Correlation (1986)
Firoozabadi, A.; Aziz, K.: Analysis and Correlation of Nitrogen and Lean-Gas Miscibility Pressure,
SPE Reservoir Engineering, pp. 575–582, November 1986. Valid for:
- Methane concentrations in the oil from 30% to 60%
- Molecular Weight of C7+ in oil is less than 200 g/mol
For N2 injection the following equation applies.
MMP = 9433 - 188000XC2-C5/(MWC7+T0.25) + 1430000[XC2-C5/(MWC7+T0.25)]2 (EOR-10)
oil, XC2-C5 is the mole% of C2 to C5 in the oil and T is temperature in ˚F.
Sebastion Correlation (1992)
Sebastion, H.M.; Lawrence, D.D.: Nitrogen Minimum Miscibility Pressures, SPE/DOE 24134,
April 1992. Valid for:
- Methane concentration in the oil up to 60%

- Sum of C2-C6 concentrations up to 30%
For N2 injection the following equation applies.
MMP = 4603 – 3283x + 4.776y – 4.008z + 2.05MWC7+ + 7.541(T + 460) (EOR-11)

x = XC1(T + 460)/MWC7+
y = XC12(T + 460)2/MWC7+
z = XC2-C6(T + 460)2/MWC7+
oil, XC1 is the mole% of C1 in the oil, XC2-C6 is the mole% of C2 to C6 and CO2 in the oil and T is
temperature in ˚F.
Zhang Correlation (2015)
Zhang, H.; Hou, D.; Kai, L.: An Improved CO₂-Crude Oil Minimum Miscibility Pressure Correlation,
Journal of Chemistry, 175940, vol. 2015, October 2015, Valid for:
- Molecular Weight of C7+ from 130 to 402.7 g/mol
- MMP up to 10,510 psi
For CO2 injection, the following equation applies.
MMP = 145.04 {a[ln(T)]b [ln(MWC7+)]c (1 + XC1N2/XC2-C6)d} (EOR-12)

a=0.000083397
b=3.9774
c=3.3179
d=0.17461
oil, XC1N2 is the mole% of C1 and N2 in the oil, XC2-C6 is the mole% of C2 to C6, CO2 and H2S in the oil
and T is temperature in ˚F.
Chen Correlation (2013)
Chen, B.L.; Huang, H.D.; Zhang, Y. et al.: An Improved Predicting Model for Minimum Miscibility
Pressure (MMP) of CO₂ and Crude Oil, Journal of Oil and Gas Technology, vol. 35, no. 2, pp. 126–
130, 2013. Valid for:
- MMP up to 4085 psi
MMP = 3.534 [(T - 32)0.8293 MWC7+0.5382 XC1N20.1018 XC2-C6-0.2316] (EOR-13)
oil, XC1N2 is the mole% of C1 and N2 in the oil, XC2-C6 is the mole% of C2 to C6 in the oil and T is
temperature in ˚F.
Yuan Correlation (1977)
Yuan, H.; Johns, R.T.; Egwuenu, A.M.; Dindoruk, B.: Improved MMP Correlations for CO₂ Floods
Using Analytical Gasflooding Theory, SPE 89359, 2005[8] Cronquist, C.: Carbon Dioxide Dynamic
Miscibility With Light Reservoir Oils, 4th Annual U.S. DOE Symposium, August 1977. Valid for:
- MMP up to 10,510 psi
MMP = 145.04 [(a + bMWC7+ + cXC2-C6 + (d + eMWC7+ + fXC2-C6/MWC7+2) T + (g + hMWC7+ +

iMWC7+2 + j XC2-C6) T2] (EOR-14)
a=-9.8912
b=0.045588
c=-0.31012
d=0.014748
e=8.0441x10-4
f=56.303
g=-8.4516x10-4
h=8.8825x10-6
i=-2.7684x10-8
j=-6.3830x10-6
oil, XC2-C6 is the mole% of C2 to C6, CO2 and H2S in the oil and T is temperature in ˚F.
Cronquist Correlation (1977)
[8] Cronquist, C.: Carbon Dioxide Dynamic Miscibility With Light Reservoir Oils, 4th Annual U.S.
DOE Symposium, August 1977. Valid for:
- Oil Gravity from 23.7˚ to 44.8˚ API

- MMP from 1070 to 5000 psi
MMP = 15.993 TX (EOR-15)

x = 0.744206 + 0.0011038MWC5+ + 0.0015279XC1N2
oil, XC1N2 is the mole% of C1 and N2 in the oil and T is temperature in ˚F.
It should be noted that all of the correlations were empirically generated. For more information
on the calculations associated with each correlation, the appropriate reference should be
reviewed. If using a correlation outside of its published limits, caution should be used when
accepting the results.
As a rule of thumb, for CO₂ miscibility, Zhang et al appears to be the optimum overall correlation
and for hydrocarbon gas miscibility, Maklavani et al appears to be the optimum correlation.
There is no option to export the MMP results to a file.
EOR.2 Reservoir Thermal Properties

The ‘Reservoir Thermal Properties’ tool (Figure EOR-3) generates thermal properties that can be
used in calculations and simulators.
Figure EOR-3: EOR / Heavy Oil Tools – Reservoir Thermal Properties

The equations for specific heat properties are as follows.
OilDen = 141.5 / (API + 131.5) (EOR-16)

OilDenResTemp = 1047OilDen / (1047 + TC - 20) (EOR-17)
WaterDenResTemp = .001 / (0.001 + 1.436x10-6(-4.8872 + 0.134186TC + 0.00212868TC2)) (EOR-18)
SpecificHeatOil = (1.6848 + 0.00391TC) / OilDenResTemp0.5 (EOR-19)
SpecificHeatWater = (4.3245 - 0.003696TC + 0.00002428TC2) (EOR-20)
SpecificHeatRock = 0.01381Shale% + 0.0083(100 – Shale%) (EOR-21)
Where: TC is the reservoir temperature in ˚C, Specific Heat is in kJ/kg-˚K.
The equations for heat capacity and thermal conductivity are as follows.
HeatCapRes= 1000 φ OilDenResTemp SpecificHeatOil (1 - Sw)

+ 1000 φ WaterDenResTemp SpecificHeatWater Sw (EOR-22)
+ 1000 (1 - φ) SpecificHeatRock ρma
ThermalCondRes = 1.7307[Cond50 - 0.00128(T - 125) (Cond50 - 0.82)] (EOR-23)

Cond50 = 0.735 - 1.3*φ + 0.363Sw0.5 (4.45 – 0.0445Shale% + 0.0165Shale%) (EOR-24)
Where: HeatCapRes is the heat capacity of the reservoir in kJ/m3-˚K, ThermalCondRes is the thermal
conductivity of the reservoir in W/m-˚K and T is the reservoir temperature in ˚F.
EOR.3 Hot/Cold Water Injection

The ‘Hot/Cold Water Injection’ tool (Figure EOR-4) generates a thermal profile of the reservoir.
Figure EOR-4: EOR / Heavy Oil Tools – Hot/Cold Water Injection

The reservoir parameters and water injection parameters are entered in the ‘Hot/Cold Water
Injection’ section. The reservoir heat capacity is estimated using the ‘Reservoir Thermal
Properties’ tool (Section EOR.2). The reservoir temperature profile is then plotted as a function
of distance from the wellbore (Figure EOR-5).
Figure EOR-5: EOR / Heavy Oil Tools – Temperature Profile
The technique to determine temperature in a reservoir at any distance from the wellbore was
presented by Lauwerier (Lauwerier, H., A., The Transport of Heat in an Oil Layer Caused by the
Injection of Hot Fluid, Applied Scientific Research, Volume 5, Issue 2, 1955), as reproduced below:
TempDist = ResTemp + (InjTemp - ResTemp)Comperfc (EOR-25)
Comperfc = (1 + 0.278393LauVar + 0.230389LauVar2 + 0.000972LauVar3 + 0.078108LauVar4)-4

LauVar = DimDist / [2 (DimTime - DimDist)0.5]
Dim_Dist = 3448.6 ThermalCondShale Dist2 / (h HeatCapRes WaterDenInjTemp SpecificHeatInjWater)
ThermalCondShale= 43.752 / (Tc + 273.15)0.55
WaterDenInjTemp = .001 / (0.001 + 1.436x10-6(-4.8872 + 0.134186Tinj + 0.00212868Tinj2))
SpecificHeatInjWater = (4.3245 - 0.003696Tinj + 0.00002428Tinj2)
DimTime = 1.098x106 t ThermalCondShale / (h HeatCapRes)2
Where: TempDist is temperature at a distance from the wellbore, ResTemp is reservoir

temperature, InjTemp is injection water temperature, T c is reservoir temperature in ˚C, Tinj is
injection water temperature in ˚C, h is reservoir thickness in m, Dist is distance from the wellbore
in m, HeatCapRes is the heat capacity of the reservoir in kJ/m3-˚K, ThermalCondShale is the thermal
conductivity of the shale in W/m-˚K and t is time in days.
The reservoir temperature profiles can be saved to a ‘csv’ file for plotting with PE² Essentials
Chart (Figure EOR-6 and EOR-7) through the ‘Export Distribution Data’ button.
Figure EOR-6: EOR / Heavy Oil Tools – Heating Temperature Profile Over Time
Figure EOR-7: EOR / Heavy Oil Tools – Cooling Temperature Profile Over Time
EOR.4 SteamFlood Residual Oil

The residual oil saturation following a steam flood can be estimated based on oil viscosity and
steamflood temperature using the ‘SteamFlood Residual Oil Saturation’ tool (Figure EOR-8).
Figure EOR-8: IOR/EOR/Heavy Oil Tools – Residual oil Saturation
The Sor correlation is based on results presented by Bursell and Pittman (Bursell, C., G. and
Pittman, G., M., Performance of Steam Displacement in the Kern River Field, Journal of Petroleum
Technology, SPE5017, August 1975). The correlation used in this tool is based on data presented
as Figure 12 in the Bursell and Pittman paper.
The tool will generate the Sor for the entered steam temperature and oil viscosity. It also presents
a table of Sor for different steam temperatures.
The correlation is presented as Equation 6-26.
Sor = 0.2736 + 1.832x10-6µ - 5.617x10-4Ts (EOR-26)
Where: Sor is the post steam flood residual oil saturation in decimal, µ is oil viscosity and Ts is
steam temperature in ˚F.
EOR.5 PCP and Sucker Rod Pump Rates

The rates generated by a PCP pump or a Sucker Rod pump can be calculated using the PCP and
Sucker Rod Pump Rates tool shown in Figure EOR-9. This is not a pump design tool, but a tool to
calculate the rates being generated at the operating conditions of the pump.
Figure EOR-9: IOR/EOR/Heavy Oil Tools – Pump Rates
For a PCP pump, the equation is as follows.
RatePCP = 5.937x10-3 (Eccen) (Rots) (Dia) (Pitch) (Efficiency) (EOR-27)
Where: RatePCP is the pump rate in bbls/d and all other terms are as input.
For a sucker rod pump, the equation is as follows.
RateSRP = 0.01781 (Speed) (Area) (Stroke) (Efficiency) (EOR-28)
Where: RateSRP is the pump rate in bbls/d and all other terms are as input.
400 PE Field Development Essentials
PE Field Development Essentials

The Field Development Essentials section contains the following program:
• Field Development Planning
• System Nodal Analysis
Field Development Planning Tool

The PE² Essentials ‘Field Development Planning’ tool can be used to generate project based
forecasts using a type curve as the basis for the production forecast (Figure FDP-1).
Figure FDP-1: Field Development Planning Tool
The type curve is generated by normalization of rate-cum data. The rate-cum data can be entered
from an Excel spreadsheet or from forecast files generated by other PE² Essentials tools. If there
is more than one well or pad in the development scenario, variations in each well’s productivity
can be generated through the incorporation of 'Statistical Parameters'.
If well flowing pressure data is required, the reservoir, wellbore and PVT parameters need to be
entered. Most of the parameters required to calculate pressure can be entered by importing a
PE² Essentials THP-BHP file.
FDP.1 Well Type Curves

Three options are available for importing well type curve data: from an Excel file, from a PE Tools
database forecast or from a separate DCA database. Figures FDP-2 and FDP-3 show an example
of importing data from Excel.
Figure FDP-2: Field Development Planning Tool – Well Type Curve Input and Generation
Figure FDP-3: Field Development Planning Tool – Excel Data

402 Field Development Planning Tool
Figure FDP-4 shows an example of importing the well type curve data from a PE² Essentials DCA
database. For this option, the wells in the database are loaded and a well is selected for use in
the Development Planning tool.
Figure FDP-4: Field Development Planning Tool – Excel Data
To generate the well type curves, the imported data is normalized to the maximum rate (Norm
Rate) and the volume initially in place (Norm Cum). For DCA data, the ‘Norm Cum’ calculation is
based on the EUR calculated from the DCA analysis results. Type curves for WGR/WOR and for
CGR/GOR are generated from water and oil/condensate production data and referenced to the
cumulative hydrocarbon production (Norm Cum).
It is possible to set up, or modify, a water profile by entering a value into the ‘Cum@WBT’ box
which represents the cumulative production at water breakthrough, checking the ‘Enter WGR
Parameter’ (or ‘Enter WOR Parameter’) and entering the Norm Cum value at a WGR/WOR of 10.
A constant CGR/GOR can be modeled by checking the ‘Model Constant CGR’ (or ‘Model Constant
GOR’) and entering the constant value into the ‘Base CGR’ (’Base GOR’) box.
Figure FDP-5 shows the results of these procedures.

Figure FDP-5: Field Development Planning Tool – Modifying Parameters
The normalized well type curve is the basic building block for the forecasting routines in the PE²
Essentials Development Planning tool. To incorporate variations in each well’s production
characteristic, a statistical option is available that uses a Gaussian distribution to vary the initial
maximum rate and the final EUR/GIIP/OOIP value for each well. Refer to Section FDP.6 for
information on use of the ‘Statistical Parameters’.
FDP.2 Well Drilling and Completion Planning

Pad drilling of unconventional wells may reduce the cost of drilling wells. The downside of pad
drilling is that once the well is drilled the time to the start of its production is delayed until the
entire pad is ready to be placed on production. The PE² Essentials Development Planning tool
includes an option to model pad drilling (Figure FDP-6).
Figure FDP-6: Field Development Planning Tool – Well Planning

The pad drilling option is implemented by checking the ‘Pad Drilling: #Wells/Pad’ option and
entering the number of wells to be drilled on each pad. If this option is not checked, individual
sequential well drilling will be incorporated into the profile.
The time to actually drill and complete all wells on the pad will determine when a production
increase will occur. For economic modelling purposes, the timing of each well on the pad is
included in the forecast so that the capital spending can be modeled properly in the Economics
tools.
FDP.3 Rig Schedule Planning and Drilling Learning Rate

If more than one well or pad is included in the project, a rig schedule has to be included in order
to drill the wells. The PE² Essentials Development Planning tool allows up to 5 rigs to be included
in the development. The time that each rig is available is specified (Figure FDP-7).
Figure FDP-7: Field Development Planning Tool – Rig/Well Planning
It should be noted that if more than one well is available at the same time, one of the wells will
be delayed. This is part of the rig scheduling routines. Also, it is assumed that all rigs can drill all
wells so when a rig is otherwise idle, it will be used to drill other pad wells.
A drilling learning curve is included in the PE² Essentials Field Development Planning tool to model
the fact that doubling of drilling experience will reduce the required time to drill a well by a
constant percentage. The learning curve model incorporated in the Field Development Planning
tool was developed by T. P. Wright in 1936 and is referred to as the Cumulative Average Model
or Wright's Model. Wright’s model can be expressed as Equation FDP.1.
Y = aXb (FDP.1)
Where: Y is the cumulative average time to drill a well, X is the cumulative number of wells, a is
the time to drill the first well and b is the Learning Slope (Equation FDP.2).
Learning Slope = Log(learning) / Log(2) (FDP.2)

The value for ‘learning’ used in the tool is defined as (100-Learning Rate)/100. So, for example,
entering a Learning Rate of 10% will yield a Learning Slope of -0.152. For an initial pad drilling
time of 480 days, the learning curve is shown in Figure FDP-8.
Figure FDP-8: Field Development Planning Tool – Learning Curve
Caution should be used when including a learning rate for individual wells. As well count
increases, drilling time may become unrealistically low.
It should be noted that Learning Rates typically range between 0% and 30%. Learning Rates above
30% are rare according to published references.
FDP.4 Facility Planning and Uptime Modelling

Facility constraints, in terms of maximum and minimum capacities, are entered and used for rate
cut-back and project shut-in criteria (Figure FDP-9). Minimum well rates, which are used to shut
in a well, are specified in the ‘Production Parameters’, Section FDP.5.
Figure FDP-9: Field Development Planning Tool – Facilities Modeling

Water production limits are not included in the PE² Essentials Development Planning tool. It is
assumed that all produced water will be handled by the facilities. When modeling conventional
or offshore projects this may be an issue and should be taken into account in the forecast.
Incorporating water limits is not straightforward because water is not a primary component of
the production (like oil or gas) but is a function of the cumulative production from a specific well.
This means that determining the optimum well to cut back to reduce the water rate while
maintaining maximum gas production will be a non-linear problem. If necessary, this type of
algorithm could be incorporated into a future version of the tool.
Uptime of the facilities can be modeled as a constant parameter or can be made to vary using
‘Statistical Parameters’ (Section FDP.6). The cost, in production, of the facilities downtime is
shown in the ‘SI Rate’ reported during the run (refer to Section FDP.7).
Modeling of the facility’s capacity limit enables the possibility of evaluating the economics of
facility debottlenecking by determining the value of the SI Rate and/or generating forecasts at
different values of facility capacity for economic comparisons.
FDP.5 Production Modeling

Production modeling can be straight forward based on the type curves or can include modeling
of the well’s bottom hole and tubing head pressures (Figure FDP-10).
Figure FDP-10: Field Development Planning Tool – Production Modeling
The values entered for ‘Minimum Well Gas Rate’ or ‘Minimum Well Oil Rate’ and ‘Maximum Well
WGR’ or ‘Maximum Well WOR’ are used to shut in the well. It should be noted that if minimum
rate is very low, the forecast may extend for a long time if the type curve has a long ‘tail’. The
absolute limit for the forecast is 30,000 days.
To model pressures, the ‘Forecast Pressure’ option is checked. The wellbore and reservoir
parameters can be imported from the PE Tools database by clicking on ‘Well Data’ and ‘Import
Well Model’. When selecting this option, the PVT parameters can also be imported from the well
model (Figure FDP-11). PVT parameters can also be imported from the PE Tools database by
clicking the ‘Import PVT Properties’ on the PVT screen (Figure FDP-12).
Figure FDP-11: Field Development Planning Tool – Importing Well and PVT Parameters
If the well and PVT parameters are not imported from the PE Tools database, they can be entered
manually (Figure FDP-12).
Figure FDP-12: Field Development Planning Tool –Well and PVT Parameters
The entered well and reservoir parameters can be varied using the ‘Statistical Parameters’ option
(section FDP.6). The parameters will be varied for every well in the model based on the entered
Gaussian distribution.
Including pressure modeling in the forecast enables the economic evaluation of reducing the
facility pressure or adding compression to the facilities. This is done by running forecasts at
different pressure limits.
Well flowing pressure is calculated based on two assumptions: a) the reservoir pressure can be
estimated using the calculated normalized rate-cum data; and b) the wellbore pressure drop can
be defined using the LIT equation.
FDP.5.1 Well Pressure Calculation

The PE² Essentials Development Planning tool is not a rigorous reservoir/production simulator.
As a result, the pressure calculation routines should be considered as estimates only.
To calculate the reservoir pressure at any time in the forecast, it is assumed that for an
unconstrained system (not choked), declining production rate is the result of declining reservoir
pressure. This allows the normalized rate-cum relationship to be converted to a pressure-
normalized cum relationship (Figure FDP-13).
Figure FDP-13: Field Development Planning Tool – Determination of Reservoir Pressure
Using the Reservoir Pressure – Norm Cum relationship, the reservoir pressure can be estimated
at any point in the forecast. With reservoir pressure, the Laminar-Inertial-Turbulent (LIT) flow
equation (gas) or the Vogel equation (oil) is then used to calculate the flowing bottomhole
pressure (FBHP).
For gas wells, the LIT equation is presented in Equation FDP.3 (ref: Tarek Ahmed, Reservoir
Engineering Handbook, Gulf Publishing, 2001; and L. Mattar, G. Brar, and M. Mumby; Theory and
Practice of the Testing of Gas Wells, Energy Resources Conservation Board,1978).
∆P2 = PR2 – FBHP2 = aqg + bqg2 (FDP.3)

a = 1424 µg Z TR (ln(0.472re/rw) + Skin) / (k h)
b = 0.128 GasG Z TR / (k1.333 rw h2)
re = (WellArea*43560)0.5
Where: PR is reservoir pressure at the current time in psi, q g is the gas flow rate in mscf/d, µg is
gas viscosity in cp, Z is the gas deviation factor, TR is the reservoir temperature in ˚R, re is the
outer radius of the reservoir, rw is the wellbore radius in ft (model assumes an 8 inch wellbore),
Skin is the wellbore skin factor, k is reservoir permeability in md, h is the net pay in ft, GasG is gas
specific gravity and WellArea is drainage area of the well in acres.
In order to check the validity of the input parameters, the absolute open flow potential (AOFP)
of the well, ∆P2 = PR2, is calculated (Equation FDP.4) and compared to the maximum rate.
AOFP = (-a + (a2 + 4b ∆P2)0.5) / 2b (FDP.4)
For saturated oil wells, Fetkovitch’s methodology for use of Vogel’s equation is presented in
Equation FDP.5 (ref: Tarek Ahmed, Reservoir Engineering Handbook, Gulf Publishing, 2001).
∆P2 = PR2 – FBHP2 = 2Pb qo / PI (FDP.5)

PI = 0.00708 k h / (µo Bo (ln(0.472re/rw) + Skin))
re = (WellArea*43560)0.5
AOFP = PI PR2/ 2Pb
Where: PR is reservoir pressure at the current time in psi, qo is the oil flow rate in bopd, Pb is the
bubble point pressure in psi, PI is the well productivity index in bopd/psi, µo is oil viscosity in cp,
Bo is the oil formation volume factor, re is the outer radius of the reservoir, rw is the wellbore
radius in ft (model assumes an 8-inch wellbore), Skin is the wellbore skin factor, k is reservoir
permeability in md, h is the net pay in ft and WellArea is drainage area of the well in acres.
The FBHP is converted to flowing tubing head pressure (THP) for reporting purposes. When using
the ‘Forecast Pressure’ option, the rate is controlled by FBHP. THP control is not implemented in
this version of the tool.
FDP.6 Statistical Modeling Using Gaussian Distributions

