Generating RP PDF

UNIVERSITY OF OKLAHOMA
GRADUATE COLLEGE
DETERMINATION OF RELATIVE PERMEABILITY FROM WELL PRODUCTION BY

CONSIDERATION OF FLUID TYPE, FORMATION HETEROGENEITY, AND SKIN
FACTOR
A THESIS
SUBMITTED TO THE GRADUATE FACULTY
in partial fulfillment of the requirements for the
Degree of
MASTER OF SCIENCE
By
AHMED ZARZOR HUSSIEN AL-YASERI
Norman, Oklahoma
2010
DETERMINATION OF RELATIVE PERMEABILITY FROM WELL PRODUCTION BY

CONSIDERATION OF FLUID TYPE, FORMATION HETEROGENEITY, AND SKIN
FACTOR
A THESIS APPROVED FOR THE

MEWBOURNE SCHOOL OF PETROLEUM AND GEOLOGICAL ENGINEERING
BY
___________________________________
Dr. Faruk Civan, Chair
___________________________________
Dr. Deepak Devegowda
___________________________________
Dr. Bor-Jier Shiau
Copyright by AHMED Z. HUSSIEN Al-YASERI 2010

All Rights Reserved.
DEDICATION
This thesis is dedicated to my parents, brothers and sisters for their love and support and to my
wife for her encouragement and patience. All of my success is because of them.
My Mother
Mothers are the lovely and greatest persons in the life. I love my mother because she is the one
who gave the life. Always she took care of me, she stayed awake all the time to make sure that I
am alright, and she got tired all the day for my comfort. She spent her life to raise me up. When I
was a child, she was feeding me. Usually, my mother advice me and provided me the good
guidances from her experience in life.
I love you mom!
ACKNOWLEDGMENTS
I would like to thank Dr. Faruk Civan, chairman of my committee, for his advice and
suggestions that help me to complete the present work. Also, I would like to thank Dr.
Deepak Devegowda and Dr. Bor-Jier Shiau for their contributions and time as members
of my committee.
Thanks again for Dr. Deepak Devegowda for his assistance and suggestions in
implementation of reservoir simulator.
Thanks to Dr. Tibor Bodi and Dr. Peter Szucs, University of Miskolc, Hungary for their
assistance in providing the data from their paper.
Special thanks to the faculty and staff of the Mewbourne School of Petroleum and
Geological Engineering, especially Shalli Young and Sonya Grant for their kindness and
willingness to help when needed.
I wish to acknowledge and thank many people for their cooperation and support during
my stay at the University of Oklahoma.
Above all, I would like to thank God for the love, support and blessings in my life.
iv
TABLE OF CONTENTS
LIST OF TABLES ........................................................................................................... v
LIST OF FIGURES .........................................................................................................vi
ABSTRACT ....................................................................................................................xvi
1. BACKGROUND ...........................................................................................................1
2. REVIEW OF THE TOTH ET AL. METHOD FOR DETERMINATION OF
RELATIVE PERMEABILITY FROM WELL PRODUCTION DATA ..................16
3. GENERATION OF SIMULATED WELL PRODUCTION DATA BY A
COMMERCIAL RESERVOIR SIMULATOR ...........................................................37
4. EVALUATION OF THE TOTH ET AL. METHOD FOR RADIAL FLOW
USING SIMULATED PRODUCTION DATA.............................................................45
4.1 EFFECT of RESERVOIR SIZE on TOTH et al. METHOD.......................................45
4.2 THE EFFECT OF (P) VALUE ON THE TOTH Et Al. METHOD ........................60
4.3 EFFECT OF THE VISCOSITY ON THE TOTH ET AL. METHOD .......................78
4.4 EVALUATING THE TWO CRITICAL EQUATIONS FOR THE TOTH ET AL.
METHOD .......................................................................................................................102
4.5 CONCLUSIONS ......................................................................................................109
5. CONSIDERATION OF SKIN FACTOR AND HETEROGENEITY
DURING THE ESTIMATION OF RELATIVE PERMEABILITY

CURVES FOR UNSTEADY STATE RADIAL DISPLACEMENT.....110
5.1 EFFECT OF SKIN FACTOR ............................................................................ .....110
5.1.1 SKIN FACTOR AND CAPILLARY PRESSURE EFFECT ................................116
5.1.2 NEGATIVE SKIN FACTOR ................................................................................119

5.2
EFFECTS
OF
HETEROGENEITY
ON
RELATIVE
PERMEABILITY
CURVES..........................................................................................................................121
5.3 CONCLUSIONS ......................................................................................................130
6. MODIFICATION AND GENERALIZATION OF THE TOTH ET
AL. METHOD FOR COMPRESSIBLE AND INCOMPRESSIBLE

