Well Testing

interpretation
Swar Manaf Taha
MSc- PEG department
Advanced Well testing
Outlines
Introduction to Well Testing
Objective of Well Test Interpretation
Data Processing
Identifying Flow Regimes
Type curves
Type curve types
Use of numerical simulation
Model identification
Parameter estimation
Results verification
Introduction to Well Testing
Tests on oil and gas wells are performed at various stages of drilling, completion and production.
The test objectives at each stage range from simple identification of produced fluids and
determination of reservoir deliverability to the characterization of complex reservoir features. Most
well tests can be grouped either as productivity testing or as descriptive/reservoir testing.
Productivity well tests are conducted to;
- Identify produced fluids and determine their respective volume ratios.
- Measure reservoir pressure and temperature.
- Obtain samples suitable for PVT analysis.
- Determine well deliverability.
- Evaluate completion efficiency.
- Characterise well damage.
- Evaluate workover or stimulation treatment.
Introduction to Well Testing
Descriptive tests:
- Evaluate reservoir parameters.
- Characterize reservoir heterogeneities.
- Assess reservoir extent and geometry.
- Determine hydraulic communication between wells.
Whatever the objectives, well test data are essential for the analysis and improvement
of reservoir performance and for reliable predictions.
Objective of Well Test Interpretation
Comprehensive interpretation of acquired data is critical for efficient
reservoir development and management because it quantifies the
parameters that characterize the dynamic response of the reservoir.
Systematic approach for interpretation must be used to obtain the most
self-consistent and correct results.
Data Processing
Transient well tests are conducted as a series of dynamic events triggered by specified
changes in the surface flow rate. During interpretation, it may be desirable to analyze just
one particular event or all events simultaneously. In either case, the data must first be
processed.
 The first step in data processing is to split the entire data set into individual flow periods.
The exact start and end of each flow period are specified. Because the sampling rate is
usually high, each transient typically includes many more data points than are actually
required. A high density of data is needed only for early-time transients. Therefore,
special algorithms are usually employed to reduce the data set to a manageable size.
Because of the nature of the pressure disturbance propagation, a logarithmic sampling
rate is preferred.
Data Processing
 The sequence of events should incorporate the recent flow rate history of the well with
the surface flow rate changes observed during the test. This enables rigorous accounting
for superposition effects. As stated previously, the shape of the pressure transient curve is
affected by the production history of the reservoir. Each change in production rate
generates a new pressure transient that passes into the reservoir and merges with the
previous pressure effects. The pressure trends observed at the wellbore result from the
superposition of all the pressure changes.
 The next step is to transform the reduced data so that they display the same identifiable
features, regardless of test type.
 After the data are transformed, the task of identifying the flow regime begins .
Identifying Flow Regimes
Identifying flow regimes, which appear as characteristic patterns displayed by the pressure
derivative data, is important because a regime is the geometry of the flow streamlines in the
tested formation. Thus, for each flow regime identified, a set of well or reservoir parameters
can be computed.
The popular Flow Regime Identification tool is used to differentiate the eight common
subsurface flow regimes on log-log plots for their application in determining and
understanding downhole and reservoir conditions.
Identifying Flow Regimes
 Radial flow: The most important flow regime for well test interpretation is radial flow,
which is recognized as an extended constant or flat trend in the derivative. Radial flow
geometry is described as flow streamlines converging to a circular cylinder
Identifying Flow Regimes
 Spherical flow occurs when the flow streamlines converge to a point. This flow regime
occurs in partially completed wells and partially penetrated formations.
Identifying Flow Regimes
Linear flow: The geometry of linear flow streamlines consists of strictly parallel flow
vectors. Linear flow is exhibited in the derivative as a positive half-slope trend. Following
figure shows why this flow regime develops in vertically fractured and horizontal wells.
The combination of linear flow data with radial flow data (in any order) can provide the
principle values of kx and kv for the directional permeabilities in the bedding plane. In an
anisotropic formation, the productivity of a horizontal well is enhanced by drilling the well
in the direction normal to the maximum horizontal permeability.
Identifying Flow Regimes
Identifying Flow Regimes
Bilinear flow: Hydraulically fractured wells may exhibit bilinear flow instead of, or in
addition to, linear flow. The bilinear flow regime occurs because a pressure drop in the
fracture itself results in parallel streamlines in the fracture at the same time as the
streamlines in the formation become parallel as they converge to the fracture. The term
bilinear refers to the simultaneous occurrence of two linear flow patterns in normal
directions.