The PE² Essentials Development Planning tool includes the option to randomly vary some of the
forecast parameters (Figure FDP-14).
Figure FDP-14: Field Development Planning Tool – Statistical Variations
Gaussian distributions can be entered for ‘Productivity’, ‘Rig Availability’ and ‘Facility UpTime’
which will incorporate variations for these modeling parameters.
As the forecast proceeds, the Gaussian distribution is randomly sampled, and modifications are
applied to the appropriate parameter to apply variability to the forecast. It should be noted that
as for other PE² Essentials distributions, entering a ‘Seed’ value will generate the same variable
results for each run. Entering a ‘Seed’ value of -1 will ensure the distribution is sampled randomly.
FDP.6.1 Statistical Well Productivity

Well productivity can be varied in two ways. The well’s type curve parameters can be varied by
entering Gaussian distributions for the initial rate and/or the EUR/GIIP/OOIP values. These
parameters will vary the initial productivity of each well as well as the ultimate recovery for each
well (Figure FDP-15).
Figure FDP-15: Field Development Planning Tool – Productivity Variations
When modeling well pressures, the reservoir permeability and/or the reservoir pay can also be
varied. This will impact each well’s productivity in terms of the flowing pressure limitation for the
well.
The ‘Minimum FBHP’ (Figure FDP-10) will be used to limit the flowing pressure for the well. Once
the FBHP target is reached, the well will continue to produce at this pressure until the minimum
rate is reached, after which the well will be shut in. THP is calculated at each time step but is not
used to control well production.
The values entered for the ‘Type Curve Parameters’ are used to vary the de-normalization of the
original type curve entered into the model. This will ensure that no two wells have the same rate-
cum profile. The entered low-mid-high values can be skewed to ensure more wells of a certain
type (high or low) are modeled.
After a forecast is generated, it is possible list the actual statistical parameters used for each well
by clicking ‘View Well Parameters’. This generates a list of the well parameters (Figure FDP-16)
and includes an option to export the data to a CSV file.
Figure FDP-16: Field Development Planning Tool – Production Well Parameters
The random seed number is also presented and can be entered into the statistical setup (Figure
FDP-15) in order to generate the same well parameters for each run.
FDP.6.2 Statistical Rig Availability

When more than one rig is included, it is possible to randomly vary the start time of when
subsequent rigs will be available (Figure FDP-17).
Figure FDP-17: Field Development Planning Tool – Rig Availability Variations

The rig will have either an early or a late start. The values entered for early and late start are
entered as positive since the parameter is relative to the base value which is normally set to zero
(on time availability). The Gaussian distribution will be randomly sampled to determine whether
the rig will have an early or late start.
FDP.6.3 Statistical Facility Uptime

The facility uptime can be varied on an annual, semi-annual or quarterly basis (Figure FDP-18).
Figure FDP-18: Field Development Planning Tool – Facility Uptime Variations
It should be noted that all pressure calculations are performed based on the gross production
rate not the net production rate which results from the facility uptime. The actual cumulative
volume produced for the time interval is based on the net production rate. Gross production rate
can be determined by adding the reported production rate and the reported SI rate for
production that is not constrained by facility maximum rate limitations.
Whenever the facility uptime is entered as 100%, the gross production rate will be the same as
the net production rate.
Modeling facility uptime allows the determination of the economic value associated with facility
upgrades to enhance the facility operation. This could be performed by evaluating the economic
value of the SI Rate at different constant uptime values or comparing the SI Rates at a constant
uptime value with random uptime values.
FDP.7 Model Output

The PE² Essentials Development Planning tool output can be displayed in monthly or daily format
(Figure FDP-19).
Figure FDP-19: Field Development Planning Tool – Forecast
The data can be saved to a ‘csv’ file in a monthly or daily format. The daily format will include the
well-by-well data along with the total project forecast. If well pressures are modeled, they will
be saved for each well. If pads are modeled, the pad production will be stored in the monthly file
along with the total project forecast.
Note - In order to use the forecast results in the PE² Essentials economics tools, the ‘Output
Frequency’ option should be set to ‘Daily’. This will ensure all characteristics of the forecast are
captured in the economics analysis.
The Development Modeling tool includes a ‘Quick Plot’ option to evaluate the results (Figure FDP-
20).
The ‘Save Graph’ button will save the current graph to a ‘png’ file.
If the ‘Quick Plot’ window is left open during a subsequent run, the ‘Update’ button can be used
to re-draw the graph with the data from a new run.
Figure FDP-20: Field Development Planning Tool – Quick Plot
For a more complete look at the results, the data can be saved to a csv file by clicking ‘Export
Forecast to CSV’. Note that the daily file includes well forecasts as well and depending on the
number of wells in the project, the file may be too big to be imported into a spreadsheet.
System Nodal Analysis Tool

The PE² Essentials ‘Nodal Analysis’ tool (Figure NOD-1) calculates the pressure drop through a
system of nodes that makes up the total production system, including reservoir, wellbore,
pipelines, ESP, gas lift and compressors.
Figure NOD-1: PE² Essentials Nodal Analysis Tool
PE Essentials Nodal Analysis tool performs a node-based analysis which means it calculates
pressure drop at different nodes in the system starting at the input node (downhole or surface)
and continues to the output node. To improve accuracy, the entire system is subdivided into
small incremental nodes to limit the pressure change over the interval.
The Nodal Analysis tool is more in-depth then the PE2 Essentials ‘Basic THP-BHP Gas Well’ and
‘Basic THP-BHP Oil Well’ tools. It includes more well options, including wellbore trajectories,
artificial lift and injectors, and includes surface pipelines which can also include compressors.
The PE2 Essentials Nodal Analysis tool includes the following:

416 System Nodal Analysis Tool
• Modeling oil, gas and water production wells

• Modeling gas and water injection wells
• Modeling of artificial lift: ESP and gas lift
• Modeling annular flow
• Modeling surface networks, including compressor
• Modeling retrograde gas wells
• Generation of simulator flow tables
• Single pressure point calculations (correlation calibration)
• Full wellbore and surface pipeline deviation surveys
• Conversion of pressure table to BHP
• Matching of oil PVT properties
• Full incorporation of reservoir IPR performance
• Gas multi-rate IPR calculations (c and n)
• Wellbore tubing correlations include:
o Hagedorn & Brown
o Beggs & Brill
o Orkiszewski
o Mukherjee & Brill
o Gray
o Mechanistic Model
• Surface pipeline correlations include:
o Beggs & Brill
For more complete information on the tubing and pipeline correlations, refer to the Appendix.
There is a wellbore ‘Pressure Matching Factor’ included in the single point pressure calculation
option that may assist with matching of known pressure data. Entering a value for this factor will
modify the frictional gradient term in the pressure drop calculations. To change the hydrostatic
gradient, modify the oil properties. Note that the pressure matching factor will be used in all
wellbore pressure drop calculations but not in the surface pipeline calculations.
The wellbore results and the surface results can be exported to a CSV file. The total system results
can be saved to a flow performance table which can then be imported into a reservoir simulator
data file.
There are a number of general references describing fluid flow in wells and pipelines. The
definitive reference is by Kermit Brown and H.Dale Beggs; “The Technology of Artificial Lift
Methods, Volume 1: Inflow Performance, Multiphase Flow in Pipes, The Flowing Well”, PennWell
Books, June 1980. Another excellent reference is by Eissa Al-Safran and James Brill; “Applied
Multiphase Flow in Pipes and Flow Assurance, Oil and Gas Production” SPE , 2017.
There are a number of example nodal analysis models included in the ‘PEE Tools Database.PEEdb’
located in the ‘Example Input Files\PEE Tools Database’ directory. Figure NOD-2 presents the
models included in the database.
Figure NOD-2: PE² Essentials Nodal Analysis Tool – Examples
NOD.1 Node-Based Analysis

A petroleum production system consists of two parts - the reservoir and the piping system. The
reservoir is where the hydrocarbons originate, and the piping system is the means to transport
the reservoir fluids from the reservoir to the processing facility. Figure NOD-3 shows a schematic
of the production system with three main nodes: reservoir; well; and surface pipeline/facilities.
Figure NOD-3: Nodal System

In order to have flow through the system, a pressure difference is needed. Fluids will flow from
the high-pressure point (reservoir) to the low-pressure point (facilities). Fluid flow through the
system is governed by the conservation of energy (first law of thermodynamics) which yields a
differential equation for steady state flow. For oil wells, the significant terms impacting the total
pressure gradient are the hydrostatic gradient (∇PHH) and the friction gradient (∇Pf). For low
pressure gas wells a third term representing the acceleration or kinetic gradient ( ∇Pa), resulting
from the expanding gas, may also be significant.
∇P = ∇Pf + ∇PHH + ∇Pa
For surface pipelines, the friction gradient (∇Pf) is the predominant term.
For upward flow (uphill for pipelines), fluids must overcome the backpressure exerted by the
effective column of fluid acting against the direction of flow. They must also overcome friction
losses due to the interaction of the fluid with the pipe wall. For downward flow (downhill for
pipelines), friction effects act against the direction of flow, but the effect of the hydrostatic
column helps overcome some of the friction losses.
Hydrostatic pressure losses are a function of the density of the fluid in the pipe but frictional
losses also depend on the fluid properties and flowing conditions within the pipe.
There are a number of calculation methods used to account for hydrostatic and frictional fluid
losses under a variety of flow conditions. Research into multi-phase flow concentrates on
predicting flow pattern, liquid holdup and pressure drop for the different flow conditions. The
research has resulted in empirical correlation models (circa 1954+) and mechanistic models (circa
1980+) for determining flowing pressure drop.
Empirical models use data obtained from laboratory tests and liquid holdup, measured by use of
quick-closing ball valves. These models were developed by correlating dimensionless numbers.
Although fluids were treated as homogeneous mixtures, gas and liquid phases were permitted
to travel at different velocities, with slippage effects being accounted for through empirical liquid
holdup correlations. Accuracy of the empirical models tends to be restricted to the range of
tested conditions.
Mechanistic modeling emerged in the early 1980’s. These models include experimental data but
are built from general mathematical principles. The mechanistic approach is to determine the
prevailing flow pattern for a given flow condition, and then to mathematically model the flow
mechanisms causing the identified flow pattern. All dominant physical parameters and
mechanisms must be incorporated in a mechanistic model to achieve accurate and reliable
predictions. Accuracy of mechanistic models tends to be better than empirical models over a
wider range of flow conditions.
Understanding slip velocity, holdup and flow maps are fundamental to understanding multiphase
flow. Holdup is the phase fraction at a node and slip is the relative phase velocity at the node.
Figure NOD-4 presents an example of holdup in a pipeline with stratified flow.
Figure NOD-4: Holdup
In stratified flow, gas will flow in the upper portion of the pipe and the liquid will flow along the
bottom. The liquid holdup is a calculation of the fractional area of the pipe containing liquid and
the gas holdup is fraction of the pipe containing gas. So liquid holdup, H L, will be:
HL = AL / (AL + AG)
Slip velocity is calculated as the difference in the in-situ phase velocities at the node where:
vslip = vg - vl
Since gas is lighter, it will generally travel faster than the liquid and this must be taken into
account in the pressure drop calculations.
All models have their own unique flow pattern map (Figure NOD-5) that are based on the in-situ
superficial velocities, vs, of the phases.
Figure NOD-5: Typical Mechanistic Model Flow Map
The superficial velocities of the liquid and gas phases are defined as the volumetric flow rate for
the phase divided by the pipe cross sectional area. These velocities can be calculated based on
the holdup, where vsl = vl * HL and vsg = vg * (1 - HL).
Flow maps are used to determine the type of flow at the node: bubble, slug, mist, etc. Once the
flow pattern is known, the appropriate pressure drop model for that flow pattern is used to
calculate the pressure drop at the node.
Hydrostatic pressure loss is determined from the fluid properties at the pressure and
temperature conditions at the node. Since pressure and temperature is also being calculated at
the node, iteration techniques are required to calculate the results.
Frictional pressure drop also depends on the fluid properties at the pressure and temperature
conditions at the node which, in addition, impacts the friction factor at the node. Friction factors
are normally based on the Fanning Friction Factor (Figure NOD-6), which is a function of the
dimensionless Reynolds number, Re.
Figure NOD-6: Typical Mechanistic Model Flow Map
Reynolds number is defined as:
Where: ρ is the in-situ density, v is the velocity, D is the pipe diameter and µ is the in-situ viscosity.
Numerical solution of Figure NOD-6 has been generated by a number of authors and is summarized as
follows.
The friction factor, f, for a Reynolds numbers in the laminar flow regime (Re ≤ 2000) is found by:
In the turbulent flow regime (Re ≥ 4000), the friction factor can be obtained from the Chen equation,
where ε is the absolute roughness of the pipe:
For the transition from laminar to turbulent flow (2000 < Re < 4000) friction factor can be estimated as:
In terms of pressure drop, the single-phase friction pressure loss is given by:
Where: f is the friction factor, ρ is the in-situ density, v is the in-situ velocity, L is the pipe length, D is the
pipe diameter and gc is gravitational conversion factor.
The single-phase hydrostatic pressure loss is given by:
Where: ρ is the in-situ density, g is the acceleration due to gravity, L is the pipe length and θ is the angle
of the pipe with respect to the horizontal
The acceleration/kinetic pressure loss is normally negligible and is ignored in multiphase calculations.
For multiphase flow, the friction pressure loss is modified by adjusting the friction factor (f), the density
(ρ) and velocity (v) to account for the multiphase mixture properties at the node and is dependant on the
flow correlation being used.
For specific information on the available correlations refer to the Appendix.
NOD.2 System Parameters

There are a number of options that can be chosen from the ‘System Parameters’ tab (Figure NOD-
7).
The type of well is specified, as well as additional options to include in the nodal analysis: artificial
lift, annular flow, surface piping and in the case of a gas well, retrograde gas.
The type of run – ‘System Analysis’ or ‘Generate Flow Table’ – is also specified on this tab.
The wellbore deviation survey and temperature survey are entered on this tab. When entering
the wellbore deviation survey, it is possible to enter the measured depth and an angle and the
tool will calculate the equivalent TVD. This is useful when modeling a future well.
Figure NOD-7: Nodal Analysis – System Parameters
Note - the angle represents the angle from vertical, for example, a horizontal leg is at 90°.
Recommendation: In order to model pressure drop along a lateral, it is recommended that an

angle of 89° be entered for the lateral.
NOD.3 PVT Parameters

The PVT parameters and the reservoir conditions are entered on the ‘PVT’ tab (Figure NOD-8).
Figure NOD-8: Nodal Analysis – PVT Parameters
The PVT parameters for all the phases should be entered in the model. It is also possible to match
oil PVT properties (Figure NOD-9) to calibrate the oil PVT correlations to actual data.
Figure NOD-9: Nodal Analysis – Matching PVT Parameters

To match oil PVT properties, enter the values to be matched and click the ‘Store/Continue’
button. Entering zero for a specific parameter will reset the match back to the correlation values.
To view the PVT values, click the ‘View PVT Data’ (Figure NOD-10).
Figure NOD-10: Nodal Analysis – PVT Properties
To view a specific property, use the drop-down menus to choose the property to plot.
Clicking ‘Export PVT Data to CSV’ will save all the PVT data to a csv file.
It should be noted that although PVT data is generated, and presented, for pressures up to 15,000
psi (100,000 kPa), the validity of the correlations at the high pressures may be an issue and should
be used with caution.
NOD.4 Reservoir Parameters

To evaluate the total system performance, reservoir data is required and is entered on the
‘Reservoir’ tab (Figure NOD-11).
Figure NOD-11: Nodal Analysis – Reservoir Parameters
The required IPR parameters are dependent on the well option chosen on the ‘System
Parameters’ tab. For an oil well, either the Oil PI or the liquid PI is entered (Figure NOD-12).
Figure NOD-12: Nodal Analysis – Oil IPR Parameters

When ‘Liquid PI’ is chosen, the IPR curve is generated using the PI value and the value for water
cut entered for ‘Water Cut’ in the ‘Flow Parameters’ box. The IPR plot in Figure NOD-11 shows a
liquid PI calculation with 25% water cut.
For a gas well, the c and n values, in terms of P2 are entered (Figure NOD-13).
Figure NOD-13: Nodal Analysis – Gas IPR Parameters
It is possible to calculate c and n by clicking the ‘Calc c & n’ button (Figure NOD-14).
Figure NOD-14: Nodal Analysis – c and n Parameters
The c and n values can be calculated using multi-rate test data or a single stabilized flowing
pressure and rate.
For a water well, the water PI is entered (Figure NOD-15).
Figure NOD-15: Nodal Analysis – Water IPR Parameters

The ‘Flow Parameters’ are entered, depending on the well type (Figure NOD-16). For injection
wells, the flow parameters include maximum injection pressure.
Figure NOD-16: Flow Parameters: Oil Well; Gas Well; Water Well; Injection Well
It is possible to run the system analysis for different flowing parameters (Figure NOD-17).
This is not the same as running multiple scenarios as is done for the lift table generation. This
performs a system analysis at the indicated conditions. It is possible to generate a system analysis
for different reservoir pressures, water cuts/WGR’s, GOR’s/CGR’s and maximum injection BHP’s
depending on the well type chosen.
Figure NOD-17: Varying System Analysis Flow Parameters
To generate a run with a different parameter, enable the appropriate check box in the ‘Flow
Parameters’ box and enter the different values in the appropriate table. To perform an analysis
for different conditions, choose the appropriate value from the drop-down menu and re-run the
analysis.
NOD.5 Wellbore Parameters

The wellbore tubing correlation is selected on the ‘Wellbore’ tab (Figure NOD-18).
Figure NOD-18: Nodal Analysis – Wellbore Parameters
Available wellbore tubing correlations include the following:

o Hagedorn & Brown
o Beggs & Brill
o Orkiszewski
o Mukherjee & Brill
o Gray
o Mechanistic Model
The Gray correlation is specifically designed for gas. The Appendix describes the models.
The wellbore completion parameters are also entered on this tab.

NOD.6 Artificial Lift Parameters

There are two options available for artificial lift: gas lift and ESP (Figure NOD-19). Note that the
‘Inc Artificial Lift’ check box on the ‘System Parameters’ tab needs to be checked before the
Artificial Lift option is enabled.
Figure NOD-19: Nodal Analysis – Artificial Lift Parameters
Although several artificial lift parameters can be entered, only one parameter can be used during
a specific run of the system analysis.
Ensure that the gas lift valve depth, ‘GLV Depth’, (Figure NOD-20) and the location of the ESP,
‘ESP Depth’ (Figure NOD-20) are entered before attempting to run the analysis. In addition, for
gas lift the maximum amount of lift gas that can be dissolved into the oil needs to be entered and
the maximum allowed gas before the ESP will cut-off needs to be entered if they are different
from the defaults.
Figure NOD-20: Nodal Analysis – Artificial Lift Parameters
When setting up the ESP data, ’Pump Frequency’, the total ‘# of Head Points’ and the ‘ESP Depth’
are entered. After the table is populated, the pump data is saved to the model by clicking the
‘Save Pump Table’ button. If only one table is being entered, it is saved automatically. If the table
is not saved, it will not be available for the run.
It is possible to edit the pump table and gas lift table by selecting the table and modifying the
appropriate parameter. For an ESP table, the ‘Save Pump Table’ has to be clicked to update the
table in the model.
Note the pump table is saved to the model but is not updated in the database. To save the data
in the PE Tools database, click the ‘Save Model to PE Tools dB’ on the main menu.
To select a specific gas lift value or pump frequency for the system analysis run, the value is
selected from the drop-down menu.
For a system analysis, only one value of gas lift value or ESP table is used for each run based on
the value selected from the drop-down menu.
NOD.7 Surface Parameters

Along with a surface pipeline, the tool includes an option to include a compressor (Figure NOD-
21). Note that the ‘Inc Surface’ check box on the ‘System Parameters’ tab needs to be checked
before the surface and compressor options are enabled. The pipeline profile and specifics are
entered into the ‘Pipeline Parameters’ table.
Figure NOD-21: Nodal Analysis – Surface Parameters
If the well type on the ‘System Parameters’ tab is a gas well (producer or injector), then the option
to add a compressor will be available. Multiple compressor tables can be entered to allow for
different compressor powers. The compressor power is used as an identifier and is not part of
the system analysis results other than to specify the compressor table that is used. Make sure
‘Save Comp Table’ is clicked after entering the compressor data.
For surface analysis, only one compressor table can be used during a run and is based on the
value selected from the drop-down menu.
NOD.8 Wellhead Results

A wellhead system analysis (Figure NOD-22) is generated by clicking the ‘Wellhead Analysis’
button.
Figure NOD-22: Nodal Analysis – Wellhead Results
The wellhead system analysis proceeds by first determining the rates, based on the IPR, for
pressure step increments of (Reservoir Pressure)/100. Then for the chosen water cut and GOR,
or WGR and CGR for a gas well, the THP is determined based on the selected tubing correlation.
Following the wellhead analysis, five THP’s are chosen internally and tubing performance curves
are generated and plotted. This is done for a QC check of the calculations.
The generated ‘Surface Performance Characteristic’ plot data can be exported to a CSV file by
clicking ‘Export Wellhead Results to CSV’ on the main menu.
NOD.9 Surface Results

A surface system analysis (Figure NOD-23) is generated by clicking the ‘Surface Analysis’ button.
This cannot be generated for a production well until the wellhead analysis has been generated.
Figure NOD-23: Nodal Analysis – Surface Results
The surface system analysis proceeds by using the rates and THP’s generated by the wellhead
system analysis and then for the chosen water cut and GOR, or WGR and CGR for a gas well, the
pressure at the outlet of the pipeline is determined. The ‘Pipeline Performance Characteristic’
plot represents the combined well/pipeline system performance. Following the surface analysis,
five outlet pressures are internally chosen, and pipeline performance curves are generated for
QC purposes.
The generated ‘Pipeline Performance Characteristic’ plot data can be exported to a CSV file by
clicking ‘Export Surface Results to CSV’ on the main menu.
NOD.10 Lift Tables

