WELL TEST INTERPRETATION

By

OKAFOR STANLEY

G2018/IPS/MSC/PPD/324

September 23, 2019

 TABLE OF CONTENTS

TABLE OF CONTENTS .................................................................................................................... 2

LIST OF FIGURES ............................................................................................................................ 3

LIST OF TABLES .............................................................................................................................. 4

1.0 INTRODUCTION ............................................................................................................................. 5

2.0 DATA INPUT .................................................................................................................................. 6

2.1 Pressure and Temperature Loading ............................................................................................ 11

2.2 Pressure Difference Plot ............................................................................................................. 14

3.0 DATA QUALITY ANALYSIS ............................................................................................................ 15

3.1 Gauge Accuracy Offset ................................................................................................................ 16

3.2 Gauge Selection .......................................................................................................................... 16

3.3 Pressure and Rate data ............................................................................................................... 17

3.4 Build-up extraction...................................................................................................................... 18

4.0 MODEL SUGGESTION .................................................................................................................. 19

4.1 Iteration one ............................................................................................................................... 19

4.1.1 Improve on log-log ................................................................................................................... 20

4.1.2 Improve on simulation ............................................................................................................. 22

4.2 Iteration two ............................................................................................................................... 23

4.3 Iteration three ............................................................................................................................. 25

5.0 IMPORTANT INTERPRETATIONS.................................................................................................. 28

5.1 Inflow Performance Relationship ............................................................................................... 28

5.2 SENSITIVITY ANALYSES ................................................................................................................ 29

2
 LIST OF FIGURES

Figure 1: Entering date and time ............................................................................................................ 6

Figure 2: Entering well and reservoir parameters .................................................................................. 7
Figure 3: Entering PVT description.......................................................................................................... 8
Figure 4: Entering analytical parameters ................................................................................................ 9
Figure 5: Entering numerical parameters ............................................................................................. 10
Figure 6: Entering initialisation model .................................................................................................. 11
Figure 7: Interface of loading gauges .................................................................................................... 12
Figure 8: Interface of loading pressure ................................................................................................. 13
Figure 9: Loaded pressure for gauges ................................................................................................... 13
Figure 11: Pressure difference plot....................................................................................................... 14
Figure 12: Unsynchronized gauges ....................................................................................................... 15
Figure 13: Synchronized gauges............................................................................................................ 15
Figure 14: Gauge separation distance .................................................................................................. 16
Figure 15: Eliminating noise to obtain good build up ........................................................................... 17
Figure 16: Good build up region ........................................................................................................... 17
Figure 17: Extracting model from build up ........................................................................................... 18
Figure 18: Generated model from build up .......................................................................................... 18
Figure 19: First iteration model selection ............................................................................................. 19
Figure 20: First model iteration run ...................................................................................................... 20
Figure 21: Regression line drawn .......................................................................................................... 20
Figure 22: Regression calculation process ............................................................................................ 21
Figure 23: Improved model for first iteration (log-log) ........................................................................ 21
Figure 24: Match for accepting K and C ................................................................................................ 22
Figure 24: Improved model for first iteration (simulation) .................................................................. 22
Figure 25: Model for second iteration .................................................................................................. 23
Figure 26: Model run for second iteration............................................................................................ 24
Figure 27: Model run for second iteration with regression .................................................................. 24
Figure 28: Model selection for third iteration ...................................................................................... 25
Figure 29: Derivative match for selected model ................................................................................... 26
Figure 30: Improved model for iteration three .................................................................................... 26
Figure 31: Inflow performance relationship curve ............................................................................... 28
Figure 32: Sensitivity on skin................................................................................................................. 29
Figure 33: Sensitivity on well bore storage constant ............................................................................ 29
Figure 34: Sensitivity on thickness (h)................................................................................................... 30

3
 LIST OF TABLES

Table 1: Iteration I model...................................................................................................................... 19

Table 2: Iteration I results before and after improvement (log-log) .................................................... 21
Table 3: Iteration I results before and after improvement (simulation) .............................................. 23
Table 4: Iteration II model..................................................................................................................... 23
Table 5: Iteration II model results ......................................................................................................... 24
Table 6: Iteration III model.................................................................................................................... 25
Table 7: Iteration III model results ........................................................................................................ 27

4
 1.0 INTRODUCTION

Well testing involves the perturbation of the reservoir and measurement of pressure variations with
time. It is used to determine well parameters like Permeability, Productivity Index (PI) and reservoir
properties like the existence of boundaries which are used to determine completion procedures for
production. In particular, well testing is very crucial in identifying the following:

 Well production capacity

 Static reservoir pressure
 Skin effect
 Flow capacity (Kh)
 Flow capability (K)
 Drainage radius of well
 Existence of heterogeneities (faults)
 Types and changes of produced fluid(s)
In carrying out this project, a well test analysis software (Saphir) developed by Kappa was used. It is a
diagnostic tool used in matching measured data to the model, taking into account the detailed production
history. Pressure changes in a well with time are measured and can be interpreted with Saphir to
determine reservoir characteristics.
The pressure and temperature data analysed are for a field with the following data:

Reservoir thickness (h) = 40 ft

Porosity (φ) = 31%
Radius Well (rw) = 0.35 ft
Formation Volume Factor (Bo) = 1.139 rb/STB
Viscosity of oil (μo) = 3.33 cp
Total Compressibility (Ct) = 14.96 E-6 psi-1
Production Rate = 2000 stb/day.

5
 2.0 DATA INPUT

Both static and dynamic parameters were entered into the software. The well and reservoir parameters
were entered into the information frame as given. The time and date were retrieved from the pressure
data of the gauges given in an ASCII file. The sequence of data input is shown in the figures below. In
step 6, an initial model is assumed before loading the gauges. This is done before prior knowledge of
the pressure response of the reservoir. The implication of this is that there is need for a detailed analysis
to be made later on, to match the model to what has been observed.

Figure 1: Entering date and time

6
Figure 2: Entering well and reservoir parameters

7
Figure 3: Entering PVT description

8
Figure 4: Entering analytical parameters

9
Figure 5: Entering numerical parameters

10
 Figure 6: Entering initialisation model

2.1 Pressure and Temperature Loading

The pressure and temperature data for all gauges are loaded in the QAQC section of the software.
The four gauges used are named CGM-B4-30376, CGM-B4-30377, CGM-B4-30385 and CGM-B4-
30400 respectively. All four gauges were loaded thus:

11
Figure 7: Interface of loading gauges

12
Figure 8: Interface of loading pressure

Figure 9: Loaded pressure for gauges 13

2.2 Pressure Difference Plot

The pressure difference plot is generated by clicking on the difference key and selecting a reference
gauge. The reference gauge selected was CGM-B4-30376. Several plots can be generated for different
reference gauges selected. By obtaining difference plots for each selected reference gauge, the condition
of a gauge can be inferred and this can be used in detecting bad gauges.

Figure 11: Pressure difference plot

From the figure above, it can be seen that CGM-B4-30385 has anomalies and should not be used for
pressure transient analysis.

14
 3.0 DATA QUALITY ANALYSIS

From the difference plots, the quality of data can be inferred. It can be said that the quality of the data
is bad as seen from the spikes in the pressure difference plot. This also shows that the gauges are
unsynchronized due to gauge movement etc. Synchronisation of gauges is necessary and this is done
by carrying out a time shift of gauge data. Care should be taken to avoid varying or changing the
pressure readings during the process.

Figure 12: Unsynchronized gauges

Figure 13: Synchronized gauges

15
3.1 Gauge Accuracy Offset

Gauges CGM-B4-30385 and CGM-B4-30376 are the upper gauges because they consistently have the
lowest pressure values. As shown in the plot above; using CGM-B4-30377 as reference, CGM-B4-
30385 and CGM-B4-30376 show positive pressure difference indicating their pressures are lower than
that of CGM-B4-30377. CGM-B4-30400 showed negative pressure difference indicative of the fact
that CGM-B4-30377 and CGM-B4-30400 are the lower gauges. The schematic below shows position
of the gauges.

Figure 14: Gauge separation distance

Gauge accuracy offset was calculated with the assumption of the worst case, that the gauges took
measurements in the water zone. Therefore, with a separation distance of 4ft, gauge accuracy offset is
calculated thus:

𝐺𝑎𝑢𝑔𝑒 𝑂𝑓𝑓𝑠𝑒𝑡 𝐴𝑐𝑐𝑢𝑟𝑎𝑐𝑦 = 𝑊𝑎𝑡𝑒𝑟 𝑃𝑟𝑒𝑠𝑠𝑢𝑟𝑒 𝐺𝑟𝑎𝑑𝑖𝑒𝑛𝑡 × 𝑂𝑓𝑓𝑠𝑒𝑡 𝐷𝑖𝑠𝑡𝑎𝑛𝑐𝑒

Gauge accuracy offset = 0.433 psi/ft × 4 ft

Gauge accuracy offset = 1.732 psi

3.2 Gauge Selection

The lower gauges CGM-B4-30377 and CGM-B4-30400 were more stable and showed consistent
pressure record. Hence analysis could be done with either of the two gauges. CGM-B4-30377 was thus
selected as the gauge for the build-up test analysis because it was the most stable.