Identifying Flow Regimes
Dual porosity or permeability: Dual-porosity or -permeability behavior occurs in reservoir
rocks that contain distributed internal heterogeneities with highly contrasting flow
characteristics. Examples are naturally fractured or highly laminated formations. The
derivative behavior for this case may look like the valley-shaped trend shown in Fig. a, or it
may resemble the behavior shown in Fig. b. The dual-porosity or -permeability flow regime
is used to determine the parameters associated with internal heterogeneity, such as
interporosity flow transmissibility, relative storativity of the contrasted heterogeneities and
geometric factors.
Identifying Flow Regimes
Identifying Flow Regimes
Identifying Flow Regimes
Slope doubling: it describes a succession of two radial flow regimes, with the slope of the
second exactly twice that of the first. This behavior typically results from a sealing fault, but
its similarity to the dual-porosity or -permeability behavior in below figure it also shows
that it can also be caused by a permeability heterogeneity, particularly in laminated
reservoirs. If slope doubling is caused by a sealing fault, the distance from the well to the
fault can be determined.
Identifying Flow Regimes
Type curves
Type curves are graphical representations used in reservoir engineering and well testing to
analyze and interpret data from production or pressure transient tests. These curves depict
the expected response of a well or reservoir under certain idealized conditions, facilitating
the interpretation of field data and the estimation of reservoir parameters.
Type curves
The original rationale for type curves was to interpret interference tests using the line source
solution. Later type curves for wellbore storage and skin effects were developed to improve
on Horner buildup analysis, which was in error whenever an apparent straight-line trend in
the transient data that was not due to radial flow in the reservoir was used to compute
estimates for k, s and p*. Over time, models capturing near-well geometry (partial
penetration, vertical fracture), reservoir heterogeneity (homogeneous, dual porosity, dual
permeability) and outer boundaries (faults, drainage boundaries, constant-pressure
boundaries) were presented as families of type curves. Since the advent of the pressure
derivative, new models have been introduced in the literature as type-curve pairs for
pressure change and its derivative. Expert well test analysts have learned to recognize
models for observed transient data as identifiable trends in the pressure
derivative.
Type curve types
 Drawdown Type Curves: Used to analyze pressure drawdown tests, where the well is
shut in and pressure is monitored as it declines over time.
 Buildup Type Curves: Used to analyze pressure buildup tests, where the well is opened to
flow after a period of shut-in, and pressure is monitored as it increases over time.
 Rate Transient Analysis (RTA) Type Curves: Used to analyze production data and
forecast future production performance based on transient flow behavior.
 Material Balance Type Curves: Used to analyze reservoir depletion and estimate
remaining reserves based on material balance principles.
Use of numerical simulation
Acquired transient data commonly contains behavior dominated by effects that are not
captured in analytical models. Typical departures from the analytical model assumptions are
multiphase flow, non-Darcy flow and complex boundary configurations that are not easily
generalized in an analytical model catalog.
Such features can be addressed with a numerical model, but commercial numerical
simulators are designed for full-field simulation with multiple wells and do not readily
adapt to the single-well focus and short time frame inherent to well testing.
If they are adapted to focus on the short-term transient behavior of a single well or a few
wells, and also designed to present the data in the form used for well test interpretation,
numerical models can provide considerable insight beyond that possible from analytical
models.
Model identification
 For the model identification stage, the analyst should recognize certain characteristic
patterns displayed by the pressure transient data. This is greatly facilitated by a
knowledge of straight-line pressure derivative response trends associated with the
formation flow geometry.
 Reservoir information collected from geoscientists assists the selection of a reservoir
model. The distinctions among the various model options consistent with the transient
test data are not always clear-cut, and more than one model may provide similar
responses. In this case the analyst may rule out most model options by consulting with
colleagues working with other, independent data. If the flow regime responses are poorly
developed or nonexistent, interdisciplinary discussion may suggest the selection of an
appropriate model and reasonable starting values for the parameter estimation stage of
the interpretation.
Parameter estimation
Once the reservoir model has been identified, it is necessary to compute the model
parameters. Using initial parameter estimates from specialized flow regime analysis,
interdisciplinary input or both resources, an initial simulation for the transient response is
computed. The initial simulated and observed responses usually differ. Modern analysis,
however, is assisted by nonlinear regression routines that automatically refine the parameter
estimates until the simulation coincides with the observed data for the essential portions of
the transient response. Thus, the first interpretation stage of model identification represents
the main challenge for the analyst.
Results verification
Several drawdown and buildup periods are typically included in a well test, and it is
common to interpret every transient and cross-check the computed reservoir parameters.
However, analysis of all the transients in a test is not always possible. In this situation,
forward modeling may help confirm the validity of a reservoir model.
Basically, forward modeling involves simulating the entire series of drawdowns and
buildups and using the reservoir model and its parameters (Fig. 31). Because the simulation
continues for much longer than an individual transient, the effects of reservoir boundaries
are more likely to be noticed. If the simulation does not match the entire pressure history,
then the assumed reservoir model should be reassessed.
Results verification