Lift tables (Figure NOD-24) are generated from the ‘Lift Tables’ tab after selecting the ‘Generate
Flow Table’ option in the ‘Run Options’ box on the ‘System Parameters’ tab.
Figure NOD-24: Nodal Analysis – Lift Tables
Note that liquid rate is used for oil wells. This was done to ensure that the well rate at high water
cuts does not yield unrealistic flow rates. For example, if an oil rate of 4000 bopd was specified
and a water cut sensitivity of 99% was input, the total liquid rate for that point would be 400,000
blpd. Such a high rate would cause issues with the friction pressure drop calculation and the
result may be an infinite pressure drop. This is captured in the table as a ‘0’ value.
All system information needed to generate the lift tables is entered on the appropriate tabs. All
appropriate options are available for inclusion in the generation of the lift tables.
For example, to include ALQ (Artificial Lift Quantity) data in the lift tables, the ‘Inc Artificial Lift’
checkbox on the ‘System Parameters’ tab has to be checked and the artificial lift data entered on
the ‘Artificial Lift’ tab. A surface network can also be included in the generation of lift tables.
The lift tables are generated by clicking the ‘Generate Lift Tables’ button. After the tables have
been generated, they can be saved in a csv file by clicking the ‘Save Table to CSV’ button or saved
as a simulator VFP table by clicking ‘Export VFP Table’ on the main menu.
Prior to saving the lift tables for use in a simulator, the data can be modified by clicking the
‘Process Curves for Simulator Use’ button. This will process all the curves to remove the unstable
portion of the tubing curves (Figure NOD-25).
Figure NOD-25: Nodal Analysis – Lift Tables
VFP table processing step is necessary in order to use the VFP data in any simulator that requires
that the VFP data to be monotonically increasing.
It should be noted that the processing of the VFP data does not modify any zero values in the
table. Correcting for zero values is up to the user. They may be corrected by reducing rates or
increasing tubing size.
NOD.11 THP/BHP Gradient Matching

To calculate a single THP/BHP pressure and calibrate the pressure drop in the well, click the
‘THP/BHP/Gradient’ button (Figure NOD-26), This option only calculates the well pressures and
does not include any surface networks included in the model.
Figure NOD-26: Converting a Single THP/BHP Point to BHP/THP Point
Prior to running this option, all the well parameters must be entered, or loaded, into the tool,
including the artificial lift parameters if required.
This option enables the results from the different correlations to be compared to help choose an
appropriate correlation for the main run.
The wellbore ‘Pressure Matching Factor’ is included to match a known pressure point. Entering
a value for this factor will modify the hydrostatic pressure drop equation which may yield a closer
match to the known pressure. This factor will also be used in the main pressure drop calculations.
Caution should be used if the correction factor falls outside of the range of 0.8 < factor > 1.2. This
may indicate that the wellbore description has a problem or the fluid properties that are used to
calculate the hydrostatic gradient term are not correct.
Note that the pressure matching factor will be used in all wellbore calculations but is not used in
the surface pipeline calculations.
The gradient information can be saved to a CSV file using ‘Export Table’.
NOD.12 THP/BHP Pressure Conversion

To import a table of well pressure/rate data and convert it to THP/BHP, click the ‘Convert
Pressure Data’ button on the main menu. This will open the ‘Pressure Conversion’ page (Figure
NOD-27).
Figure NOD-27: Converting a Table of Well Pressure Data
The pressure/rate data can be imported from a well in the PE Tools database or an Excel
spreadsheet by clicking on the ‘Load Data’ button. Figure NOD-28 shows the import of data from
the database.
Figure NOD-28: Loading Well Pressure Data

Only database wells that contain the type of pressure requested in the ‘Pressure Data’ box will
be listed and available for importing. Once the well pressure data is imported (Figure NOD-29), a
correlation model is chosen, and conversion is performed by clicking the ‘Convert …” button. The
label on the button will confirm which conversion process will occur.
Figure NOD-29: Converting Well Pressure Data
After the pressure data has been converted, it is possible to save the data to a CSV file by clicking
the ‘Export Data to CSV File’.
The converted pressures can be saved to the well in the PE Tools database by clicking the ’Save
Pressure to PE Tools db’ button. This will update the well in the PE Tools database with the
pressure data.
Appendix Nodal Analysis
A1 Hagedorn and Brown Correlation

In 1965 Hagedorn and Brown developed an empirical multi-phase pressure drop model
(Hagedorn, A. R., Brown, K. E., “Experimental Study of Pressure Gradients Occurring During
Continuous Two-Phase Flow in Small Diameter Conduits”, Journal of Petroleum Technology,
April, 1965).
Modifications have been applied to the original Hagedorn and Brown (M-HB) correlation
significantly extended the validity of the calculations. The modifications include the assumption
that liquid holdup will be zero when the calculated value is less than the no-slip liquid holdup,
and the Griffith correlation applied to bubble flow (Griffith, P. and Wallis, G. B., “Two-Phase Slug
Flow”, Journal of Heat Transfer, August, 1961).
The basic pressure drop equation, disregarding the kinetic energy term, for the MHB technique
is as follows.
144 dP/dz = ρa + f Mt2 / (7.413x1010 D5 ρa) (A1-1)

ρa = HLρL + (1 - HL)ρg
Where: dP/dz is the pressure gradient in psi/ft, ρa is the average in-situ density in lbm/ft3, ρL is the
liquid in-situ density in lbm/ft3, ρg is the gas in-situ density in lbm/ft3, f is the friction factor, Mt is
the total mass flow rate in lbm/d, D is the pipe diameter in ft and HL is the liquid holdup.
Liquid holdup, HL, is defined as the fraction of an element of pipe which is occupied by liquid at
some instant as HL = Volume of liquid in a pipe element / volume of the pipe element. The value
for liquid holdup varies from zero for single phase gas flow to one for single phase liquid flow. Oil
holdup is determined as HO = 1 - HL.
Hagedorn and Brown generated a number of empirical correlations to determine H L. These

correlations are presented In Figure A1-1.
Figure A1-1: Hagedorn and Brown Liquid Holdup Correlations

Calculation of the liquid holdup is the key parameter required for calculating pressures for the
M-HB technique (as well as for all other analytical multiphase techniques). For M-HB, HL is
calculated using the charts in Figure A1-1 and the following dimensionless numbers.
NLv = 1.938 VsL (ρL/σ)0.25

Ngv = 1.938 Vsg (ρL/σ)0.25
ND = 120.872 D (ρL/σ)0.5
NL = 0.15726 µL (ρLσ3)-0.25
Where: NLv is the liquid velocity dimensionless number, VsL is the superficial liquid velocity in ft/sec,
Vsg is the superficial gas velocity in ft/sec, σ is the gas liquid interfacial tension in dynes/cm, Ngv is
the gas velocity dimensionless number, ND is the pipe diameter dimensionless number, D is the pipe
inside diameter in feet and µL is the liquid viscosity in cp.
The equation to solve the first chart in Figure A1-1 is as follows (Guo, Sun and Ghalambor).
CNL = 10a (A1-2)

a = -2.6985 + -0.551b2 + 0.54785b3 – 0.12195b4
b = log(NL) + 3
The equation for the second chart, which uses CNL from Equation A1-2, is as follows.
HL/ψ = -0.10307 + 0.61777c - 0.63295c2 + 0.29598c3 - 0.0401c4 (A1-3)

c = log(X) + 6
X = 0.7643 NLv P0.1 CNL / Ngv0.575 ND (A1-4)
Finally, the third chart is solved with Equation A1-5 for ψ, the secondary correction factor, and
liquid holdup, HL, is calculated as follows.
Ψ = 0.91163 – 4.82176Y + 1232.25Y2 – 22253.6Y3 + 116174.3Y4 (A1-5)

Y = Ngv NL0.38 / ND2.14
Ψ = 1 if Y < 0.01
HL = Ψ (HL/ψ)
The friction factor is calculated using Equations A1-6 and A1-7.
f = [1.14 - 2Log(δ/ID + 21.25/Re0.9)]-2 (A1-6)

Re = 0.022 Mt / (D µLHL µg(1-HL)) (A1-7)
Where: Re is the multiphase Reynolds Number, D is the pipe inside diameter in feet, µL is the
liquid viscosity in cp and µg is the gas viscosity in cp.
Whenever flow is in the bubble flow regime, the pressure drop equation is based on the Griffith
correlation as follows.
144 dP/dz = ρa + f ML2 / (7.413x1010 D5 HL2 ρL) (A1-8)
Where: dP/dz is the pressure gradient in psi/ft, ρa is the average in-situ density in lbm/ft3, ρL is
the liquid in-situ density in lbm/ft3, f is the friction factor, ML is the liquid mass flow rate in lbm/d,
D is the pipe diameter in ft and HL is the liquid holdup.
Liquid holdup and Reynolds Number for the Griffith correlation (bubble flow) is as follows.
HL = 1 – 0.5{1 + 1.25Vm – [(1 + 1.25Vm)2 – 5Vsg]0.5}

Vm = VsL + Vsg (A1-9)
Re = 0.022 ML / (D µL) (A1-10)
Bubble flow exists under the following condition.
Vsg / Vm < 1.017 - 0.2218 Vm2 / D
Since fluid properties are a function of pressure, which will be a function of the location of the
node, PE² Essentials Nodal Analysis uses a variable depth increment to calculate pressure drop.
Note that all the other PE Essentials Nodal Analysis correlations will be presented in a general
format. For more specific information, the supplied references should be reviewed.
A2 Beggs and Brill Correlation

The Beggs and Brill correlation was published in 1973 (Beggs H., Brill J.; "A Study of Two-Phase
Flow in Inclined Pipes", JPT (May 1973)). The advantage of this correlation was that it was tested
for all inclinations and is used for the surface pipeline calculations in the Nodal Analysis tool.
This empirical model was derived from experimental data for vertical, horizontal, inclined uphill
and downhill flow of gas-water mixtures. As a result, it is one of the few published correlations
capable of handling flow in all directions. It was developed measuring the flow of water and air
through 1" and 1-1/2" sections of acrylic pipe that could be inclined at different angles from the
horizontal.
A3 Orkiszewski Correlation
The Orkiszewski correlation was published in 1967 (Orkiszewski, J.; "Predicting Two Phase
Pressure Drop In Vertical Pipes”, Journal of Petroleum Technology, 19(6); SPE-1546-PA).
A4 Mukherjee and Brill Correlation

The Mukherjee and Brill correlation was published in 1983 (Mukherjee H., Brill J.P.; “Liquid
Holdup Correlations for Inclined Two-Phase Flow”, Journal of Petroleum Technology, Vol. 35, No.
5).
The Mukherjee-Brill model can be applied to wells with different inclination angles and it has
been widely applied for wet gas wells.
A5 Gray Correlation
The Gray correlation was developed by H.E. Gray in 1974 specifically for wet gas wells (Gray, H.
E; "Vertical Flow Correlation in Gas Wells", User manual for API 14B, Subsurface controlled safety
valve sizing computer program, API). Although this correlation was developed for vertical flow, it
has been modified for vertical and inclined pipe pressure drop calculations.
A5 Mechanistic Model Correlation

The Petalas and Aziz mechanistic model (Petalas N., and Aziz, K., "A Mechanistic Model for
Multiphase Flow in Pipes", Journal of Canadian Petroleum Technology, June 2000, Volume 39,
No. 6) is valid for all pipe inclinations, geometries, and fluid properties.
The model was not built for a specific set of data or fluid properties. Instead, first principles were
applied to the possible flow patterns that are observed at different inclinations. For this reason,
it should be applicable to all pipe inclination and fluid properties. The model is a refinement of a
previous study by the authors (1996) where subsets of a database of over 20,000 laboratory
measurements and data from approximately 1,800 wells were used.
The following Figure NOD.A5-1 is reproduced from the referenced paper and shows the flow
chart to determine a flow regime in the mechanistic model. It is a very in-depth process and
results in a significant amount of computations and as a result is the slowest correlation in the
Nodal Analysis tool.
Figure NOD.A5-1: Petalas and Aziz Mechanistic Model Flow Regime Determination
444 PE Recovery Factor Essentials
PE Recovery Factor Essentials

Recovery Factor Essentials includes the ‘Recovery Factor Analysis’ tool (Figure 4-1) that is
comprised of six models for determining recovery factor:
• Deterministic / Probabilistic Oil Recovery Factor

• Deterministic Gas Recovery Factor
• Unconventional Oil/Gas Recovery Factor
• Reservoir Complexity Index Oil Recovery Factor
• Empirical North Sea Recovery Factor (not included in tool)
• Artificial Neural Network Oil Model
Recovery Factor Analysis Tool

Recovery Factor Essentials includes the ‘Recovery Factor Analysis’ tool (Figure RFA-1)
Figure RFA-1: PE² Essentials - Recovery Factor Analysis Tool

The tool that is comprised of six models for determining recovery factor:
• Deterministic / Probabilistic Oil Recovery Factor (Section RFA.1)

• Deterministic Gas Recovery Factor (Section RFA.2)
• Unconventional Oil/Gas Recovery Factor (Section RFA.3)
• Reservoir Complexity Index Oil Recovery Factor (Section RFA.4)
• Empirical North Sea Recovery Factor (not included in tool) (Section RFA.5)
• Artificial Neural Network Oil Model (Section RFA.6)
It should be noted that estimating recovery factors is not an exact science. It is based on empirical
correlations and results should not be accepted as a final value. The purpose of the Recovery
Factor Analysis Tool is to generate recovery factors using different techniques. It is up to the user
to determine which value (if any) is most representative of the reservoir of interest.
RFA.1 Deterministic/Probabilistic Oil RF

Entering values in the 'Deterministic / Probabilistic API Oil Recovery Factor' section (Figure RFA-
2) will yield deterministic calculations of the oil recovery factors for solution gas drive and water
drive reservoirs based on the 'API' equations. It is also possible to generate a probabilistic
recovery factor profile using Monte Carlo simulation and the API oil RF equations.
Figure RFA-2: Deterministic / Probabilistic Recovery Factor

446 Recovery Factor Analysis Tool
RFA.1.1 Deterministic Oil Recovery Factor

The deterministic recovery factors are based on the API recovery factor equations as follows.
SGD RF = 41.815 (φ(1 - Sw)/Bob)0.1611 (0.001k/µob)0.0979 Sw0.3722 (Pbp/Pab)0.1744 (RFA-1)

WD RF = 54.898 (φ(1 - Sw)/Boi)0.0422 (0.001kµwi/µoi)0.077 Sw-0.1903 (Pi/Pab)-0.2159 (RFA-2)
Where: SGD RF is the solution gas drive recovery factor in %, WD RF is the water drive recovery
factor in %, Bob is oil Bo at the bubble point, Boi is the oil Bo at the initial pressure, k is permeability
in md, µob is oil viscosity at the bubble point, µoi is oil viscosity at initial pressure, µwi is water
viscosity at initial pressure, Pbp is the bubble point pressure, Pab is the abandonment pressure, Pi
is the initial pressure, Sw is the initial water saturation in decimal and φ is the porosity in decimal.
It should be noted that the WD RF does not represent the recovery for a waterflood. The WD RF
correlation is for a natural water drive reservoir. Also, there is no specific API recovery factor
equation for a gas drive reservoir.
As rules of thumb:
- The recovery factor for a solution gas drive, or depletion drive oil reservoir, is in the range
of 5% to 20%
- The recovery factor for a gas cap drive oil reservoir is in the range of 10% to 30%
- The recovery factor for a water drive oil reservoir is in the range of 30% to 80%
An alternate approach for calculating ultimate oil recovery factor is to calculate the unit recovery
factor in bbls/ac-ft and multiply by the area of reservoir, as follows.
For an undersaturated oil reservoir with Pi above Pb, unit recovery to the bubble point pressure
of the depletion drive oil reservoir, Ndep, with no water drive is:
Ndep = 7758φ (Pi - Pb) [co + cr + Sw(co - cw)] / [Boi (1 + co(Pi - Pb))] (RFA-3)
For saturated oil, the unit recovery for a solution gas drive oil reservoir, NSGD, with no water drive,
is based on residual gas saturation as follows:
NSGD = 7758φ [(1 - Sw)/Boi - (1 - Sw - Sgr)/Boab] (RFA-4)
Sgr = 0.625 – 1.3125φ (RFA-5)
The unit recovery for a water drive oil reservoir, NWD, is:
NWD = 7758φ [(1 – Sw)/Boi – Sor/Boab] (RFA-6)
Where: φ is porosity in decimal, Pi is initial reservoir pressure in psi, Pb is bubble point pressure
in psi, co is oil compressibility in psi-1, cr is rock compressibility in psi-1, cw is water compressibility
in psi-1, Boi is initial oil Bo, Sw is initial water saturation in decimal, Sgr is residual gas saturation in
decimal, Boab is oil Bo at abandonment pressure Sor is residual oil saturation in decimal.
RFA.1.2 Probabilistic Oil Recovery Factor

Probabilistic oil recovery factor is an optional calculation performed using Monte Carlo
simulation on the API equations. The P90, P50, P10 normal distributions are used as input to the
simulation and the number of iterations are specified. The probabilistic recovery factor results
for a solution gas drive and a water drive oil reservoir are presented. The ‘Expected Value’ based
on the probabilistic results are also presented (Figure RFA-3).
Figure RFA-3: Probabilistic Oil Recovery Factor
For each iteration of the Monte Carlo simulator, the fluid properties are re-calculated based on
the randomly sampled pressure. This ensures that all parameters that make up the API equations
are converted to probabilistic distributions.
To generate a new probabilistic simulation for the same input distribution (if Seed=-1), just click
the 'MC Simulation' button again. The 'Seed' value is included so the same probabilistic results
can be re-generated. By entering a specific seed value, the same random Gaussian distribution
will be used for the simulation. Entering '-1' for the seed will generate random seed numbers so
the results will be different for each simulation run. The default number of Monte Carlo
simulations is 10,000. This results in a relatively smooth probabilistic distribution curve.
RFA.2 Deterministic Gas RF

The deterministic recovery factor for a gas reservoir is based on material balance calculations.
For a volumetric gas reservoir, recovery factor is dependent on abandonment pressure (Figure
RFA-4).
Figure RFA-4: Deterministic Gas Recovery Factor

If a water drive is present, gas recovery will be reduced because of residual, or bypassed, gas.
The gas recovery factor equations are as follows.
Vol RF = 100 (1 - Bgi/Bgab) = 100 (1 – PabZi/PiZab) (RFA-7)

Limited WD RF = 100 (SgPi/Zi - SgrPab/zab) / (SgPi/Zi) (RFA-8)
Strong WD RF = 100 (Sg - Sgr) / Sg (RFA-9)
Sgr = 0.625 – 1.3125φ (RFA-10)
Where: Vol RF is the volumetric gas recovery factor in %, Weak WD RF is the gas recovery factor
for a reservoir with a limited water drive in %, Strong WD RF is the gas recovery factor for a
reservoir with a strong water drive in %, Pab is the abandonment pressure, Pi is the initial pressure,
Sg is the initial gas saturation (1 - Sw) in decimal, Sgr is residual gas saturation in decimal, Bgab is
the gas formation factor at Pab, Bgi is the gas formation factor at Pi, Zi is the Z-factor at Pi, and Zab
is Z-factor at Pab.
As rules of thumb:
- If a compressor is not installed, Pab can be estimated at 100psi/1000ft of depth
- The recovery factor for a volumetric reservoir is in the range of 80% to 90%
- The recovery factor for a weak water drive reservoir is in the range of 70% to 80%
- The recovery factor for a strong water drive reservoir is in the range of 60% to 70%
- High permeability will tend to move recovery factor to the top of the range
No probabilistic calculations for gas recovery have been implemented in the PE² Essentials
Recovery Factor Analysis tool.
RFA.3 Unconventional Oil/Gas RF

For unconventional oil and gas reservoirs (Figure RFA-5), there are no industry standard
published correlations for recovery factors. The correlations presented here were generated for
use in PE² Essentials using public domain recovery data for unconventional reservoirs.
Figure RFA-5: Unconventional Oil/Gas Recovery Factor

For the unconventional empirical recovery factor, recovery data from a 2013 report published by
the US EIA (www.eia.gov/analysis/studies/worldshalegas/pdf/overview.pdf) was used to build
the gas recovery correlations. The published data included in-place volumes and EUR volumes
for 137 shale reservoirs in 41 countries.
Figure RFA-6 presents a plot of the gas in place versus EUR data for unconventional gas reservoirs
and Figure RFA-7 presents the oil in place versus EUR data for unconventional oil reservoirs.
Figure RFA-6: Unconventional Gas GIIP vs EUR
Figure RFA-7: Unconventional Oil OIIP vs EUR
The R2 for the gas plot is 0.9497 and the R2 for the oil plot is 0.9756. The trends apparent on these
plots yielded the following correlation equations for the unconventional recovery factor.
Gas RF = 10.2798 GIIP0.1169 (RFA-11)

Oil RF = 3.9581 OOIP0.0329 (RFA-12)
Where: Gas RF and Oil RF are in %, GIIP is in Tscf and OOIP is in billion bbls.
The Recovery Factor Analysis tool also presents an estimate of ‘Potential Field IP’ for full
development of an unconventional oil or gas field. Note – no assumption on well count is made
for this estimate. These calculations were presented in an Oil and gas Journal article
(www.ogj.com/articles/print/vol-110/issue-12/exploration-development/evaluating-
production-potential-of-mature-us-oil.html) and are as follows.
Qg = 230 (GasRF GIIP/100)1.0664 (RFA-13)

Qo = 260 (OilRF OOIP/100)0.7088 (RFA-14)
Where: Qg is potential gas field IP in mmscf/d, GIIP is initial gas in place in Tscf, GasRF is calculated
gas recovery factor in % (Equation 4-11), Qo is potential oil field IP in mbopd, OOIP is initial oil in
place in billion bbls and OilRF is calculated oil recovery factor in % (Equation RFA-12).
The ‘Potential Field IP’ is the potential production that may be achieved if the field was fully
developed. No assumptions have been made concerning number of wells required to achieve
this rate.
RFA.4 Reservoir Complexity Index Oil RF

Oil recovery factor based on Reservoir Complexity Index, RCI, is an EPCI enhancement to the work
published by Wickens and Kelly (Wickens, L. M., Kelly, R., Rapid Assessment of Potential Recovery
Factor: A New Correlation Demonstrated on UK and USA Fields, SPE134450, 2010). Wickens and
Kelly used data from 24 UK North Sea fields in a linear correlation for RCI-RF with an applied
weighting to maximize the R² of the correlation. The database used by Wickens and Kelly was
increased to include 43 Norwegian North Sea fields. The complete list of fields and their scores
are presented in Table RFA-1.
The scores listed in Table RFA-2 were generated using the criteria in Table RFA-1.
Table RFA-1: RCI Scoring Criteria