16
3.3 Pressure and Rate data

After selecting a gauge, the pressure data has to be analysed and a region selected for analysis. A build
up analysis requires a good build up region. This can be achieved using the box and delete tools. A
corresponding rate profile is generated, using the split tool, based on the selected build up and
production regions. Furthermore, the rates can be specified. A production rate of 2000 rb/d is used as
given.

Figure 15: Eliminating noise to obtain good build up

Figure 16: Good build up region

17
3.4 Build-up extraction

Using the extract key in the analysis section, the good build-up region previously selected is extracted.

Figure 17: Extracting model from build up

Figure 18: Generated model from build up

18
 4.0 MODEL SUGGESTION

As already stated, a model that describes the observed behaviour from the build-up analysis must be
selected. Model selection is based on experience and a basic understanding of the principles of well
testing. Iterations of different models are carried out in a bid to try and fit the observed behaviour with
an explainable trend, which will aid in the description of reservoir behaviour. Analytical fits are made
to the model, which are then improved. The different iterations carried out are stated thus:

4.1 Iteration one

Wellbore storage model Constant

Well model Vertical (Finite radius)
Reservoir model Homogeneous
Boundary model Infinite acting

Table 1: Iteration I model

Figure 19: First iteration model selection

19
 Figure 20: First model iteration run

Figure 21: Regression line drawn

4.1.1 Improve on log-log

The good transient was highlighted by a regression line and the model was improved on log-log plots.

20
 Figure 22: Regression calculation process

Figure 23: Improved model for first iteration (log-log)

Results before and after applying the improve function are given in the table respectively:

Skin 4.37654 3.52448

Permeability (mD) 299.045 269.938
Flow capacity (mD.ft) 11961.8 10797.5
Reservoir pressure (psia) 4668.03 4677.9
Wellbore storage constant(rb/psi) 3.95226E-4 2.28546E-4

Table 2: Iteration I results before and after improvement

(log-log)
21
4.1.2 Improve on simulation

Accepting K and C from match of build-up gives:

Figure 24: Match for accepting K and C

Accepting all data from match of build-up gives:

Figure 24: Improved model for first iteration (simulation)

22
The values of the parameters from the two processes; accepting only K, C and accepting all parameters
are given respectively:

Skin 4.37654 3.99708

Permeability (mD) 298.915 330.849
Flow capacity (mD.ft) 11956.6 11916.3
Reservoir pressure (psia) 4668.42 4667.5
Wellbore storage constant(rb/psi) 2.48752E-4 2.43134E-4

Table 3: Iteration I results before and after improvement (simulation)

Improve on log-log produced a more matched model than improve on simulation as seen in the first
iteration. Therefore, subsequent iterations will be done based on log-log improvements.

4.2 Iteration two

Wellbore storage model Constant

Well model Vertical (Finite radius)
Reservoir model Homogeneous
Boundary model Single fault

Table 4: Iteration II model

Figure 25: Model for second iteration

23
 Figure 26: Model run for second iteration

Figure 27: Model run for second iteration with regression

Skin 3.8849 3.37755

Permeability (mD) 283.627 264.047
Flow capacity (mD.ft) 11345.1 10561.9
Reservoir pressure (psia) 4671.59 4685.89
Wellbore storage constant(rb/psi) 3.95226E-4 2.15732E-4

Table 5: Iteration II model results

24
4.3 Iteration three

Wellbore storage model Changing

Well model Vertical (Finite radius)
Reservoir model Homogeneous
Boundary model Single fault

Table 6: Iteration III model

Figure 28: Model selection for third iteration

25
Figure 29: Derivative match for selected model

Figure 30: Improved model for iteration three

26
Skin 6.78329
Permeability (mD) 375.369
Flow capacity (mD.ft) 15014.7
Reservoir pressure (psia) 4659.03
Wellbore storage constant(rb/psi) 2.06129E-4

Table 7: Iteration III model results

27
 5.0 IMPORTANT INTERPRETATIONS

To complete the interpretation, the representative model for the well and reservoir has to be examined
with other tools to validate the choice of model. These include;

 Inflow Performance Relationship (IPR) Plot

 Sensitivity analysis

5.1 Inflow Performance Relationship

The IPR curve is a plot of bottomhole flowing pressure against production rate. It quantifies the flow
rate of a well as a function of drawdown and describes the ability of a well to produce fluids at varying
BHP. It is useful in estimating well capacity, tubing string design, scheduling of an artificial lift method.

Figure 31: Inflow performance relationship curve

28
5.2 SENSITIVITY ANALYSES

Figure 32: Sensitivity on skin

Figure 33: Sensitivity on well bore storage constant

29
 Figure 34: Sensitivity on thickness (h)

It can thus be inferred from the comparison and sensitivity analysis done that the well most closely
fits the third iteration:
 A changing wellbore storage
 Homogeneous Reservoir
 Single sealing fault Boundary

30