Table RFA-2: RCI Scores and Reported Recovery

The scores were normalized, and a power correlation was used for the normalized RCI with
exponent weighting to maximize the R² of the normalized RCI-RF correlation. The resulting RCI-
RF correlation is presented as Equation RFA-15.
RCImax = 5StructExp + 5VisExp + 5AreaExp + 5PermExp

RCIcalc = StructScoreStructExp + ViscosityScoreVisExp + AreaScoreAreaExp + PermScorePermExp
RCInorm=RCIcalc/RCImax
RF = 76.727 - 134.27 RCInorm (RFA-15)
Where: RF is oil recovery factor in %, the Scores are assigned based in the criteria presented in
Table RFA-2 and the exponents were chosen to maximize the R2 of the correlation.
The calculated normalized RCI and RCI-based oil RF are presented as output (Figure RFA-8).
Figure RFA-8: RCI-based Oil Recovery Factor

RFA.5 Alternative North Sea Oil RF

While investigating recovery factors, a number of analyses using regression were attempted for
North Sea oil fields. Many correlations were examined including API, OOIP and kh (Figure RFA-9).
Figure RFA-9: Possible Correlation Plots for Oil Recovery Factor
The kh plot appeared to show a general semi-log correlation with RF. Assuming that the kh trend
is valid, a high, mid, low correlation was generated for the data (Figure RFA-10).
Figure RFA-10: Semi-log kh Correlation for Oil Recovery Factor

High RF = 7.029 log(kh) + 16.815 (RFA-16)

Mid RF = 12.626 log(kh) - 20.252 (RFA-17)
Low RF = 19.692 log(kh) – 72.263 (RFA-18)
Based on the kh correlation plot, this correlation should be used only if kh is greater than
~5000md-ft.
Two other potential oil recovery factor correlations were generated as well, a kh/OOIP
correlation (Figure RFA-11) and an api-kh/OOIP correlation (Figure RFA-12).
Figure RFA-11: Semi-log kh/OOIP Correlation for Oil Recovery Factor
Figure RFA-12: Semi-log api-kh/OOIP Correlation for Oil Recovery Factor

The kh/OOIP and api-kh/OOIP plots appear to show a general semi-log correlation with RF. If the
trends actually exist, a high, mid, low correlation was generated for the data as follows.
RF = 20.244 log(kh/OOIP) + a (RFA-19)

RF = 20.244 log(api kh/OOIP) + b (RFA-20)
Where: a and b are listed on the plots and represent the high, mid, low constants.
These correlations (Figures RFA-10, RFA-11 and RFA-12) should only be used to estimate ranges
of recovery.
As an example, assume a reservoir has a kh of 300md, 150ft of pay and contains 250mmbbls of
35˚api oil. The correlating function for Figure RFA-12 is 6300 (api-md-ft/mmbbls). The resulting
range of recovery factors is 59.9% / 41.9% / 23.9%.
It should be noted that there are no recovery factors above 65% in this data set. A calculated RF
that exceeds 65% should be set at 65%.
RFA.6 Artificial Neural Network (ANN) Oil RF Model

The ANN RF model is a new oil recovery model developed using an open source neural network
model. There are a number of Neural Networks programs available that are based on the
OpenNN system. Figure RFA-13 shows the inputs required for the PE2 Essentials ANN Oil RF
model. More information on the model was published in the March 2019 issue of World Oil.
Figure RFA-13: Artificial Neural Network Oil RF Model

RFA.6.1 Introduction to Artificial Neural Networks (ANN)

Neural networks, a technique of machine learning, is a branch of artificial intelligence which
attempts to model high-level data interactions by using complex architectures which perform
multiple transformations.
The fundamental basis for neural network technology is the Universal Approximation Theorem
(http://en.wikipedia.org/wiki/Universal_approximation_theorem). This theorem states that any
continuous function that maps a set of real numbers to another set of real numbers can be
approximated to some degree of accuracy by a feed-forward neural network with a single hidden
layer and a finite number of hidden units (neurons or perceptrons) which contain non-linear
transfer functions - Figure RFA-14. In almost all implementations of ANN’s, the non-linear transfer
function is a hyperbolic tangent (tanh).
Figure RFA-14: Artificial Neural Network Model
Neural networks are particularly well-suited for modeling complex non-linear relationships which
cannot be easily modeled by traditional linear regression methods, for example, oil recovery
factors.
Full-featured ANN programs are normally combined with genetic algorithms, statistics/linear
regression, and fuzzy logic to automatically find optimal or near-optimal solutions for the
problem. For more information on these concepts refer to:
https://en.wikipedia.org/wiki/Genetic_algorithm and https://en.wikipedia.org/wiki/Fuzzy_logic.
ANN’s may be a branch of artificial intelligence but they are basically high tech statistical
calculators. As with all statistical analysis the results are completely dependant on how ‘well-
behaved’ the input data is.
The author has successfully used ANN’s in a number of implementations in the oil industry, for
example, predicting IP30 (30-day IP), IP90 and IP365 for hydraulically fractured horizontal wells.
This model could then be used to optimize well spacing and the hydraulic fracture programs of
future horizontal wells.
RFA.6.2 Building the ANN Oil RF Model

The validity of any ANN model is dependant on how well the system has been ‘trained’. Normally
a large dataset of well-behaved data is used for training and a different set of data is used as a
blind test once the model is built. ANN systems will subdivide the training data set into a training
set and a validating subset. This is not the same as using a set of blind test data. The validating
subset is still part of the training subset.
Based on the API RF equations (Equation RFA.1 and RFA.2) and general reservoir engineering
principals, the reservoir parameters used to build the ANN were: STOOIP, porosity, permeability,
viscosity, Oil API and net pay. Although not the only parameters that impact recovery, these
parameters are considered to be the major components impacting long-term recovery of oil from
a reservoir. The final form used as input to the ANN were: Log(STOOIP), Log(kh), Log(viscosity),
Log(phi-h) and Oil API. The logarithm of values was used to place the max/mins in a reasonable
range and to pre-process the data prior to use in the ANN.
For the ANN Oil RF model, data from 264 sandstone/clastic reservoirs were used. A random
subset of 46 reservoirs were removed from the data set to be used as a blind test. Of the
remaining 218 reservoirs, 20% were used as validating data for training. Figure RFA-15 presents
the available data for the sandstone reservoir data set.
It is obvious from Figure RFA-15 that the data has a large scatter. This is not unexpected when
considering reservoir recovery factors. It is one of the problems with trying to generate a recovery
factor correlation that is universally valid.
Nevertheless, each of the ANN inputs do tend to show a trend with respect to recovery factor.
These trends will help the ANN system find an optimal solution.
It should be noted that the PE2 Essentials ANN Oil RF Model continues to be developed as more
data becomes available. The most current model was published in the March 2019 issue of World
Oil.
A second data set comprised of 38 carbonate/dolomite reservoirs was used to develop the ANN
Oil RF model for carbonate reservoirs. A random subset of 5 reservoirs were removed from the
data set to be used as a blind test.
Figure RFA-16 presents the available data for the carbonate reservoir data set.
It is obvious that the carbonate ANN model will not be as robust because of the small dataset.
Figure RFA-15: Sandstone Data Set for ANN Oil RF Model

Figure RFA-16: Carbonate Data Set for ANN Model
The big unknown for ANN’s is the number of neurons to include in the model. With well-behaved
data, it may be possible to minimize the neurons so that they are less than or equal to the number
of inputs to the model – 5 or 6 in this case.
It was found that using a small number of neurons could not handle the problem. Specifically,
the lower the neuron count the less likely the ANN was able to model the high and low RF trends.
Figure RFA-17 shows the result, when 4 neurons were used in the ANN model. It was apparent
that the ANN had trouble finding an optimum solution that would cover the entire range of RF’s
for the low neuron counts.
Figure RFA-17: ANN Model – 4 Neurons
Multiple models were built to try and find an optimum number of neurons that would yield an
optimum solution that was valid over the entire range of RF’s. Figure RFA-18 shows the R2
correlation for ANN’s built using different numbers of neurons.
Figure RFA-18: ANN Model – Correlation Coefficients
From Figure RFA-18 there appears to be a maximum correlation coefficient for an ANN with 10
neurons.
Figure RFA-19 shows the result for an ANN trained with 10 neurons. Figure RFA-19 includes both
the training data set and the blind test data set. There is still a scattered in the results but with
the scatter in the input data, this was expected.
Figure RFA-19: ANN Model – 10 Neurons
As a final check, a second Neural Network program was used to build a second ANN with 10
neurons. Figure RFA-20 shows the two models. Both models are included in PE2 Essentials.
Figure RFA-20: ANN Oil RF Models – Sandstone Reservoirs
Model 2 appears to handle the blind test data better than Model 1, but Model 1 appears to model
the high/low RF trends better than Model 2.
For carbonate reservoirs, 6 neurons were found to be optimum to model the RF’s (Figure RFA-
21).
Figure RFA-21: ANN Oil RF Models – Carbonate Reservoirs
Because of the small dataset, the ANN model looks very good, but caution should be used when
using the results from this model.
RFA.6.3 ANN Oil RF Model – Sandstone Model

Figure RFA-22 summarizes the ANN Oil RF Model for sandstone reservoirs.
Figure RFA-22: PE2 Essentials ANN Oil RF Models – Sandstone Reservoirs

The equation that represents the ANN model for sandstone reservoirs is presented below.
Scaled_Log_OOIP = (Log_OOIP - 2.65783) / 0.699151

scaled_log_kh = (log_kh - 4.41412) / 0.991662
scaled_API = (API - 33.5211) / 6.89976
scaled_log_Viscosity = (log_Viscosity - 0.186206) / 0.568663
scaled_Log_Poro_h = (Log_Poro_h - 3.30592) / 0.494583
a = Tanh(-5.3314 - 0.653912*scaled_Log_OOIP + 4.64736*scaled_log_kh -

8.25336*scaled_API + 3.30679*scaled_log_Viscosity + 3.32009*scaled_Log_Poro_h)
b = Tanh(-1.21679 + 1.47136*scaled_Log_OOIP - 1.70876*scaled_log_kh -

0.151557*scaled_API - 2.59019*scaled_log_Viscosity + 1.19481*scaled_Log_Poro_h)
c = Tanh(-7.71927 - 2.77029*scaled_Log_OOIP + 2.82045*scaled_log_kh +

d = Tanh(-1.25949 + 2.11782*scaled_Log_OOIP - 2.26974*scaled_log_kh -

0.14775*scaled_API - 3.29086*scaled_log_Viscosity + 1.21194*scaled_Log_Poro_h)
e = Tanh(1.55945 - 0.249741*scaled_Log_OOIP + 0.564794*scaled_log_kh -

0.328641*scaled_API - 0.0815263*scaled_log_Viscosity - 0.248626*scaled_Log_Poro_h)
f = Tanh(-2.73454 + 0.763912*scaled_Log_OOIP - 1.97289*scaled_log_kh +

0.138494*scaled_API + 2.13543*scaled_log_Viscosity - 1.59942*scaled_Log_Poro_h)
g = Tanh(-0.211486 + 0.754921*scaled_Log_OOIP - 0.0672823*scaled_log_kh +

0.84095*scaled_API - 0.0102544*scaled_log_Viscosity - 0.949527*scaled_Log_Poro_h)
h = Tanh(-5.77792 + 0.160162*scaled_Log_OOIP - 1.71346*scaled_log_kh -

i = Tanh(1.05659 - 3.47005*scaled_Log_OOIP - 0.586121*scaled_log_kh -

j = Tanh(-4.67433 - 1.40063*scaled_Log_OOIP + 3.41236*scaled_log_kh -

scaled_RF= -2.06301 + 0.271591*a + 2.75565*b + 0.397435*c - 2.35473*d + 3.00811*e +

0.777111*f + 0.976147*g - 0.885672*h + 0.406045*i + 0.215362*j
ANN_RF = (0.5 * (scaled_RF + 1.0) * (78.4 - 4) + 4) (RFA-21)
Equation RFA-21 is the expression for ANN Model 1 in the PE2 Essentials RF tool.
RFA.6.4 ANN Oil RF Model – Carbonate Model

Figure RFA-23 summarizes the ANN Oil RF Model for carbonate reservoirs. The equation
(Equation RFA-22) that represents the ANN model for carbonate reservoirs is presented below.
Figure RFA-23: PE2 Essentials ANN Oil RF Models – Carbonate Reservoirs
scaled_Log_OOIP=(Log_OOIP - 3.02819) / 0.636082

scaled_log_kh=(log_kh - 3.40965)/ 0.909265
scaled_API_Gravity=(API - 34.2652) / 7.58305
scaled_log_Viscosity=(log_Viscosity - 0.101507) / 0.790694
scaled_Log_Poro_h=(Log_Poro_h - 3.23401) / 0.523619
a=Tanh(-1.25222 + 4.32191*scaled_Log_OOIP - 0.117155*scaled_log_kh +

2.23329*scaled_API_Gravity + 2.91222*scaled_log_Viscosity - 5.90619*scaled_Log_Poro_h)
b=Tanh(-1.98078 - 0.690941*scaled_Log_OOIP + 3.41286*scaled_log_kh -

3.52012*scaled_API_Gravity - 6.97704*scaled_log_Viscosity - 2.65092*scaled_Log_Poro_h)
c=Tanh(2.44436 - 1.78361*scaled_Log_OOIP - 3.22423*scaled_log_kh +

1.56915*scaled_API_Gravity + 1.10927*scaled_log_Viscosity + 1.46621*scaled_Log_Poro_h)
d=Tanh(-1.10494 - 2.3695*scaled_Log_OOIP + 7.75508*scaled_log_kh -

6.85185*scaled_API_Gravity - 14.7533*scaled_log_Viscosity - 6.9812*scaled_Log_Poro_h)
e=Tanh(4.11763 - 0.223371*scaled_Log_OOIP + 2.64977*scaled_log_kh +

1.30149*scaled_API_Gravity + 0.934554*scaled_log_Viscosity + 0.961708*scaled_Log_Poro_h)
f=Tanh(0.135445 - 2.61746*scaled_Log_OOIP - 0.704636*scaled_log_kh +

4.23633*scaled_API_Gravity + 1.83716*scaled_log_Viscosity - 1.18279*scaled_Log_Poro_h)
ANN_RF = 14.9587 + 16.6988*a - 14.3414*b + 8.04666*c + 16.7378*d + 17.123*e - 12.9095*f

(RFA-22)
PE Asset Valuation Essentials

The Asset Valuation Essentials section contains the following:
• Basic Project Economics
• Comprehensive Asset Performance Evaluation
Project Economics Analysis Tool

The PE² Essentials Project Economics tool is a scoping economics model that can use either an
imported forecast or forecasts generated by other PE² Essentials tools (Figure ECO-1).
Figure ECO-1: PE² Essentials Basic Project Economics Tool
The tool generates before tax net present value (NPV) and can perform full-cycle and look-back
economic analysis. This version has the option to perform simplified corporate economics. The
Project Economics tool can be used to evaluate single or multi-well projects.
466 Project Economics Analysis Tool
As a caveat, although this program has been built from fully debugged routines that have been
used in the past, there are so many permutations and combinations possible with this model that
it is difficult to test all combinations that may be used. If you end up with anomalous results that
do not seem to make sense, send the information to the author for evaluation.
ECO.1 Oil and Gas Prices

Oil and gas prices can be entered as a single value and then modified using the appropriate
escalation factor (Figure ECO-2). Alternatively, an oil/gas price forecast can be loaded through
the 'Oil/Gas Price Forecast' button on the main screen.
Figure ECO-2: Project Economics Tool – Oil/Gas Prices
To facilitate loading of historical oil and gas prices, an Excel spreadsheet (‘Historical and
Forecasted Oil and Gas Prices.xlsx’) that includes oil and gas prices from January 1974 has been
included in the “Example Input Files\Excel Files” directory. This file also includes a price forecast
(2017+). To incorporate a different price forecast, modify the data in the ‘Forecast Prices to 2028’
spreadsheet.
After clicking the 'Oil/Gas Price Forecast' button, the Price Import screen will show the oil and
gas prices currently stored with the model (Figure ECO-3). To change out these values, import
the new data.
The oil and gas prices can be imported from an Excel file or loaded from a price deck saved in the
PE Tools database (‘Import Price Deck From PE Tools db’). If the price data is imported from an
Excel file, the price deck can be saved to the PE Tools database.
When a price forecast file is loaded, the 'Oil Price ($/bbls)' and 'Gas Price ($/mscf)' on the main
screen (Figure ECO-1) will be set to '-1'. To disable the loaded price forecast, enter a value for oil
and/or gas price in the appropriate box. It is possible to disable just the gas price forecast or just
the oil price forecast by entering the relevant price and leaving the other price as '-1'. The price
forecast can be re-enabled by re-entering '-1' for the price.
Figure ECO-3: Project Economics Tool – Oil/Gas Price Import
All oil and gas prices, whether a single value or imported, should be entered in today's currency.
The oil and gas escalation factor (Section ECO.2) is used to convert to money-of-the-day. The
‘Price Esc Cap’ will place an upper limit on the price escalation. For example, if an oil price of $45
is entered, or is the last value in the oil price forecast, and the escalation cap is set at 1.2 then oil
price will escalate annually to a maximum of $54 and then stay constant for the remainder of the
forecast period.
Note that the oil/gas price forecast is saved with the model using 'Save Model' so the price
forecast does not have to be re-entered.
ECO.2 Escalation Factors

The cost of a number of operating and capital expense parameters can be escalated over time to
take inflation into account (Figure ECO-4).
Figure ECO-4: Project Economics Tool – Escalation Factors

The following parameters can be escalated:

▪ Oil Price
▪ Gas Price
▪ Capital Costs:
• Well Capital Costs
• Facility / Project Capital Costs
▪ Operating Costs:
• Head Office Overhead
• Fixed Well Operating Costs
• Variable Well Costs
▪ Oil/Gas Pipeline and Transportation Costs
▪ Oil/Gas Processing Costs
▪ Water Disposal Costs
The Annual Escalation Rates are entered in the ‘Escalation Factors’ section and costs are
escalated using Equation ECO.1.
EscFactor = (1 + EscRate/100)t (ECO.1)
Where: EscFactor is the annual escalation, EscRate is the escalation rate and t is the time in years.
ECO.3 Discount Rate, Internal Rate of Return and Discount Year

The annual discount rate is entered into the model and the internal rate of return (IRR) is
calculated during the economic forecast (Figure ECO-5). To facilitate full cycle, or look-back,
analysis the year to start discounting can be set as shown in Figure ECO-6.
Figure ECO-5: Project Economics Tool – Discount Rate and IRR
Figure ECO-6: Project Economics Tool – Forecast Year to Start Discounting
Since the economic analysis assumes a ‘0’ year start date, for full cycle economics the start of
discounting will be at the end of the history.
To take the time value of money into account, all future revenue is converted to a common
reference point in time. This is assumed to be the current year or the present (hence the term,
‘present value’). This is achieved by discounting future net cash flow. Discounting converts a
future sum of money into the equivalent of present-day cash.
The rate used for discounting future cash flow is called the discount factor and is entered into
the model in the ‘Annual Discount Rate’ input box (Figure ECO-5). The annual discount factor is
calculated at mid-year using Equation ECO.2.
DiscFactor = (1 + DiscRate/100)-(t - t0 + 0.5) (ECO.2)
Where: DiscFactor is the annual discount applied to the cash flow, DiscRate is the discount rate
entered on the main sheet and t is the time in years, t0 is the year to start discounting (defaults
to year 1) and 0.5 specifies mid-year discounting.
The present value, PV, of a net cash flow, netCF, received at some future time, t, is given by
Equation ECO.3.
PVt = netCFt DiscFactort (ECO.3)
Where the subscript t is the time in years when the cash flow is received.
The net cash flow at a given time netCFt is calculated using Equation ECO.4.
netCFt = TotalNetRevenue – TotalOpCosts – TotalCapEx (ECO.4)
Each revenue and cost stream has the appropriate escalation factors applied and then the
discount factor is applied to generate the net present value.
The Internal Rate of Return, IRR, is the discount factor that will make the net present value, NPV,
equal to zero. IRR is found by iterating on the annual discount factor until the NPV is 0.
ECO.4 Operating, Capital and Sunk Costs

The operating costs and capital costs are entered in appropriate sections of the model (Figures
ECO-7, ECO-8 and ECO-9).
Figure ECO-7: Project Economics Tool – Operating Costs

Figure ECO-8: Project Economics Tool – Well Capital Costs
Figure ECO-9: Project Economics Tool – Facility/Project Capital Costs
The costs are dependent on the forecast being generated. In terms of ‘Well Capital Costs’, if a
multi-well forecast is being generated then the unit costs should be an average for the item. For
example, not all wells would have a ‘Total Measured Well depth’ of 4500 feet. Instead this
represents the average depth for all wells so the total cost for the multi-well project is valid.
For a multi-well development, it is assumed that the subsequent well will be drilled after the
previous well is online. Total well costs will be escalated dependent on when the well is
producing. The individual well total cost is calculated using Equation ECO.5.
TotalWellCost = (DailyCost)(WellMD)/ROP + CompCost + TieInCost (ECO.5)
Where: DailyCost is the daily drilling cost, WellMD is the measured depth of the well, ROP is the
rate of penetration, CompCost is the cost to complete the well and TieInCost is the cost to tie-in
the well, clean up the well and includes any other costs associated with the well.
The annual operating costs are escalated based on the escalation rates entered into the model.
TotalOpCostsEsc = (HeadOffice)(EscFactorOpEx) +
(FixedCost) (EscFactorOpEx) +
(VariableCosts)(Qg)(ProdDays)(EscFactorOpEx) +
(VariableCosts)(Qo)(ProdDays)(EscFactorOpEx) + (ECO.6)
(GasTransFee)(Qg)(ProdDays)(EscFactorOilGasTrans) +
(OilTransFee)(Qo)(ProdDays)(EscFactorOilGasTrans) +
(GasProcessFee)(Qg)(ProdDays)(EscFactorOilGasProcess) +
(OilProcessFee)(Qg)(ProdDays)(EscFactorOilGasProcess) +
(WaterProcessFee)(Qw)(ProdDays)(EscFactorWaterProcess)
The annual capital costs are also escalated based on the entered escalation rate. The cost for the
initial wells (Section ECO.5) is considered to be sunk cost and is not escalated. Subsequent well
costs are escalated based on when they are available.
TotalCapCostsEsc = (OngoingCapCosts)(EscFactorCapEx) + (ECO.7)

(TotalWellCost)(EscFactorCapEx)
All costs are discounted after escalation.
Sunk costs are not escalated or discounted and are calculated using Equation 9-8.
SunkCapCosts = PreStartProjCosts + (TotalWellCost)(#InitialWells) (ECO.8)
Note when using the preceding equations, caution must be used to ensure all terms are
consistent, thousands$ or millions$.
ECO.5 Well Count and Well Profile

Initial and final well counts are entered prior to the economic run (Figure ECO-10).
Figure ECO-10: Project Economics Tool – Well Count
A minimum initial well count of one well is required to run the economics. If the purpose of the
run is to model a single well, the final count is set equal to the initial well count. If multiple wells
are to be included, the final well count is set appropriately.
A base well profile is entered by clicking ‘Load Well Profile’ (Figure ECO-11). Production profiles
can be imported from the PE Tools database; from Excel; or from a separate DCA database.
Refer to the information button on the Production Profile Import page for information
concerning Excel formats. When importing from Excel, always make sure to check that the ‘Units’
and the ‘Fluid Type’ parameters are set properly before importing the data.
It is possible to import production data from an Excel spreadsheet; from a PE² Essentials PE Tools
database, or a DCA database as shown in Figure ECO-11.
Figure ECO-11: Project Economics Tool – Import Production Forecast
If data is imported from an Excel file, the data can be saved to the PE Tools database with the
‘Save to PE Tools db’ button on the Production Profile Import page.
It should be noted that the production profile (Figure ECO-12) is saved with the economics model
and does not need to be imported every time the Economics tool is loaded.
Figure ECO-12: Project Economics Tool – Imported Production Data
It is possible to read a standalone DCA database file and import forecasts or history+forecasts of
one or more of the wells in the DCA database. When “DCA Database” is chosen for import a
screen pops up to choose the well data to be imported (Figure ECO-13).
Figure ECO-13: Project Economics Tool – Import Production Forecast
When there is more than one well used in the forecast, it is possible to add variability to the well
profiles by checking 'Vary Well Profiles' (Figure ECO-10). The first well will use the loaded
production profile, but subsequent wells will use a profile that has been randomly modified by a
factor between 0.75 and 1.25. Figure ECO-14 shows the single well profile (left) and a 5-well
project profile (right) incorporating variable well rates.
Figure ECO-14: Project Economics Tool – Multi- Well Production Data

ECO.6 Economic Results

The net gas and oil/condensate prices are escalated prior to being discounted. This escalated
price is then applied to the production stream to generate the revenue stream. Finally, costs are
added, and discounts are applied to generate the net present value and corresponding plots for
the total project are generated (Figures ECO-15 and ECO-16).
Figure ECO-15: Project Economics Tool – Plots of Economic Analysis
Figure ECO-16: Project Economics Tool – Table of Economic Analysis
The economic equations are as follows:
NetGasPriceesc = (NetGasPrice)(EscFactorgas) (ECO.10)

NetOilPriceesc = (NetOilPrice)(EscFactoroil) (ECO.11)
NetGasRevenueesc = (NetGasPriceEsc)(Qg)(ProdDays) (ECO.12)
NetOilRevenueesc = (NetOilPriceEsc)(Qo)(ProdDays) (ECO.13)
TotalNetRevenueesc = (NetGasRevenuedisc)(NetOilRevenuedisc) (ECO.14)
CashFlowesc = TotalNetRevenueesc - TotalOpCostsEsc - TotalCapCostsEsc (ECO.15)
(ECO.16)
CumCashFlowesc =∑(CashFlowesc)t - SunkCapCosts

CashFlowescDisc = (CashFlowesc)(DiscFactort) (ECO.17)
CumCashFlowescDisc =∑(CashFlowescDisc)t - SunkCapCosts (ECO.18)
NPV = CumCashFlowescDisc (ECO.19)
All cash flow calculations are presented on an annual basis except for the first year where the
economics are presented on a monthly basis (Figure ECO-15) so the well start-ups can be
observed when multi-well forecasts are made.
The reported economic parameters (Figure ECO-16) are as follows:

• ‘NPV’ is CumCashFlowescDisc
• ‘Econ Life’ is determined as the time when annual cash flow goes negative
• ‘Total Cap Costs’ is the total project capital costs
• ‘Max Exposure’ is maximum undiscounted cash required for the project
• ‘PIR’ is NPV / Total Cap Costs
• ‘Break Even Year’ is the year when cumulative cash flow goes positive
ECO.7 Corporate Economics

After the main project economics have been run, it is possible to run a corporate economic
analysis of the project. Clicking the ‘Corporate Economics’ button opens up the corporate
economics tool (Figure ECO-17).
Figure ECO-17: Project Economics Tool – Corporate Economics

The Corporate Economics tool takes the cash flow forecast generated for the project and
calculates before and after-tax economics for the company based on entered tax and royalty
rates.
This tool includes input for the following parameters:
• Cost Recovery Uplift – when wells are drilled on penalty, the operator can recover an
uplift portion of the well costs. As an example, if the penalty is 300% and the well costs
$5 million, then a cost recovery uplift of be $10 million would be entered.
• Working Interest Before P/O – this is the company interest prior to payout of the capital
costs.
• Working Interest After P/O – this is the company interest after payout of the capital costs.
• Over-Riding Royalty – This is the ORR that the company has to pay to a silent partner.
• Corporate Income Tax Rate – This is the corporate income tax rate.
• Time to Depreciate CapEx – This is the time over which the CapEx is depreciated for
income tax purposes.
• Royalty Rate Before P/O – this is the government royalty rate prior to payout of the capital
costs.
• Royalty Rate After P/O – this is the government royalty rate after payout of the capital
costs.
After the corporate economics has been run, the results can be saved to a csv file by clicking the
‘Save Company Results’ button on the main screen. The ‘Save Project Results’ button will save
the gross project economic analysis to a csv file.
ECO.8 Project Economic Example – Marcellus

The final step in the analysis of the Marcellus well is to evaluate the full cycle economics of the
P90, P50 and P10 forecasts. To generate the economics, Figure ECO-18 presents the cost for a
Marcellus well (source: http://ir.eqt.com)
Figure ECO-18: Project Economics Tool – Example Costs

Since this is a full cycle economic analysis the historical oil and gas prices from April 2014 were
loaded into the Project Economics tool (Figure ECO-19).
Figure ECO-19: Project Economics Tool – Price History/Forecast
The assumptions for this exercise are that the well was drilled in early 2014 and started
production in April 2014. The well has 21 months of available history which ends at December
2015. To perform full cycle economics, no escalation on the realized prices to the end of 2016
was applied. Start of discounting was set at 2.75 years (23 months from April 2014).
The production forecasts, including the historical production, generated by the DCA and Monte
Carlo DC Forecast programs are shown in Figure ECO-20 and were used to evaluate the full cycle
economics of this Marcellus well.
Figure Eco-20: Project Economics Tool – Marcellus Full Cycle Forecasts

The assumptions used to generate the full cycle economic analysis are as follows:
• Well capital costs were based on a 6400-foot lateral
• US$6mm total well costs
• US$0mm/yr head office
• US$50k/yr fixed well costs
• US$0.25/mscf variable costs
• US$0.5/mscf transportation
• US$0.5/mscf processing
• Production of the well was assumed to start April 2014
• Discounting applied after 2.75 years
After all the parameters were entered into the model (Figure ECO-21), each production forecast
was imported (Figure ECO-22) and economics for each forecast was run.
Figure ECO-21: Project Economics Tool – Marcellus Model: P50 Analysis

Figure ECO-22: Project Economics Tool – Marcellus Model: Import Forecasts
The full cycle economic results are presented in Figure ECO-23 and the IRR results are presented
in Figure ECO-24.
From the economic analysis, it is obvious that for the assumed historical/forecast gas prices, the
Marcellus economics are positive and show a 7-year to 10-year payback of costs.
Figure ECO-23: Project Economics Tool – Economic Results: Marcellus Model
Figure ECO-24: Project Economics Tool – IRR Results: Marcellus Model

480 Asset Economic Evaluation Tool
Asset Economic Evaluation Tool

The PE² Essentials ‘Asset Economic Evaluation’ tool is a full-featured economics model used for
Comprehensive Asset Performance Evaluation (CAPE) (Figure CPE-1).
Figure CPE-1: PE² Essentials Comprehensive Asset Economic Performance Evaluation Tool
The tool generates before- and after-tax net present value (NPV) and internal rate of return (IRR)
for the asset and can perform full-cycle and look-back economic analysis. CAPE includes a number
of generic fiscal regimes.
CAPE can perform single asset analysis and can combine single assets into a multi-asset database.
This database can be password locked for read-only use for audit trail purposes. To add/modify
the database, it has to be unlocked using the locking password. If a password is lost, a master
password can be used to override the lost password.
The state of the database, locked or unlocked, can be determined by the state of the ‘Lock
db’/’UnLock db’ buttons. If the the database is locked as read only, the ‘UnLock db’ button will
be enabled and vice-versa. When individual assets are being evaluated, the ‘Lock db’ and ‘UnLock
db’ buttons are disabled.
CPE.1 CAPE Introduction

Petroleum economics (Asset Valuation) involves the application of the techniques of economic
analysis at every stage in the development of oil and gas exploration and production projects.
The economics of oil and gas projects are affected by a range of factors, including:
• Level of knowledge about the oil or gas field (subsurface description and
characterization, surface facilities, etc.)
• Location (onshore, offshore—shallow water, deep and ultra-deep water—), type and
number of wells
• Market conditions (commodities prices, global supply and demand, worldwide E&P
environment)
• Effect of tax/royalty systems, production sharing and service contracts and overall
fiscal systems stability
By analyzing factors like these, Economists are able to assist in making investment decisions, such
as deciding whether or not to drill an exploration well or whether or not to develop an oil/gas
production project. They are crucial in the negotiations around Production Sharing Contracts and
purchasing oil and gas properties. Economists are also involved in the assessment and
management of the technical, economic and other risks associated with the different phases of
an oil or gas projects.
Given the volatility in commodities (oil, gas, oil products) prices today, the economic evaluation
of upstream oil and gas investments is essential. Business decisions involving asset acquisitions,
lease-buy assessments, exploration drilling options, oil and gas field developments, equipment
purchases, and fiscal negotiations all require detailed economic analysis. CAPE covers cash flow
analysis, deriving and understanding economic indicators and detailed profitability and fiscal
analysis.
CAPE provides detailed economic modelling using cash flow analysis and related sensitivity
analysis, Spider diagrams and tornado plots. The tool currently includes two different fiscal
systems commonly used in the oil & gas industry:
• Tax Royalty Contracts
• Production Sharing Contracts
• Service Contract
The benefits of economic evaluation of projects and thus of a thorough CAPE workflow are:
• Make investments decisions with greater confidence
• Standardize business processes across organization
• Improve the consistency of economic evaluation
• Incorporate risk assessment (by a thorough sensitivity analysis)
• Capture probabilistic analysis
Figure CPE-2 presents the workflow as implemented in the PE² Essentials CAPE tool.
Figure CPE-2: PE² Essentials Asset Economic Evaluation - Workflow
CPE.2 Fiscal Terms for E&P Projects

Fiscal terms (Figure CPE-3) for upstream investments refer to the agreement between a local
government and an oil and gas exploration company to explore, develop and produce
hydrocarbons.
Figure CPE-3: Asset Economic Evaluation - Fiscal Regimes

The objective of a host government is to maximize wealth from its natural resources by
encouraging appropriate levels of exploration and development activity. In order to accomplish
this, governments must design fiscal systems to attract oil and gas companies. The objectives of
the oil and gas companies are to build equity and maximize wealth by finding and producing oil
and gas reserves at the lowest possible cost and highest possible profit margin.
In a competitive world, areas with the least favorable geology, the highest costs, and the lowest
prices at the wellhead would offer the best fiscal terms, while areas with the best geology, the
lowest costs, and the highest prices at the wellhead would offer the toughest fiscal terms.
The role of the host government is to design a fiscal system where exploration and development
rights are acquired by those companies who place the highest value on these rights. Competitive
bidding can help achieve this objective. In the absence of competition, efficiency must be
designed into the fiscal terms.
Regardless of the fiscal system used, the bottom line is the economic issue of how costs are
recovered, and profits divided. In order to accomplish this, fiscal systems are designed to:
• Provide a fair return to the state and to the industry
• Avoid undue speculation
• Have clarity and stability
• Limit undue administrative burden
• Provide flexibility
• Create healthy competition and market efficiency
The design of an efficient fiscal system must take into consideration the political and geological
risks as well as the potential rewards. Detailed economic modeling using discounted cash flow
analysis (as implemented in CAPE) is the best way to evaluate division of profits. Factors that limit
a company’s profits in a contract, such as cost recovery allowance, a government’s right to back-
in for an extra share of production, or an additional tax, can be modeled for the purposes of
contract negotiation or project evaluation. Division of profits is commonly referred as contractor
take or share and government take or share.
CPE.3 Tax and Royalty Contracts (TRC)

To calculate the contractor’s share in a tax and royalty contract, the first item to be deducted
from the gross revenues from oil and gas production are the royalties. Gross revenues less royalty
equals net revenue.
The next item are deductions, which include operating costs, depreciation, and amortization, and
intangible drilling costs. These are deducted from net revenue to arrive at taxable income.
The third item is taxation. Revenue remaining after royalty and deductions is called taxable
income. Taxable income might be taxed in two layers: provincial and federal (e.g. Canadian
terms). The remaining revenue after taxation is the contractor’s share of the revenue.
TRC, or concessionary contracts (Figure CPE-4) are relatively simple and are not as widely used
internationally as the more complex production sharing contracts.
Figure CPE-4: Tax and Royalty Regimes (TRC/Concessions)
Royalties are regressive because they are levied on gross revenues. For less profitable ventures,
the relative percentage of royalty increases. The further from gross revenues that taxes are
levied, the more progressive the system becomes.
CPE.4 Production Sharing Contracts (PSC)

In most contractual systems, the facilities put in place by the contractor within the host
government domain become the property of the state either the moment they are landed in the
country or upon startup or commissioning. Sometimes, the title to the assets or facilities does
not pass to the government until the attendant costs have been recovered.
Contractual systems are divided into service contracts and production sharing contracts. The
difference between them depends on whether or not the contractor receives compensation in
cash or in kind (crude). Figure CPE-5 presents the PSC specifics.
Figure CPE-5: Production Sharing Contracts (PSC)
Basic features of production sharing contracts are as follows:

• The title of the hydrocarbons remains with the host State/Government
• The State maintains management control and the contractor is responsible for the
execution of petroleum operations in accordance to the terms of the contract
• The contractor is required to submit annual work programs and budgets for scrutiny
and approval of the State/Government
• The contract is based on production sharing and not profit-sharing basis
• The contractor provides all financing and technology required for the operations and
bears all risks of the project
• During the term of the contract, after allowance for up to a specified percentage of
annual production for recovery of costs, the remaining production is split between
the contractor and State/Government
• Equipment purchased and imported by the contractor becomes property of the
State/Government (service company equipment and leased equipment are exempt
from this stipulation)
CPE.5 Service Contracts (SC)

A Service Contract is basically a PSC but the contractor gets paid a fee in term of xx$ for every
unit volume produced. As a result, refer to Section CPE.4 for more information.
CPE.6 PSC/SC Advantage versus TRC

When evaluating fiscal terms (Figures CPE-4 and CPE-5), the focus is on division of profits –
government share and contractors share. The most dramatic difference between the two fiscal
regimes has to do with how much taxation is imposed.
PSC/SC terms can be more flexible. There may be more opportunities to come up with more
balanced / attractive terms. For example, reduced income taxes paid within the PSC/SC instead
of paying higher separate income taxes which can be >70%. Another possibility is negotiating a
PSC/SC without royalties.
Under a PSC/SC, there is a higher net income after tax in the early years of production because
of the built-in cost recovery mechanisms. With high cost recovery provisions, increased
government share comes at a relatively later stage of production.
It is possible to obtain the same economic results under a variety of systems. The challenge is to
make the most of the flexibility offered by a PSC/SC (i.e. negotiable points).
The PSC/SC is more likely to be a win-win situation. It has more balanced fiscal terms which yields
an improved business climate.
CPE.7 Ring Fencing

A ring fence is a protection-based transfer of assets from one destination to another, usually
through the use of offshore accounting. A ring fence is meant to protect the assets from inclusion
in an investor's calculable net worth or to lower tax consequences.
In project asset valuation, ring-fencing occurs when a portion of a company's assets or profits are
financially separated without necessarily being operated as a separate entity. Reasons that
governments use ring fencing include:
• Regulatory reasons,
• Creating asset protection schemes with respect to financing arrangements, or
• Segregating into separate income streams for taxation purposes (cost recovery, tax
pools, etc.)
CAPE assumes that each asset has its own ring fence, so calculations are specific to that asset.
CPE.8 Typical Project Life Cycle

Economics is a continuous activity. Figure CPE-6 schematically illustrates the cumulative cash
flow for a typical E&P project over its life.
Figure CPE-6: Project Life Cycle
The first analysis normally concerns the exploration economics prior to any discovery. This may
consist of the analysis of a completely new venture, a bid on an acreage block, a review of the
prospects within currently held acreage, or the economics of a single prospect. The economic
analysis is required not only for deciding whether or not to proceed to invest, but also to rank
the opportunity with other opportunities. also, for currently held acreage the decision may be to
divest.
Once economically recoverable quantities of oil/gas have been discovered within a prospect,
economic analysis is required to decide whether further appraisal activities are desirable and, if
so, economically justified. The purpose of these appraisal activities, such as the drilling of one or
more appraisal wells or the acquisition of additional seismic data, is to reduce the uncertainty in
the prospect to a sufficiently low level as to enable proper development decision to be taken.
Once appraisal activities have been completed, the next decision is whether or not to commit
large funds to field development: The Final Investment Decision or FID. After the FID for field
development has been taken, economic analysis is still required. Field Development Plans can be
changed, and these changes require economic analysis. In this context, for example, acceleration
projects can be considered – investments to accelerate, but not necessarily increase, the
recovery of the reserves – or EOR (Enhanced Oil Recovery) projects.
Performing a “postmortem” or look back economic analysis, after the end of a project, enables
an evaluation of past investment decisions and to judge whether, in retrospect, the investment
made the return envisaged at the time of the FID. It can be of great value in highlighting where
consistent biases have been introduced in the past, with a view to avoiding these in the future.
This post investment review is often carried out by an integrated team.
Towards the end of a field’s life it will be necessary to decide when and how to decommission
(abandon) the project. Here again the economist will be required to analyze the alternatives.
Table CPE-1 presents general metrics to evaluate investment opportunities.
Table CPE-1: Investment Metrics
CPE.9 CAPE – Data Input

Figure CPE-7 shows the screen for entry of the general model parameters.
Figure CPE-7: Asset Economic Evaluation - Model Parameters
Data can be input manually; from the PE Tools database; or, from an Excel spreadsheet.
A file called “CAPE Input Data.xlsx” is included in the “CAPE Results File” directory. This file is used
to import parameters into CAPE from a spreadsheet. It is imperative that the format of this file is
not changed since CAPE assumes the data is available in a fixed location and format.
Using the data template ensures all the required data for CAPE is in the proper format and makes
input into CAPE simple and straightforward. Selecting each section of the CAPE Model will have
an option to read the data template file for importing of the appropriate data (Figure CPE-8). This
is accessed with the appropriate ‘Import ….’ button.
Figure CPE-8: Asset Economic Evaluation Data Input
The first time the Excel file is to be accessed, it must be linked to the CAPE tool using the ‘Link
Excel File’ button. Following this the Excel data is read simply by clicking the ‘Excel Import’ button.
It is also possible to import data from the PE Tools database (Figure CPE-9). To do this, the CAPE
Model data must have been stored to the PE Tools database from the appropriate CAPE Model
screen. The available models will be listed, check the model and ‘Load …” the data.
Figure CPE-9: Asset Economic Evaluation - PE Tools Database Data Input Option
CPE.10 CAPE – Model Parameters

Figure CPE-10 shows the entry of the general model parameters.
Figure CPE-10: Asset Economic Evaluation - Model Parameters
Oil and gas prices can be entered as a single value and then modified using the appropriate
escalation factor. Alternatively, an oil/gas price forecast can be loaded through an Excel file using
the “Import Price Data”. Note that oil and gas price forecasts are entered in $US regardless of the
currency used for analysis (refer to Section CPE.14).
It should be noted that all import tables function the same way. If a constant value is entered for
a specific parameter, then the table is populated with that constant. All cells in the table can be
edited so that specific values can be used for specific years. When a table is not made up of a
constant value, the box containing the relevant value will display a “-1”. The “-1” indicates that
the yearly data is variable.
The Model Parameters also includes the annual discount factor to be used for base NPV
calculations as well as the “Start Year” which is the year the economic analysis begins – the actual
production data can start later than the start year. If a start month greater than 1 is entered, the
cumulative production for the first year of production will be reduced. For look-back economics
that includes production history, the start of discounting can start at a later date.
Finally, the Model Parameters sheet includes the “Asset Name”. The asset name is defaulted but
can be changed to anything the user requires. Note, a “:” should not be used in the name and
will be removed if entered in the Asset Name box.
CPE.11 CAPE – Production Data

Figure CPE-11 shows the entry of production data.
Figure CPE-11: Asset Economic Evaluation - Production Data

It is possible to enter production data manually; import the data from an Excel spreadsheet; or,
import the data from a PE Tools database. The following PE² Essentials generated forecasts can
be imported from the PE Tools database into CAPE:
• PE² Essentials Unconventional Forecast Tool
• PE² Essentials Basic Reservoir Simulator Tool
• PE² Essentials StreamTube WaterFlood Tools
• PE² Essentials Misc/Immisc CO2 WAG WF Tool
• PE² Essentials Gas Material Balance Tool
• PE² Essentials Oil Material Balance Tool
• PE² Essentials Decline Curve Analysis Tool
• PE² Essentials Monte Carlo DC Forecast Tool
• PE² Essentials Field Development Planning Tool
If data is imported from an Excel file, the data can be saved to the PE Tools database.
It is possible to read a separately stored DCA database file and import forecasts or
history+forecasts of one or more of the assets in the DCA database (Figure CPE-12).
Figure CPE-12: Asset Economic Evaluation – Import DCA Production Forecast
When importing DCA assets, it is possible to delay the start-up of the asset as shown in Figure
CPE-12.
From the Data Import screen, it is also possible to clear the CAPE database or delete the asset
chosen on the main CAPE screen.
An asset can be easily duplicated by loading the asset, importing a new production forecast,
modifying the asset name on the ‘Asset Parameters’ page and then saving the asset with the new
forecast. The copied asset can then be imported into the CAPE database.
CPE.12 CAPE – Capital Cost

Figure CPE-13 shows the screen for entry of capital cost data. Project capital costs are entered as
unescalated costs.
Figure CPE-13: Asset Economic Evaluation – Import Unescalated Capital Costs

To enter well costs, the ‘Total Well Cost’ to drill and complete a well is entered. The annual well
cost will be added to the ‘Well Cost’ column as wells are added to the ‘Inc Well’ column. Figure
CPE-14 shows use of this input option.
Figure CPE-14: Asset Economic Evaluation – Well Scheduling and Costs
The second option is to manually enter the timing and number of incremental wells, as well as
the total cost for the incremental wells, in the table.
The abandonment costs are set-up by entering the ‘Total Abandonment Costs’; how many years
the cost is to be included and the start year for the abandonment costs (Figure CPE-14).
Note that a ‘-1’ in the input boxes indicates that the capital cost data was manually entered.
CPE.13 CAPE – Operating Cost

Figure CPE-15 shows the entry of operating cost data. Operating costs are entered as unescalated
costs.
Figure CPE-15: Asset Economic Evaluation – Import Operating Cost
Operating costs can be entered as a constant value and escalated by the escalation factors which
is entered on the Model Parameters page, or they can be entered for each year in the forecast.
When manual entry is used, the corresponding box is changed to “-1”
CPE.14 CAPE – Fiscal Parameters

The required fiscal parameters are dependent on whether the “Taxes and Royalties Regime”, the
“Production Sharing Contract” or the “Service Contract” is selected for the Fiscal Regime.
The Fiscal Parameters include an option to apply start-of-year, mid-year or year-end discounting.
Figure CPE-16 presents an example of the impact on the economics for different discounting options.
Figure CPE-16: Asset Economic Evaluation – Sensitivity to Discounting
Figure CPE-17 shows the entry of fiscal parameters.
Figure CPE-17: Asset Economic Evaluation – Fiscal Parameters
For the PSC tax regime, if Oil Royalty and/or Government Share is calculated based on cumulative
oil volume then the historical cumulative oil is entered in the ‘Cum at Start of Forecast (MM)’ box
so that the proper percentage is used in the forecast.
The required currency is entered on this page as well. The built-in currencies are $US, $CDN, GBP
and NOK. There is an option to add additional currencies in the drop-down menu. Any added
currency will be stored with the model. Once the currency is chosen the conversion rates can be
entered as a constant in the “Base Conv” box or entered on a yearly basis in the table.
It should be noted that, although CAPE will perform analysis in any currency, the oil and gas price
forecast (Section CPE.10) must be entered in $US. This was implemented based on the fact that
most, if not all, public oil/gas price forecasts, and sales of oil and gas, are normally based on $US.
CPE.15 CAPE – Sensitivities and Analysis Results

Once all the parameters are entered, the analysis is performed by clicking “Run Model”. After
performing the base analysis, a sensitivity analysis can be performed on production, CapEx, OpEx,
oil price and gas price. Sensitivity is performed at +/-30% in increments of 5% (Figure CPE-18).
Figure CPE-18: Asset Economic Evaluation – Sensitivities
A table of project and corporate economic results as well as a table of NPV results are presented
on the main page (Figure CPE-1).
Following analysis, it is possible to export the input/output results to an Excel file. CAPE uses
three Excel templates located in the “Bin\PE Essentials CAPE Libs” directory called
“TemplateSC.xlsx “, “TemplatePSC.xlsx” and “TemplateTRC.xlsx”.
The Excel file is generated by clicking “Export Results” on the main menu. This opens an export
page (Figure CPE-19) which allows the generation of an Excel file or a csv report file.
Figure CPE-19: Asset Economic Evaluation – Export Analysis Results
Checking “Save Corporate Results” and “Save to Excel File” will generate an Excel file containing
the analysis results and plots as well as the input parameters used to generate the results. This
file is saved in the “CAPE Results File” directory with a unique name so that many runs can be
made and saved without overwriting the previous files.
The Excel file contains all the economic run data and input data and automatically generates
tables and plots that are useful for viewing the results. Figures CPE-20 and CPE-21 show examples
of the results and plots included in the Excel file.
Figure CPE-20: Asset Economic Evaluation – Excel Results Plots
Figure CPE-21: Asset Economic Evaluation – Excel Results Tables/Plots

CPE.16 CAPE – Purchase Options

The Asset Economic Evaluation tool includes an option to evaluate the value of purchasing an
equity position in a company’s working interest. This option is accessed by the clicking ‘Purchase
Options’ on the main menu.
The target equity interest is entered in ‘Equity Purcahse (%)’ and a range of ‘Acquisition Cost’ for
the equity. The tool will plot the NPV at different discount rates (Figure CPE-22).
The IRR can also be plotted by clicking the ‘Plot IRR’ option (Figure CPE-23).
Figure CPE-22: Asset Economic Evaluation – Equity Purchase Options, NPV
After the analysis is completed, the results can be saved to a spreadsheet by clicking ‘Save Results
to Spreadsheet’. This file is saved in the ‘Asset Evaluation Results File’ directory with a unique
name so that many runs can be made and saved without overwriting the previous run results.
Figure CPE-23: Asset Economic Evaluation – Equity Purchase Options, IRR

502 General Appendix
General Appendix
The Appendix contains the following sections:
▪ Appendix 1 – Conversion Factors
▪ Appendix 2 – Workflow Examples

• W.1 – Granite Wash Well
• W.2 – North Sea Recompletion
• W.3 – Hydraulic Fracture Uplift
• W.4 – Sequential Wells vs Pad Drilling
• W-5 – Horizontal Frac/NoFrac Well Comparison
• W-6 – Simulated Well Sensitivities
• W-7 – Eagle Ford PDA Example
• W-8 – CAPE: Marcellus 6-Well Pad
▪ Appendix 3 – Decline Curve Models and Diagnostic Formulations
▪ Appendix 4 – Concepts of Production Data Analysis

Appendix 1 – Conversion Factors

Conversion Factors Conversion Factors
Input Conversion Output Property Input Conversion Output
Property
Units Operator Factor Units Variable Units Operator Factor Units
2
ft * 2.29567E-05 = acres barrels * 0.1364 = tonnes (metric)
2 3
acres * 43560.17 = ft barrels * 159 = m
2
m * 0.000247104 = acres barrels * 42 = US gallons
2
acres * 4046.87 = m Crude Oil US gallons * 0.00325 = tonnes (metric)
2 2 3
m * 0.000001 = km US gallons * 0.003785 = m
Area 2 2
km * 1000000 = m US gallons * 0.02381 = barrels
2
m * 0.0001 = ha barrels/day * 49.8 = tonnes/year
2 3
ha * 10000 = m 109sm * 35.3147 = Bscf
3
acres * 0.404687 = ha 109sm * 0.9 = Million Tonnes Oil Equivalent
3
ha * 2.47104 = acres 109sm * 0.74 = Million Tonnes LNG
9 3
cc/gm * 32.1958 = scf/ton 10 sm * 35.7 = TBTU
3
scf/ton * 0.03106 = cc/gm 109sm * 6.6 = MBOE
In-Situ Gas Content 3 3
m /t * 32.1958 = scf/ton Bscf * 0.028317 = 109sm
scf/ton * 0.03106 = m3/t Bscf * 0.025 = Million Tonnes Oil Equivalent
3
gm/cm * 62.4280 = lbm/ft Bscf * 0.021 = Million Tonnes LNG
3
lbm/ft * 0.0160185 = gm/cc Bscf * 1.01 = TBTU
Density 3 3
kg/m * 0.062428 = lbm/ft Bscf * 0.19 = MBOE
3 3 3
lbm/ft * 16.01845 = kg/m Million Tonnes Oil Equivalent * 1.11 = 109sm
(Density of Air = 1.222 kg/m3) Oil API * 141.5 / (API+131.5) = Oil Specific Gravity Million Tonnes Oil Equivalent * 39.2 = Bscf
(Density of Water = 1000 kg/m3) Oil Specific Gravity * 141.5/API - 131.5 = Oil API Million Tonnes Oil Equivalent * 0.82 = Million Tonnes LNG
3
Oil API * 141.5/(API+131.5)*62.4 = lbm/ft Million Tonnes Oil Equivalent * 39.7 = TBTU
3 3
stb/Mscf * 0.00561458 = m /m Natural Gas and LNG Million Tonnes Oil Equivalent * 7.33 = MBOE
3 3 3
m /m * 178.1076 = stb/Mscf Million Tonnes LNG * 1.36 = 109sm
Condensate Gas Ratio 3 3
stb/MMscf * 5.61E-06 = m /m Million Tonnes LNG * 48 = Bscf
3 3
m /m * 178107.6 = stb/MMscf Million Tonnes LNG * 1.22 = Million Tonnes Oil Equivalent
3 3
scf/stb * 0.1781076 = m /m Million Tonnes LNG * 48.6 = TBTU
Gas Oil Ratio 3 3
m /m * 5.61458 = scf/stb Million Tonnes LNG * 8.97 = MBOE
3
inch * 2.54 = cm TBTU * 0.028 = 109sm
cm * 0.3937 = inch TBTU * 0.99 = Bscf
Length inch * 25.4 = mm TBTU * 0.025 = Million Tonnes Oil Equivalent
ft * 0.3048 = m TBTU * 0.021 = Million Tonnes LNG
m * 3.2808 = ft TBTU * 0.18 = MBOE
3
bar * 14.5038 = psi MBOE * 0.15 = 109sm
psi * 0.0689474 = bar MBOE * 5.35 = Bscf
bar * 100 = kPa MBOE * 0.14 = Million Tonnes Oil Equivalent
kPa * 0.01 = bar MBOE * 0.11 = Million Tonnes LNG
Pressure
psi * 6.89476 = kPa MBOE * 5.41 = TBTU
kPa * 0.145038 = psi kilocalorie (kcal) * 4.187 = kJ
2
psi * 70.3077 = g/cm Kcal * 3.968 = BTU
2
g/cm * 0.0142232 = psi kJ * 0.23883 = kcal
psi/ft * 22.621 = kPa/m kJ * 0.948 = BTU
kPa/m * 0.044207 = psi/ft BTU * 0.25202 = kcal
psi/ft * 3.28084 = psi/m BTU * 1.05485 = kJ
Energy
psi/m * 0.3048 = psi/ft 1 kilowatt-hour (kWh) * 860 = kcal
3
psi/ft * 0.006944 = lb/ft kWh * 3600 = kJ
3
lb/ft * 144 = psi/ft kWh * 3412 = BTU
Pressure Gradient 3
psi/ft * 2306.682 = kg/m kcal * 0.001162791 = kWh
3
kg/m * 0.000433523 = psi/ft kJ * 0.000277778 = kWh
3
psi/ft * 2.306682 = gm/cm BTU * 0.000293083 = kWh
3
gm/cm * 0.433523 = psi/ft hp * 0.7457 = kW
Power
psi/ft * 19.23077 = ppg kW * 1.341 = hp
3
ppg * 0.052 = psi/ft acre-ft * 1233.48 = m
o * o
C (C * 9/5) + 32 = F BTU/ft-˚F * 0.5797 = W/m-˚K
o * o
F (F - 32) * 5/9 = C BTU/ft3-˚F * 14.911 = kj/m3-˚K
Temperature o * o
K 1.8 = R BTU/lb-˚F * 0.23885 = kj/kg-˚K
o * o
R 0.5555 = K BTU/scf * 0.037334 = MJ/sm3
3 3
m * 35.3147 = ft cp (gas) * 0.001 = μPa.s
3 3 2
ft * 0.0283168 = m in * 6.4516 = cm2
Miscellaneous
3
m * 6.29 = bbl lbf * 4.44822 = N
3
bbl * 0.158983 = m lbm * 0.453592 = kg
Volume 3
ft * 0.178107 = bbl miles * 1.609 = km
3 2
bbl * 5.6146 = ft miles * 2.58999 = km2
L * 0.00629 = bbl Section * 640 = Acres
bbl * 158.9825 = L Section * 258.999 = ha
1lb * 0.4536 = kg ton * 0.90748 = tonne
kg * 2.2046 = lb
lb * 0.000453593 = metric t
Weight metric t * 2204.62 = lb
metric t * 0.9072 = short ton
metric t * 1.016 = long ton
short t * 0.8929 = long ton
504 Appendix 2 – Workflow Examples
Appendix 2 – Workflow Examples
W.1 Granite Wash Well

The following information is excerpted from http://www.naturalgasintel.com/granitewashinfo.
The Granite Wash is a liquids-rich tight sand ~160 miles long and ~30 miles wide and covering
parts of Western Oklahoma and the Texas Panhandle (Figures W.1-1 and W.1-2).
Figure W.1-1: Granite Wash Counties in Oklahoma and Texas
Figure W.1-2: Granite Wash Location

The Granite Wash is one of the deeper unconventional formations in North America, lying at
depths between 10,000'-14,500'and is made up of a number of layered zones. These zones are
listed as "A", "B", etc., as shown in Figure W.1-3. Gas in the Granite Wash tends to be liquids-rich,
with natural gas liquids and condensate typically accounting for 30-40% of well production.
Figure W.1-3: Granite Wash Stratigraphy
The Granite Wash reportedly had 8.8 Tscf of technically recoverable natural gas as of January 1,
2013, according to the Energy Information Administration.
The purpose of this example is to evaluate the full cycle, as well as the point forward, economics
of a horizontal well completed in the Granite Wash and place on production in April 2010.
W.1.1 GW-01 Well

Data from a hydraulically fractured horizontal Granite Wash well, GW-01, (Figure W.1-4) has been
evaluated using the PE² Essentials software. The Excel file, ‘GW-01.xlsx’ contains the production
data from this well (“Workflow - Granite Wash Example”).
Figure W.1-4: Granite Wash Well (GW-01) Daily Production Data
Because of the scatter in the daily data, the data was averaged over a 1-week interval (Figure
W.1-5) and this averaged data was used for analysis.
Figure W.1-5: Granite Wash Well (GW-01) Weekly Production data

Figure W.1-6 is the wellbore and completion diagrams for this horizontal well.
Figure W.1-6: GW-01 Wellbore and Completion Diagrams

Figure W.1-7 shows the wellbore trajectory for the GW-01 horizontal well showing the
hydraulically fractured intervals.
Figure W.1-7: GW-01 Wellbore Trajectory
W.1.2 Gas Analysis

Figure W.1-8 indicates that the gas CGR has an average value of 16.4 bbls/mmscf.
Figure W.1-8: GW-01 - Condensate-Gas ratio
Figure W.1-9 lists the components measured on a surface sample (ref: 590 psi and 80˚F).
Figure W.1-9: GW-01 Laboratory Gas Analysis
The GW-01 laboratory gas analysis (SG=0.668, BTU=1133) was used as a starting point for history
matching the raw gas components to yield an average CGR of ~16.4 bbls/mmscf, using the PVT
model in the ‘Unconventional Forecast’ Tool. Figure W.1-10 presents the final estimation of raw
gas properties for the GW-01 well.
Figure W.1-10: GW-01 Estimated Raw Gas Analysis
These gas properties were used for analysis.
Prior to proceeding with a history match, the flowing THP data must be converted to BHP for
matching.
A wellbore model was built (Figure W.1-11) and the THP data was imported and converted to
BHP (Figure W.1-12). The BHP data was then transferred to the ‘GW-01.xlsx’ file for import into
the history matching tool.
Figure W.1-11: GW-01 Wellbore Model
Figure W.1-12: GW-01 THP to BHP

W.1.3 Production History Match

Using the ‘Unconventional Forecast’ tool, the production data was history matched in order to
estimate reservoir parameters. For this exercise, the 7-day averaged data was used.
The Numerical Model was used to generate the initial history match. The base model that was
built for this reservoir is ‘PE_Essentials_Unconventional_GW-01_Base.dvx’ located in the
“Workflow - Granite Wash Example\Unconventional” directory.
The weekly averaged historical production data was imported into the History Match model
(Figure W.1-13). A preliminary history match was performed using the analytical model (Figure
W.1.14).
Figure W.1-13: GW-01 Import of Production History
Figure W.1-14: GW-01 Unmatched and Matched History, Analytical Model

The history match shown in Figure W.1-14 (right) is an acceptable match. But, since a numerical
model more rigorously models the reservoir, it is preferred for long term forecasting and as a
result the numerical model was history match as well.
Note that history matching with the numerical model is time consuming and may or may not
yield a more accurate solution. As is the case for all history matching; there are different solutions
that appear to be equally valid. The final history match, by any technique, is not necessarily the
most accurate solution, but it is sufficient to match the historical data.
The parameters obtained from the match using the analytical model were refined in the
numerical model and the comparison of the analytical (left) and the numerical (right) history
match results are presented in Figure W.1-15.
Figure W.1-15: GW-01 Production History Match – Analytical and Numerical Models
The numerical history match model is saved as ‘PE_Essentials_Unconventional_GW-

01_NumericalMatch.dvx’ and this model will be used for long term forecasting. The main
difference between the analytical and the numerical history match models is that the numerical
model has a slightly lower permeability than the analytical model.
Figure W.1-16 presents the final history matched reservoir parameters; Figure W.1-17 presents
the final hydraulic fracture parameters generated for the history match; and Figure W.1-18
presents the well parameters. These figures include the analytical model match parameters as
well for comparison.
Figure W.1-16: GW-01 History Match – Reservoir Parameters (Left-Analytical, Right-Numerical)
Figure W.1-17: GW-01 History Match – Fracture Parameters (Left-Analytical, Right-Numerical)
Figure W.1-18: GW-01 – Wellbore Parameters (Left-Analytical, Right-Numerical)
Figures W.1-19 and W.1-20 present the forecast results using the history match parameters
from the numerical model.
Figure W.1-19: GW-01 – Unconventional Forecast
Based on initialization of the numerical simulator, GIIP was calculated to be 2.887 Bscf and 50-
year recovery factor was calculated to be 95.4%. It should be noted that the history data file used
for plotting on Figure W.1-20 was generated with the PE² Essentials DCA tool.
Figure W.1-20: GW-01 – Unconventional Gas Rate Forecast

W.1.4 Monte Carlo Volumetrics: Gas Volume

Using the ‘Monte Carlo Volumetrics’ tool with the reservoir parameters presented in Section
W.1.3, a Monte Carlo simulation was run to estimate gas volume for P90, P50, P10 and expected
value (EV) cases (Figure W.1-21). The expected value for GIIP is 2.74 Tscf with an expected
recovery of 2.55 Tscf (RF = 93.1%) for this well.
Figure W.1-21: GW-01 – Monte Carlo Gas Volumetrics
The results of this analysis were used to direct a ‘Monte Carlo DC Forecast’ analysis.
W.1.5 Decline Curve Analysis

Using the ‘Decline Curve Analysis’ tool, the DCA parameters were determined using the average
weekly gas production rates (Figure W.1-22).
It is possible to generate different solutions for the DCA parameters depending on the low/high
ranges and ‘X Range’ entered for the analysis. This analysis was generated based on the
assumption that the GW-01 well will exhibit a ‘Super Hyperbolic’ decline which is common for
unconventional reservoirs.
Figure W.1-22: GW-01 – DCA Using Weekly Averaged Rates
Figure W.1-23 shows that DCA using the daily and monthly rates yields results that are similar to
the weekly rate DCA. The difficulty with using the daily rates is that, since the data is very
scattered, a proper analysis is difficult to determine.
Figure W.1-23: GW-01 – DCA Using Daily and Monthly Rates
Figure W.1-24 presents the supplementary DCA plots for the weekly data analysis.
Figure W.1-24: GW-01 – Supplementary DCA Plots (Weekly Data)
The DCA results were entered as P50 in the ‘Monte Carlo DC Forecast’ tool. Ranges for P10/P90
were entered and history data, which was generated from the DCA tool using ‘Save History’, was
imported to generate probabilistic forecasts (Figure W.1-25).
Figure W.1-25: GW-01 – Monte Carlo DC Parameter Generation and Forecast

It is possible to generate a number of different solutions for the Monte Carlo DC parameters
depending on whether or not a constant ‘seed’ is used (refer to Section 2.4). This analysis
represents just one realization for the P90/P50/P10/EV decline parameters.
In order to generate consistent results, the ‘Initial Daily Flow Rate’ and ‘Annual Nominal Decline
Factor’ were entered as constant values. This will limit the Monte Carlo analysis to generation of
b, Dlim and Qf values. After generating the DC parameters, they can be saved by clicking ‘Save
Simulation’. This will store the different realizations which can then be imported into the ‘Decline
Curve Analysis’ tool to generate the deterministic DC forecasts.
The Monte Carlo simulation can also be performed without importing history (Figure W.1-26).
Figure W.1-26: GW-01 – Monte Carlo DC Parameter and Forecast Generation
The advantage of using the ‘Decline Curve Analysis’ tool to generate the deterministic DCA
forecasts is that the water and condensate rates will also be generated for economics analysis
purposes.
The Monte Carlo parameters were imported into the ‘Decline Curve Analysis’ tool and the
different forecasts were generated with the results shown in Figure W.1-27.
Note that in order to save all the forecasts in the DCA database, copies of the base well (weekly
data) were generated for each realization. All forecasts were saved to CSV files.
Figure W.1-27: GW-01 – DCA Forecasts for P90/P50/P10/EV
Figure W.1-28 presents the remaining gas rate and cum gas forecasts and Figure W.1-29 presents
the full cycle well forecasts for the GW-01 well.
Figure W.1-28: GW-01 – Comparative Remaining Forecasts for P90/P50/P10/EV

Figure W.1-29: GW-01 – Comparative Full Cycle Forecasts for P90/P50/P10/EV
The forecasts yielded the following results:

P90: Total EUR = 2358 mmscf
Remaining Gas = 1128 mmscf


EV: Total EUR = 2711 mmscf

These forecasts were then imported into the ‘Scoping Economics’ tool to evaluate the full cycle
economics of the well as well as the current economics for the well.
W.1.6 Scoping Economics

The ‘Scoping Economics’ tool was used to evaluate the economic value of the GW-01 well in
terms of the remaining value of the well and the full-cycle economics of the well (look-back
evaluation).
In order to perform full cycle economics, the historical gas and oil prices are required. The Excel
file, ‘Historical and Forecasted Oil and Gas Prices.xlsx’, located in the “Example Input Files\Excel
Files” directory was used to import historical prices from August 2008 as well as import a price
forecast. Figure W.1-30 shows the gas prices used for the remaining value (Forecast Gas Price)
and the full cycle (Historical / Forecast Gas Price) economic analysis.
Figure W.1-30: GW-01 – Gas Price Data
The assumptions used for the point-forward economic run are: overhead cost for the well is
$100k/year; variable well cost is $0.76/mscf; gas transportation fee is $0.50/mscf: and water
disposal cost is $0.25/bbl. For full-cycle economics, a 2010 well completion cost of $4.5mm was
assumed. No escalations were applied for either forecast.
Figure W.1-31 shows the import screens for the oil and gas price forecasts. The spreadsheet was
‘‘Historical and Forecasted Oil and Gas Prices.xlsx’ and the left sheet is the import for the
incremental forecast, or point forward, economics and the right sheet is for the look back, or full
cycle, economics.
Figure W.1-31: GW-01 – Oil/Gas Price History/Forecast
Note that discounting for the full cycle economics starts after 6.75 years which is equivalent to
January 2017 and 2 years for the incremental forecast. The discounting is delayed in the
incremental forecast since it starts in January 2015 (end of history is December 2014).
Figures W.1-32 and W.1-33 show the incremental economic runs and Figures W.1-34 and W.1-
35 shows the full cycle economic runs.
Figure W.1-32: GW-01 – Point Forward Economic Runs
Figure W.1-33: GW-01 – Point Forward Discounted Cum Cash Flows

Figure W.1-34: GW-01 – Look Back (Full Cycle) Economic Runs
Figure W.1-35: GW-01 – Point Forward Discounted Cum Cash Flow

As is the case for most wells, the point forward economics of the GW-01 well is robust with an
NPV of $2.4mm at the P50 level. Since the GW-01 well was placed on production in 2010, during
a period of low gas prices, look back economics (P50) yields an IRR of 6% and a payout of 11 years
(2021).
Under 2017 economic conditions, this well may not be economic – current well costs for a Granite
Wash well are in the $7 - $8 million range and gas price forecasts remain low.
W.1.7 Material Balance Type Curve Generation

The ‘Gas Material Balance’ tool can be used to generate type curves for a given well or area. In
most cases, especially for unconventional reservoirs, depleting reservoir pressure is not known.
Without reservoir pressure, it is still possible to build a multi-tank model to history match the
production characteristics of the well using the ‘Low Perm Material Balance’ tool.
There is no reservoir pressure data available for GW-01 so only cumulative production is available
for analysis (Refer to Figure W.1-36).
Figure W.1-36: GW-01 – Multi-Tank Gas MB Model Import Production Data
The multi-tank GW-01 model is comprised of a 0.5 Bscf tank communicating with a 6 Bscf tank.
Figure W.1-37 shows the parameters used to build a multi-tank material balance model of the
GW-01 well. The reservoir and well parameters were obtained from the data presented in Section
W.1.3. Figure W.1.38 shows the history match of the production data.
Figure W.1-37: GW-01 – Multi-Tank Gas Material Balance Model
Figure W.1-38: GW-01 – Multi-Tank MB Gas Model Production History Match / Forecast
To use this multi-tank model as a forecast generator for a future well, the well and tank
parameters can be modified, and a new forecast is generated using the future well parameters.
W.2 North Sea Well Recompletion

A North Sea well stopped producing because of wellbore issues and will require a workover
costing $5mm to return the well to production. The well is a late field life well and produces with
a high water cut. The problem is to determine if it is economically justifiable to perform the
workover.
Figure W.2-1 shows the historical production from the well prior to shut-in.
Figure W.2-1: NS-01 – Production History
The economic parameters for this analysis are as follows:

• Well workover cost: US$5mm
• Well overhead: US$2mm/yr
• Fixed well costs: US$250k/yr
• Variable well costs: US$31.45/m3 ($5/bbl)
• Transportation costs: US$12.58/m3 ($2/bbl)
• Water disposal (injection costs): US$3.15/m3 ($0.5/bbl)
• Gas processing fee: 0 (flared)
• Annual discount rate: 10%
• No price escalation
W.2.1 Decline Curve Analysis

Using the ‘Decline Curve Analysis’ tool, the DCA parameters were determined for the daily oil
production rates (Figure W.2-2 and Figure W.2-3).
Figure W.2-2: NS-01 – Import Production Data
Figure W.2-3: NS-01 – Decline Curve Analysis

A high hyperbolic exponent was expected for this analysis since water injection was occurring at
a reported voidage replacement ratio of 1.0.
The supplementary plots for this analysis are presented in Figure W.2-4.
Figure W.2-4: NS-01 – Decline Curve Analysis – Supplementary Plots
The ‘Monte Carlo DC Forecast’ tool was used to generate P10/P50/P90/EV parameters for use in
the DCA tool (Figure W.2-5). A constant seed value was entered so the results could be
regenerated in the future
The DCA results were used for the P50 input parameters.
Figure W.2-5: NS-01 – Monte Carlo DC Simulation Results
Figure W.2-6 presents the Monte Carlo DC simulation results.
Figure W.2-6: NS-01 – Monte Carlo DC Simulation Results

The Monte Carlo DC results were transferred to the DCA tool and forecasts were generated. To
generate the water production forecast, the ‘Use WOR’ option was implemented. The oil cut
option could also have been used. Figure W.2-7 is a comparison of the oil rate, cumulative oil,
GOR and water cut forecast results for the NS-01 well and Figure W.2-8 presents the final
volumes.
Figure W.2-7: NS-01 – Forecast Results
Figure W.2-8: NS-01 – Volumetric Results

W.2.2 Scoping Economics

Using the economic parameters and the DCA forecasts, the ‘Scoping Economics’ tool was used to
evaluate the viability of the workover. Figure W.2-9 presents the economic runs for each case
and includes a plot of the oil and gas price forecast used in the analysis.
Figure W.2-9: NS-01 – Economic Evaluation
NPV (at 10%) are as follows:

• P90: US$10.0mm
• P50: US$20.4mm
• P10: US$27.0mm
Remaining economic reserves are:

• P90: 118.2 x 103 m3
• P50: 194.4 x 103 m3
• P10: 235.1 x 103 m3
Economics are based on the assumption that the well will be shut in as soon as the annual cash
flow goes negative (Figure W.2-10).
Figure W.2-10: NS-01 – Discounted Annual Cash Flow
Time to payout the recompletion costs are:

• P90: 10 months
• P50: 10 months
• P10: 9 months
Based on all the parameters included in this analysis, the economics for the recompletion of the
NS-01 well are very robust. The recompletion of this well should proceed.
W.3 Hydraulic Fracture Uplift

A single stage simulation model of a horizontal well (500 ft stage by 1000 ft well spacing) was
built using the ‘PE² Essentials Basic Reservoir Simulator’. A base 10-year forecast was generated
then a hydraulic fracture was added to the model. Another 10-year forecast was generated and
the economics of fracturing the well was evaluated.
The simulator files are ‘PE_Essentials_BasicReservoirSimulator_Base Horizontal Gas Well.dvx’

and ‘PE_Essentials_BasicReservoirSimulator_Horizontal Gas Well, 1-Fracture.dvx’ and are
located in the “Workflow - Hydraulic Fracture Uplift\Simulator” directory.
The economic file is ‘PE_Essentials_Economics FractureUplift.dvx’ and is located in the

“Workflow - Hydraulic Fracture Uplift\Economics” directory. The 2016+ oil and gas price forecast
was imported from the ‘Historical and Forecasted Oil and Gas Prices.xlsx’ file located in the
“Example Input Files\Excel Files” directory.
The model grid is 15 x 19 x 3 with x-gridding that includes a 2 ft block in the center of the grid for
fracture placement (Figure W.3-1).
Figure W.3-1: Model Grid – Single Stage Hydraulic Fracture
To place a hydraulic fracture in the well, the permeability in the grids from (8, 5) to (8, 15) was
increased to 5 md (Figure W.3-2). This is equivalent to a fracture permeability of 480md assuming
that the fracture width is 0.25”. This is calculated by kh = kfWf where k is 5md, h is 2ft and Wf is
0.25” (0.0208ft).
Figure W.3-2: Hydraulic Fracture Input Parameters
The reservoir parameters, fluid parameters, relative permeability and wellbore parameters are
presented in Figures W.3-3 to W.3-6.
Figure W.3-3: Reservoir Parameters

Figure W.3-4: Gas PVT Parameters
Figure W.3-5: Relative Permeability Parameters

Figure W.3-6: Horizontal Wellbore Parameters
The lateral section of the well was placed in the center of the zone (layer 2). The entire lateral,
except for block 1 and 15 were open to flow at a minimum BHP of 500 psi (Figure W.3-7).
Figure W.3-7: Well Schedule

The modeled fracture stage encompasses 11.5 acres and contains 1.7 Bscf of gas in place. Figure
W.3-8 shows the production forecast for the unfractured and fractured cases.
Figure W.3-8: Forecast Comparison – Unfractured vs Fractured
Two economic runs were made: unfractured (Figure W.3-9) and fractured Figure W.3-10).
Figure W.3-9: Economic Analysis – Unfractured Case

Figure W.3-10: Economic Analysis – Fractured Case
The incremental cost assumption for the fractured case is that the cost of a single hydraulic
fracture stage would be $750k. Both cases include $100k for shared incremental costs for the
completion and tie-in of the well.
A comparison of the economic results is shown in Figure W.3-11.
Figure W.3-11: Economic Results – Comparison
From this analysis, one hydraulic fracture stage will deliver an incremental 612 mmscf of gas over
ten years; will take an additional 7 months to pay out the fracture costs; and delivers an extra
$1.6million of NPV.
W.4 FDP: Sequential Wells vs Pad Drilling

Development of an unconventional oil field is going to be expanded with the addition of 24 wells.
The operator has two rigs currently available and needs to decide whether to drill 4 pads with 6
wells on each pad or to sequentially drill each well.
For pad drilling, the two rigs would each drill one pad which is then placed on production and
then the rigs would drill the third and fourth pads. In this scenario 12 wells would be placed on
production initially, but up-front capital costs would be significant. For sequential drilling, the
production rate would increase at a slower pace, but the up-front capital costs would be
significantly reduced. To account for the delay in production start-up while the pads are being
drilled a carrying cost of 5% will be added to the up-front capital cost.
For this example, the following assumptions were made:

• Statistical variations are included on the well performance parameters
• Facility costs are the same for either scenario
• Carrying costs for the pads is 5% of the capital to drill the first 2 pads
• Pad completion costs have a savings of 10% over sequential completions
• Additional $50k/well to tie-in sequential wells to common facilities
• All other economic parameters are the same
• Facility uptime is a constant 95%
• No limits on maximum facility rate capacity
• Rig moves take 5 days for sequential drilling
W.4.1 Production Type Curve

Using the ‘Decline Curve Analysis’ tool, the DCA parameters were determined for the production
type curve which was available for the project (Figure W.4-1).
Figure W.4-1: Unconventional Oil – Production Type Curve

Figure W.4-2 shows the decline curve analysis results which were then imported into the
Development Planning tool (Figure W.4-3) to generate the normalized type curve.
Figure W.4-2: Unconventional Oil – DCA Results
Figure W.4-3: Unconventional Oil – Type Curve

W.4.2 Project Forecast

Figure W.4-4 shows the inputs to the Field Development Planning tool for the two scenarios.
Figure W.4-4: Development Planning Tool – Sequential Drilling vs Pad Drilling
For both scenarios, the statistical variations were enabled for the production parameters. In both
models, the Seed was set to be the same so the well productivity variations would be the same.
Figure W.4-5 shows the statistical parameters used for both scenarios.
Figure W.4-5: Development Planning Tool – Statistical Parameters

The forecasts are shown in Figure W.4-6 (sequential drilling) and Figure W.4-7 (pad drilling).
Figure W.4-6: Development Planning Tool – Sequential Drilling Forecast
Figure W.4-7: Development Planning Tool – Pad Drilling Forecast

The forecast results are similar (Figure W.4-8) but the peak rates are significantly different.
Figure W.4-8: Development Planning Tool – Forecast Results (Top: Sequential, Bottom: Pad)
Figures W.4-8 and W.4-9 show that the main differences between the two scenarios are the peak
oil rate (7992 bopd versus 2965 bopd), and the oil rate during the first year of production.
Figure W.4-9: Development Planning Tool – Forecast Comparison
The economic advantage of the rate acceleration that occurs with pad drilling may be offset by
the carrying costs of the pad drilling capital.
W.4.3 Project Economic Comparison

The economic difference between the two scenarios is assumed to be in the capital costs and the
early rate profile. There should be cost savings associated with pad drilling and these are modeled
as a 10% reduction in completion costs as shown in Figure W.4-10.
Figure W.4-10: Scoping Economics Tool – Capital Comparison
The pad drilling economic model also included a ‘Pre-Startup Project Cost’ of $1.527million which
represents a 5% carrying cost on the $30.552million required to pre-drill the first two pads
($2.546million/well * 12wells).
The economic analysis results (NPV10) of the two scenarios is presented in Figure W.4-11.
Figure W.4-11: Scoping Economics Tool – Economic Comparison
Overall, the economics are similar: IRR for the sequential scenario is 27% and IRR for the pad
scenario is 28.7%. One potentially significant difference between the two scenarios is that the
pad drilling scenario has a 3-year payout of all its costs, including carrying costs, versus a 4-year
payout for the sequential drilling scenario. This could be significant because of the rapid decline
in oil rate evident from the type curve – the earlier payout would reduce the project risk.
For information purposes, Figure W.4-12 shows the economic run of the pad drilling scenario.
Figure W.4-12: Scoping Economics Tool – Economic Analysis of the Pad Drilling Scenario
All files used in this analysis are located in the “Workflow\Unconventional Oil Example” directory.
W.5 Multi-Frac/NoFrac Gas Well

NOTE - The grid input limitations for the current version of the PE² Essentials Basic Reservoir
Simulator will allow only 7 hydraulic fractures to be included in a model, assuming 1 grid block
separating the fracture grid. This is not a limitation of the simulator but of the PE² Essentials input
interface. This was designed because an IMPES simulator is not well suited to massive, complex
models. It would be better to model a single fracture stage and upscale the results as done in the
PE² Essentials Unconventional Forecast tool and as demonstrated in Appendix W.3. Nevertheless,
as an exercise, a horizontal gas well containing 7 hydraulic fractures was built and forecast results
were compared to an unfractured horizontal well.
For this exercise, the simulator files are located in the “Workflow - Multi-Frac Horizontal Well”
directory. The two models have the same grid structure.
‘PE_Essentials_BasicReservoirSimulator_No Fracs, k=0.007.dvx’ is the base model and
‘PE_Essentials_BasicReservoirSimulator_7 Fracs, k=0.007 FCD=2.4, xf=300ft.dvx’ is the fractured
well model. A facture design model, ‘PE_Essentials_Hydraulic_Fracture_Analysis.dvx’, was used
to determine fracture parameters.
The horizontal gas well model has a 29x19x1 grid (Figure W.5-1) with uniform permeability of
0.007md and porosity of 0.06.
Figure W.5-1: Basic Reservoir Simulator – Model Grid
The grid was set up with three grid blocks between the fracture grids.
The PE² Essentials ‘Hydraulic Fracture Design’ tool was used to evaluate fracture parameters for
this model (Figure A2-5.2).
Figure W.5-2: Hydraulic Fracture Design – Multi-Frac Well
The fracture design indicated that for a fracture half length, xf, of ~300ft, the optimum FCD would
be 2.4. For IMPES stability, the model’s fracture grid width (dy) was set to 2ft so the equivalent
grid permeability for a 300ft model fracture would be 2.5md as calculated below.
kf wf = FCD k xf
kf = (2.4)(0.007)(300) / 2
kf = 2.5md
Figure W.5-3 shows the input fracture parameters for the model.
Figure W.5-3: Basic Reservoir Simulator – Model Fractures
Figure W.5-4 shows the x-y grid and Figure W.5-5 shows the permeability distribution for the
multi-frac well.
Figure W.5-4: Basic Reservoir Simulator – Model Grid

Figure W.5-5: Basic Reservoir Simulator – Fracture Model permeability
The reservoir parameters for the two models are the same except for the resulting average
permeability (Figure W.5-6).
Figure W.5-6: Reservoir Parameters (Left: No Fracture, Right: Fractured)
The lateral section of the well was placed in the center of the layer (Figure W.5-7). The entire
lateral, except for block 1 and 29 were open to flow at a minimum BHP of 500 psi (Figure W.5-8).
Figure W.5-7: Basic Reservoir Simulator – Well Model
Figure W.5-8: Basic Reservoir Simulator – Well Schedule
A 10-year forecast was generated for each model incorporating the ‘Save Simulation Run Data’
option for plotting with PE² Essentials Chart. The forecast results (Figures W.5-9 and W.5-10)
indicate that the fractured well model recovered 82.3% of the gas in place versus 55.2% for the
unfractured well model.
Figure W.5-9: Basic Reservoir Simulator – Non-Fractured Well Forecast
Figure W.5-10: Basic Reservoir Simulator – Fractured Well Forecast

Figures W.5-11 and W.5-12 show a comparison of the pressure profile along the wellbore at 102
days and at the end of the forecast, respectively. The top plot is for the unfractured well and the
lower plot is for the fractured well.
Figure W.5-11: Basic Reservoir Simulator – Wellbore Grid Pressure, 102 days
Figure W.5-12: Basic Reservoir Simulator – Wellbore Grid Pressure, 3650 days
Figure W.5-12 indicates that the reservoir grid pressure along the lateral at the end of the
forecast was significantly less for the fractured well than for the unfractured well, as expected.
PE² Essentials Chart was used to compare the two forecasts. The run files were imported into
Chart and the two forecasts plotted for comparison (Figure W.5-13).
Figure W.5-13: Chart – Rate Comparison, Fractured vs Non-Fractured Well
Figure W.5-13 shows some of the rate acceleration component that is evident in all fractured
wells. This is apparent by the gas rate for the fractured well falling below the gas rate for the
unfractured well after 9 years of production.
W.6 Single-Well Simulation Models

A single well simulation model can be used to look at a number of sensitivities associated with
production performance and recovery characteristics of various parameters. This section
presents a number of models and evaluates the sensitivity of one parameter for each model.
Except for GOC and OWC, all models have the same rock and fluid characteristics as shown in
Figure W.6-1.
Figure W.6-1: General Reservoir and Fluid Parameters
The grids were set appropriately for each sensitivity case.
The models used in this section are included in the “Workflow - Single Well Sensitivity Models”
directory.
W.6.1 Vertical Oil Well – Drainage Area Sensitivity

The production performance of a vertical well was evaluated to evaluate the effects of drainage
area. This type of study may be useful in the case of a new discovery with limited available data.
A 15x15x5 pseudo-radial grid was built for an oil well in the center of the field. The well was
completed in all five layers. There is no oil/water contact and the reservoir was initially above the
bubble point pressure (no gas cap). This comparison varied only the drainage area from 160 acres
to 640 acres (refer to Figures W.6-2 and W.6-3).
Figure W.6-2: Reservoir Grid – Vertical Oil Well, 160 Acre Drainage
Figure W.6-3: Reservoir Grid – Vertical Oil Well, 640 Acre Drainage
The vertical well location and well parameters are shown in Figure W.6-4
Figure W.6-4: Wellbore Parameters – Vertical Oil Well, Drainage Area Comparison
OIIP after initialization and the 10-year recovery for the two models are shown in Figure W.6-5.
Figure W.6-5: Vertical Well Model – Initialization and 10 Year Forecast

Figures W.6-6 presents a comparison of the forecasts for the two models. The middle curves on
the graph show the GOR forecasts. As expected, the GOR for the smaller drainage area increases
faster than for the larger drainage area because the reservoir pressure (upper two curves) falls
faster for the smaller area.
Figure W.6-6: Vertical Well Model – 10-Year Forecast (Reservoir Pressure, GOR, Oil Rate)
Although the larger drainage area allows for higher rates and slower decline in overall reservoir
pressure. The recovery factor after 10 years was only 4.1% for the larger drainage area versus
7.6% for the smaller drainage area. This indicates that one oil well in a 640-acre drainage area
may not be optimum for recovery.
W.6.2 Vertical Oil Well – Water Coning/Completion Interval Comparison

An oil/water contact (OWC) at the top of the fourth layer was added to the 160 acres, vertical oil
well model used in Section W.6.1 to evaluate the effect of different well completions.
This sensitivity compares the forecast for three cases; the well completed in the top three layers;
the well completed in the top four layers; and the well completed in all five layers. The well’s oil
rate was set at 200 bopd for all cases and the recovery and well performances were compared.
For this case, the OIIP was 1.98 mmbbls and WIIP was 2.76 mmbbls. Figure W.6-7 shows the
reservoir properties with the addition of the OWC.
Figure W.6-7: Vertical Well Model – Addition of OWC at Top of Layer 4
Figure W.6-8 shows the well location in the grid for the case of the well completed in the top
three oil bearing layers.
Figure W.6-8: Vertical Well Model with OWC – Well Location, Completion in Oil Layers
Figure W.6-9 presents the 10-year forecast results for oil rate and reservoir pressure and Figure
W.6-10 presents the GOR and water cut.
Figure W.6-9: Vertical Well Model with OWC – 10-Year Forecast (Reservoir Pressure, Oil Rate)
Figure W.6-10: Vertical Well Model with OWC – 10-Year Forecast (GOR and Water Cut)
Oil rate for the wells completed into the water layer are higher than oil rate for the well
completed in just the oil layers. This is the result of slowing water encroachment into the oil layer
by reducing pressure in the water zone. The GOR is highest in the well completed in all layers
because of pressure depletion resulting from the higher fluid (oil+water) recovery.
Figures W.6-11 to W.6-13 show the water saturation after 10 years at the well location in layer
3 which is the lowermost oil bearing layer. Water encroachment is significanlty reduced when
the water layers are included in the completion.
Figure W.6-11: Vertical Well Model with OWC – Water Saturation (Layer 1-3 Completion)
Other parameters such as the impact of vertical permeability and the impact of the relative
permeability curves (end points and curvature), could also be evaluated with this model. The
overall goal would be to optimize recovery with minimal water/gas coning. This sort of analysis
can be important if production facilities have limited water handling capacity or if water disposal
is costly.
W.6.3 Horizontal Oil Well – Lateral Length Comparison

Production performance of a horizontal well can be compared to evaluate the effects of lateral
length, zone thickness, vertical-to-horizontal permeability, and the position of the lateral in the
pay interval, as well as vertical distance from an oil/water or gas/oil contact.
A simple grid (Figure W.6-14) was constructed to evaluate the impact of a number of lateral
lengths. This model has grid dimensions of 21x21x5, with each cell 530ft by 530ft by 5ft in length,
width, and thickness. The different wellbore models are shown in Figure W.6-15
Figure W.6-14: Reservoir Grid – Horizontal Oil Well, Lateral Length Comparison
Figure W.6-15: Wellbore Model – Horizontal Oil Well, Lateral Lengths: 1590ft, 3710ft, 5830ft
An oil-water contact (OWC) was placed at the top of the fourth layer and the well was placed in
the center of the second layer. All laterals were centered in the grid model as shown in Figure
W.6-16.
Figure W.6-16: Horizontal Oil Well – Location of the 5830ft Lateral
Figures W.6-17 to W.6-19 show the pressure profile along the lateral at the end of the 10-year
forecast.
Figure W.6-17: Horizontal Oil Well – 10-Year Forecast, Pressure Along 1590ft Lateral
Figures W.6-20 to W.6-22 show the water saturation profile along the lateral at the end of the
10-year forecast. This quantifies the magnitude of the water encroachment into the well.
Figure W.6-20: Horizontal Oil Well – 10-Year Forecast, Water Saturation Along 1590ft Lateral
Figure W.6-23 shows the reservoir pressure and the oil rate performance for the different lateral
lengths and Figure W.6-24 shows the GOR and Water Cut performance.
Figure W.6-23: Horizontal Oil Well – 10-Year Forecast (Reservoir Pressure, Oil Rate)
Figure W.6-24: Horizontal Oil Well – 10-Year Forecast (GOR, Water Cut)
The 10-year oil recovery was 2.8% for the 1590ft lateral, 3.9% for the 3710ft lateral and 4.7% for
the 5830ft lateral. As expected, recovery increases as lateral length increases - all other factors
remaining the same.
Water breakthrough and water cut responses were similar for all lateral lengths. GOR was slightly
variable for the different lateral lengths and can be attributed to the higher reservoir pressure
depletion for the longer laterals, resulting in higher recovery and pressure depletion.
The saturations in each layer at the end of the 10-year forecast for the 5830ft lateral are
presented below. Figure W.6-25 shows the gas saturation in the top layer (above the wellbore)
indicating the formation of a secondary gas cap. Figure W.6-26 shows the water saturation along
the lateral indicating water encroachment into the well layer and Figure W.6-27 shows the water
saturation below the lateral. Figure W.6-28 shows the oil saturation in the fourth layer, which
was initially water saturated (Sw=100%), indicating the loss of oil into the aquifer because of
pressure depletion and the formation of the secondary gas cap. Finally, Figure W.6-29 shows the
oil saturation in the bottom layer.
Figure W.6-25: Horizontal Oil Well – 10-Year Forecast, Gas Saturation, Layer 1 (Secondary Gas
Cap)
Figure W.6-26: Horizontal Oil Well – 10-Year Forecast, Water Saturation, Layer 2 (Water
Encroachment)
Figure W.6-27: Horizontal Oil Well – 10-Year Forecast, Water Saturation, Layer 3 (Water
Encroachment)
Figure W.6-28: Horizontal Oil Well – 10-Year Forecast, Oil Saturation, Layer 4 (Oil Movement
into Water)
Figure W.6-29: Horizontal Oil Well – 10-Year Forecast, Oil Saturation, Layer 5 (Minor Oil
Movement into Water)
Most of the oil movement into the aquifer could be stopped with water injection to maintain
pressure in the reservoir.
W.6.4 Horizontal Oil Well – Fracture Spacing/Stage Size

The cost of completing a well with hydraulic fractures is dependent on the number of fracture
stages placed in the well. The number of stages in a given lateral is a function of the size of each
stage and the length of the lateral. A fracture stage can be a couple of hundred feet to as low as
50ft.
To evaluate the recovery sensitivity to the size of stages in a horizontal hydraulically fractured
well, the well model presented in Section W.5 was converted to an oil well with no water or gas
contacts and used for this analysis.
The model has a 29x19x1 grid with 7 hydraulic fractures. The fractures are placed so that there
are 3 grid blocks between the fractures. For this analysis, the length of the fracture stage is
determined by the total length of the three blocks between the fractures.
Seven models were used for this sensitivity analysis and are included in the “Workflow\Single
Well Sensitivity Models” directory.
Table W.6-1-1 lists the parameters for the seven cases built for this analysis. The table also
includes the 10-year recovery factor for each case.
10 Year
Lateral
Stage OIIP Recovery
Length
(ft) (mbbls) Factor
(ft)
(%)
450 1102 3164 34.4
375 893 2564 35.0
300 736 2114 35.6
225 553 1589 36.5
150 370 1064 37.8
75 188 539 40.2
45 114 329 42.0
Table W.6-1: Sensitivity Cases – Fracture Stage Size
One impact of reducing the fracture spacing in an oil well is the resulting increase in GOR (Figure
W.6-30). The increased GOR is the result of the increased drawdown in the reservoir as recovery
is increased and reservoir pressure is reduced.
Figure W.6-30: Fracture Spacing – 10-Year Forecast, GOR
Figure W.6-31 presents a plot of the recovery as a function of fracture spacing and Figure W.6-
32 presents a plot of the normalized recovery factors (normalized to the recovery factor value
for a stage length of 225ft) as a function of fracture spacing.
Figure W.6-31: Fracture Spacing – 10-Year Forecast, Recovery Factor
Figure W.6-32: Fracture Spacing – RF (Normalized to RF at 225ft)
Figure W.6-32 shows that, for this model, a stage size greater than 225ft does not significantly
reduce recovery factor but reducing the stage size may significantly increase recovery factor. The
economics of decreasing the fracture spacing, and increasing total number of stages in the well,
has to be evaluated to determine the optimum fracture spacing (number of stages and stage
size).
W.7 PDA – Eagle Ford Example

Historical production data was available from an Eagle Ford tight oil well (refer to “Workflow -
Eagle Ford Example” for input/output files).
One complexity for this well was that the well was initially produced through casing from July 31
to Aug. 6. Tubing was then run into the well and the well was returned to production. Although
PE² Essentials Production Data Analysis includes an internal routine to convert THP/CHP to BHP,
this complexity cannot be handled. In order to generate BHP, two PE² Essentials THP-BHP Oil Well
models were constructed:
PE_Essentials_Oil_THPBHP EagleFordCasingFlow.dvx (Figure W.7-1 and W.7-2)
PE_Essentials_Oil_THPBHP EagleFordTubingFlow. dvx (Figure W.7-3 and W.7-4)
Figure W.7-1: PE² Essentials THP-BHP Model – Casing Flow Model

Figure W.7-2: PE² Essentials THP-BHP Model – CHP to BHP Conversion
Figure W.7-3: PE² Essentials THP-BHP Model – Tubing Flow Model

Figure W.7-4: PE² Essentials THP-BHP Model – THP to BHP Conversion
The BHP was merged into the production data and imported into PE² Essentials PDA (Figure W.7-
5). The reservoir parameters and PVT parameters were entered (Figure W.7-6).
Figure W.7-5: PE² Essentials Production Data Analysis – Eagle Ford Example
Figure W.7-6: PE² Essentials Production Data Analysis – PVT Data
The PDA tool will automatically remove the zero rates for analysis purposes. To further improve
the analysis, the rate spikes were removed from the data (Figure W.7-7.
Figure W.7-7: PE² Essentials Production Data Analysis – Data Edit: Remove Rate Spikes
The existence of boundary-dominated flow was confirmed by placing a unit-slope line on the data
plot on the “Flow Regime Identification” sheet (Figure W.7-8). This confirms that flowing material
balance, simulation and conventional decline curve analyses are possible with this data set.
Figure W.7-8: PE² Essentials Production Data Analysis – Flow Regime Identification
There are many more plots available in the “Flow Regime Identification” sheet that can be used
to confirm the existence of bilinear, linear and boundary dominated flow (BDF).
W.7.1 Flowing Material Balance – Eagle Ford Example

Since BDF was confirmed to exist, a flowing material balance was performed on the data. It
should be noted that flowing P/Z and flowing PI is available for FMB of a gas well. These
parameters are not available for the oil FMB because oil production is a much more complex
process to model (Refer to Section 8.5.9).
A caveat when performing oil FMB, it is very important to load the proper material balance model
to estimate declining reservoir pressure. For a gas well, the default straight-line material balance
model will work since the gas recovery process is normally a straight-line P/Z process which
extrapolates to initial gas in place. For oil reservoirs, the material balance is a complex process of
depletion, gas and water drives and as a result, the default straight-line material balance model
will not be valid so an appropriate material balance model should be loaded.
Figure W.7-9 shows the FMB analysis using an imported depletion drive material balance model
generated with the PE² Essentials Oil Material Balance tool (Oil_Material_Balance Results
Depletion Drive EagleFord.csv). Resulting oil in place was 578 mbbls.
Figure W.7-9: PE² Essentials Production Data Analysis – Flow Material Balance Analysis
It should be noted that oil pseudo pressure was used for analysis. This gives a more rigorous
solution taking varying oil properties into account. Figure W.7-10 presents the analysis using
pressure. The result is a lower oil in place (465 mbbls). Section W.7.2 evaluates the difference.
Figure W.7-10: PE² Essentials Production Data Analysis – FMB – Pressure Based
W.7.2 Analytical Simulation – Eagle Ford Example

Since BDF was confirmed to exist and an oil in place was determined, the analytical simulator was
used to determine the remaining reservoir parameters for this Eagle Ford well. Figure W.7-11
shows the simulation results using the FMB pseudo pressure analysis results.
Figure W.7-11: PE² Essentials Production Data Analysis – Analytical Simulation, 578 mbbls Case
Figure W.7-12 presents the simulation results for 465 mbbls. Note that it would still be possible
to match the historical pressures by increasing the permeability to 2.5 md from 1.8 md.
Figure W.7-12: PE² Essentials Production Data Analysis – Analytical Simulation, 465 mbbls Case
W.7.3 Numerical Simulation – Eagle Ford Example

With the estimated reservoir parameters resulting from the analytical simulation a PE² Essentials
Basic Reservoir Simulator well model was built to confirm the analytical simulator results, fine-
tune the reservoir parameters, and generate a forecast for the well. Figure W.7-13 shows the
parameters used to build the simulation model and Figure W.7-14 is the simulation run.
Figure W.7-13: PE² Essentials Production Data Analysis – Numerical Simulation, 578 mbbls Case
Figure W.7-14: PE² Essentials Basic Reservoir Simulator – Eagle Ford, 578 mbbls Case
The results from the PDA analytical simulation and the Basic Reservoir Simulator are similar.
W.7.4 Decline Curve Analysis – Eagle Ford Example

Since boundary dominated flow was evident from the PDA, the data was exported to a DCA
database file and loaded into the PE² Essentials Decline Curve Analysis tool (Figure W.7-15).
Figure W.7-15: PE Decline Curve Analysis – Eagle Ford
Figure W.7-16: PE Decline Curve Analysis – eDCA, Eagle Ford

The eDCA tool was used to evaluate the Eagle Ford data Figure W.7-16). The LOPL model was
used with a regression start time of 0.3 yrs to restrict the analysis to the BDF period, and a total
production period of 3 years. The option to “Include Equiv Arps” was checked to generate the
Arps parameters for forecasting with the main DCA tool.
It is also possible to generate a match using the cum oil data for this well. In this case, the match
starts at time zero so the match results do not match the rate-derived results. Figure W.7-17
shows the cum match.
Figure W.7-17: PE Decline Curve Analysis – eDCA, Eagle Ford, Cum Match Based
After returning to the main DCA screen, the water cut and GOR trends were estimated (Figure
W.7-18) and a forecast was generated (Figure W.7-19).
Figure W.7-18: PE Decline Curve Analysis – Water Cut and GOR Trends
Figure W.7-19: PE Decline Curve Analysis – DCA Forecast
From the DCA forecast, the ultimate oil recovery is forecast to be 62.3 mbbls (10.6%) with a
remaining recovery, to a minimum of 1 bopd of 35.5 mbbls.
It should be noted that a DCA forecast generated from the cum-based eDCA match yielded an
ultimate recovery of 53.1 mbbls and a remaining recovery of 26.3 mbbls. This reduced recovery
is a result of including the early time, low rate period (time <0.3 years) in the cum-based match.
At this point the data could be exported for Monte Carlo forecasting or additional forecasting
and economic analysis could be performed.
W.8 Asset Evaluation – 6-Well Pad

The proposed plan is to produce the Marcellus by drilling two 6-well production pads. The
question to be answered: is it better to drill the two pads simultaneously with two rigs or
sequentially with one rig. A pad cannot be placed on production until all 6 wells are completed
and tied into the facilities.
W.8.1 Marcellus Historical Wells - DCA

In order to base the economics on real Marcellus data, a DCA database containing 12 Marcellus
wells was built for this project. The DCA database containing the base data is located in the
“Workflow - Marcellus Pads” directory and is called “PE_Essentials_DCA_DataBase
Marcellus.dvx”.
The first step is to perform a DCA on the historical data and generate a forecast for each well.
The Marcellus-11 well in the database was used to demonstrate the steps used for this example.
Since the entire history will be used as a “forecast” for the CAPE analysis, eDCA was used to
determine the Arp parameters for the entire history plus a 10-year forecast. Figure W.8-1 shows
the first step: calculate the LOPL parameters for a 10-year history/forecast.
Figure W.8-1: eDCA – Marcellus-11: LOPL Analysis
The next step is to fit an Arps model to the data (Figure W.8-2) and then modify the Dlim until
the history/forecast matches the LOPL model results.
Figure W.8-2: eDCA – Marcellus-11: Arps Analysis and Match
The results are then transferred to the main DCA page by clicking “Save Params” and a forecast
is generated by clicking “Run Forecast” (Figure W.8-3) on the main DCA page.
Figure W.8-3: eDCA – Marcellus-11: Generate Forecast for CAPE
This was done for all 12 Marcellus wells in the database and the results are stored in the
“PE_Essentials_DCA_DataBase Marcellus Inter.dvx” file for import into the Asset Evaluation tool.
Table W.8-1 present the summary of the eDCA of the twelve Marcellus wells.
Table W.8-1: PE Decline Curve Analysis – eDCA Results: Marcellus
W.8.2 Marcellus Pad Production – Asset Evaluation Import Parameters

The Marcellus history/forecast was then imported into the Asset Evaluation tool (Figure W.8-4).
Figure W.8-4: Asset Evaluation – Import DCA Forecast

For this example, the second pad is available immediately, 5.5 months (~27 days/well) after
production start-up or delayed to 1 year after production start-up. The drilling of the second pad
was modeled by delaying the availability of the second group of 6 wells as shown in Figure W.8-
5. Figure W.8-5 compares the three production profiles that were evaluated: no delay for second
pad; 5.5-month delay for second pad, and 1-year delay for second pad.
Figure W.8-5: Asset Evaluation – Comparison of Marcellus Forecast
In both delay cases, the capital for the second pad is spent in the first year but the production
availability is different.
The economic data used for these sensitivity cases for the Marcellus development were similar
to the economic data presented in Section W4. The following figures – Figure W.8-6, Figure W.8-
7, Figure W.8-8, Figure W.8-9 – present the input data used for this analysis.
Note – the TRC fiscal regime was used with an assumed gas royalty rate of 18% and a corporate
income tax rate of 35%.
The oil and gas price forecast was imported from the “Historical and Forecasted Oil and Gas
Prices.xlsx” spreadsheet located in the “Example Input Files\Excel Files” directory
Figure W.8-6: Asset Evaluation – Marcellus: Model Parameters
Figure W.8-7(a): Asset Evaluation – Marcellus: Capital Costs and Pad Timing, No Delay
Figure W.8-7(b): Asset Evaluation – Marcellus: Capital Costs and Pad Timing, 5.5 Month Delay
Figure W.8-7(c): Asset Evaluation – Marcellus: Capital Costs and Pad Timing, 1-Year Delay
Figure W.8-8: Asset Evaluation – Marcellus: Operating Costs
Figure W.8-9: Asset Evaluation – Marcellus: Fiscal Terms

W.8.3 Marcellus Pad Comparison – Asset Evaluation Results

The analysis results for the three scenarios are presented in Figures W.8-10 and W.8-11.
Figure W.8-10: Asset Evaluation – Marcellus: NPV Results
Figure W.8-11: Asset Evaluation – Marcellus: Economic Results
From the analysis, the simultaneous production of both pads has higher economic value. A large
part of this is the rate acceleration resulting from having 12 wells available at the start of
production.
If a delay is required, delaying for 1-year appears to have slightly better economics because of
more efficient use of the tax pools by the delayed pad, but this could just be the result of yearly
averaging in the calculations.
W.9 Hydrate Protection Requirement

A gas well is produced through a pipeline to a central surface facility and then exported. The PE
Nodal tool can be used to generate the different pressures in the system if actual pressure-
temperature values are unknown.
The following is given for this example:

Bottomhole Pressure / Temperature: 2500 psi / 120 F
Wellhead Pressure / Temperature: 1750 psi / 80F
Pipeline Outlet Pressure / Temperature: 1500 psi / 70F
Central Manifold Export Pressure / Temperature: 900 psi / 65F
Separator Pressure / Temperature: 500 psi / 45F
W.9.1 Hydrate Formation Curves

The PE Essentials Hydrate Tool was used to generate the pressure-temperature hydrate curves
for 0% methanol (Figure W.9-1), 10% methanol (Figure W.9-2), and 20% methanol (Figure W.9-
3) and the csv files were exported for each case. The example model included in PE Essentials
was used to generate the hydrate curves.
Figure W.9-1: Generation of Hydrate Curves for 0% Methanol


W.9.2 Graphical Analysis

The PE Essentials Graph Tool was used to evaluate the requirement for methanl to protect the
production from hydrates. To plot the analysis, the production pressure/temperature data was
copied to a csv file in a format similar to the Hydrate tool’s csv plots.
Figure W.9-4 shows the format to the hydrate tool files and Figure W.9-5 presents the csv file
built for the production data.
Figure W.9-4: Hydrate Tool CSV File
Figure W.9-5: Production Data CSV File
The key to the production data csv file is the layout and the first line in the file has to be the same
as the first line in the Hydrate Tool’s csv file. The first line is used to confirm that compatible files
are being loaded.
All the csv files were then loaded into PE Graph (Figure W.9-6) and the analysis plot was built
(Figure W.9-7).
Figure W.9-6: PE Graph - Import of csv File

Figure W.9-7: PE Graph – Plot Generation
Figure W.9-8 shows the production scenario in terms of pressure/temperature.
Figure W.9-8: PE Graph – Hydrate Evaluation
From Figure W.9-8, there is a risk of hydrates when pressure falls below 800 psi. This occurs
between the central manifold and the separator. Based on the analysis, a methanol solution of
<=10% injected after the central manifold would protect this system from hydrate formation.

PE Essentials

Uploaded by

Copyright:

Available Formats

PE Essentials

Uploaded by

Document Information

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

PE Essentials

Uploaded by

Copyright:

Available Formats

Petroleum Engineering

& Economics Essentials

Don LeBlanc, PEng

Tools and Techniques to Evaluate

Don LeBlanc, PEng

Mr. Don LeBlanc, PEng has an in-depth background in petroleum

PVT.3 Oil Properties ................................................................................................................................ 73

Hydraulic Fracture Design Tool ....................................................................................................... 141

UNC.1 Analytical Model ........................................................................................................................ 226

OMB.2.1 Schilthuis Steady State Aquifer Model ........................................................................................................................................ 312

FDP.5 Production Modeling .................................................................................................................. 406

Asset Economic Evaluation Tool ..................................................................................................... 480

PE² Essentials is a comprehensive suite of software tools comprised of Reservoir/Production

Figure 1: PE² Essentials

The tools included in the current version of the program are:

IOR/EOR/Heavy Oil Essentials

Asset Management Essentials

Field Development Essentials

Asset Valuation Essentials

Figure 2: PE² Essentials – Authorizing the Program

PE Asset Management Essentials

Production Database Tool

Figure PDB-1: PE² Essentials Database Tool

Figure PDB-2: Plotting of Database Data – Multi-Well Plot

Figure PDB-3: Summing Wells in Production Database

PDB.1 Data Import

Figure PDB-4: PE² Essentials Database - Importing Data

Figure PDB-5: PE² Essentials Database – Data Table List

PDB.2 Database Editing

Figure PDB-6: PE² Essentials Database - Editing Data

Figure PDB-7: PE² Essentials Database – Delete Well

Figure PDB-8: PE² Essentials Database – Change Well Name

Figure PDB-9: PE² Essentials Database – List Wells

Figure PDB-10: PE² Essentials Database – Append Data

Figure PDB-11: PE² Essentials Database – Production Well Data Edit

Figure PDB-12: PE² Essentials Database – Injection Well Data Edit

Note – it is strongly recommended that the PE Production Database be backed up on a regular

Production Data Analysis Tool

Figure PDA-1: PE² Essentials Production Data Analysis Tool

The tool is comprised of a number of different modules:

PDA.1 Reservoir Data Entry

Figure PDA-2: PDA – Reservoir Data

Figure PDA-3: PDA – Choosing a Well

Figure PDA-4: PDA – Choosing Pressure Source for analysis

PDA.2 Production Data Import

Figure PDA-5: PDA – Production Data Import

Figure PDA-6c: PDA – Update Well in PE Tools Database

PDA.3 PVT Data Entry

Figure PDA-7: PDA – PVT

Figure PDA-8: PDA – PVT Import

PDA.4 Oil and Gas Pseudo Pressure

PDA.4.1 Gas Pseudo Pressure

This transformation of pressure to pseudo-pressure is an exact transformation, which accounts

PDA.4.2 Oil Pseudo Pressure

Oil pseudo pressure is defined as follows:

PDA.5 Data Validation / Diagnostics

Figure PDA-9: PDA – Data Validation /Diagnostics

Figure PDA-10: PDA – Data Editing

Figure PDA-11: PDA – Data Editing: Removing Spikes

Figure PDA-12: PDA – Multi-Well Plotting