FLUIDS AND EVALUATION BY MEANS OF A RESERVOIR
SIMULATOR ............................................................................................131
6.1 FORMULATIONS OF RELATIVE PERMEABILITIES UNDER LINEAR
FLOW.............................................................................................................................134
6.2 FORMULATION FOR RELATIVE PERMEABILITIES UNDER RADIAL
FLOW.............................................................................................................................138
6.3 FORMULATION FOR FLUID SATURATIONS....................................................139
6.4 EVALUATION OF THE NEW TECHNIQUE USING A RESERVOIR
SIMULATOR..................................................................................................................143
6.5 APPLICATION AND VERIFICATION ..................................................................144
6.6 CONCLUSIONS ......................................................................................................146
RFERENCES.................................................................................................................159
APPENDIX A NOMENCLATUR ...............................................................................166
vi
LIST OF TABLES
Table 2.1 Petrophysical data for example 2.1..................................................................32
Table 2.2 Petrophysical data for example 2.2 ..................................................................37
Table 4.1 Petrophysical parameters for the case studies ..................................................46
Table 4.2 Production data for example 4.1.1 ...................................................................47
Table 4.3 Np, Wp and Wi for example 4.1.1 ...................................................................48
Table 4.4 Constant parameters for examples (4.1.1), (4.12), (4.13) ................................56
Table 4.5 Petrophysical parameters for the case studies ..................................................60
Table 4.6 Petrophysical parameters for the case studies .................................................69
Table 4.7 Constant parameters for (P) =34.45 bar .........................................................78
Table 4.8 Constant parameters for (P) = 74.98 bar ........................................................78
Table 4.9 Petrophysical parameters for the case studies .................................................79
Table 4.10 Production data for the core sample ...............................................................95
Table 4.11 Use Toth et. al method to recalculate relative permeability curves ...............95
Table 4.12 Petrophysical parameters for example 4.3.5 ..................................................98
Table 5.1 Petrophysical parameters for example 5.1 ......................................................114
Table 6.1 Petrophysical parameters for example 5.1, 5.2, 5.3 and 5.4 .........................147
vii
LIST OF FIGURES
Fig.(1.1) Permeability definition ......................................................................................11
Fig.(1.2) General relative permeability curves..................................................................11
Fig.(1.3) Drainage and Imbibition displacement ..............................................................12
Fig.(1.4) Steady state and unsteady state method for core sample ...................................12
Fig.(1.5) Three section core for Penn -State method ........................................................13
Fig.(1.6) General Welges plot .........................................................................................13
Fig.(1.7) Pressure volume relationship .............................................................................14
Fig.(1.8) Fluid density vs. pressure for different fluid types ............................................14
Fig.(1.9) Flow regimes .....................................................................................................15
Fig.(2.1) Oil and water Production data for example 2.1 .................................................31
Fig.(2.2) Displacement equation for example 2.1.............................................................32
Fig.(2.3) Welges plot for example 2.1 .............................................................................32
Fig.(2.4) Relative permeability ratio curve for example 2.1.............................................33
Fig.(2.5) Relative permeability curves for example 2.1 ...................................................33
Fig.(2.6) Water fractional curve (after breakthrough time) for example 2.2 ...................34
Fig.(2.7) Displacement equation for example 2.2 ............................................................35
Fig.(2.8) Cumulative water influx for example 2.2 ..........................................................35
Fig.(2.9) Relative permeability ratio curve for example 2.2 ............................................36
Fig.(2.10) Relative permeability curves for example 2.2 .................................................36
Fig.(3.1) Three dimension shape from Eclipse with six injection wells and one production
well in the center (=60o) ................................................................................................39
Fig.(3.2) Two dimension shape from Eclipse with six injection wells and one production
viii
well in the center (=60o) .................................................................................................39

Fig.(3.3) Two dimension shape from Eclipse result .........................................................40
Fig.(3.4) Three dimension shape from Eclipse result .......................................................41
Fig.(3.5) Three dimensional shape from Eclipse result ....................................................42
Fig.(3.6) One slice shape for Radial flow .........................................................................43
Fig.(3.7) Radial flow but only for one slice (From Eclipse result) ...................................43
Fig.(3.8) Time vs. rate for one slice and the whole reservoir ............................................43
Fig.(4.1) Assumed relative permeability curves for examples 4.1.1, 4.1.2, and 4.1.3 .....46
Fig.(4.1.1) Production data from Eclipse for example 4.1.1 ............................................46
Fig.(4.1.2) Displacement equation for example 4.1.1 ......................................................49
Fig.(4.1.3) Cumulative water volume produced for example 4.1.1 ..................................49
Fig.(4.1.4) Relative permeability ration (assumed and calculated) for example 4.1.1 .....50
Fig.(4.1.5) Relative permeability curves (assumed and calculated) for example 4.1.1 ....50
Fig.(4.1.8) Cumulative water volume produced for example. 4.1.2 .................................52
Fig.(4.1.9) Relative permeability ration (assumed and calculated) for example 4.1. 2 ....52
Fig.(4.1.10) Relative permeability curves (assumed and calculated) for ex. 4.1.2 ..........53
Fig.(4.1.11) Production data from Eclipse for example 4.1.3 ..........................................53
Fig.(4.1.12) Displacement equation for example 4.1.3 ....................................................54
Fig.(4.1.13) Cumulative water volume produced for example. 4.1.3 ...............................54
Fig.(4.1.14) Relative permeability ration (assumed and calculated) for ex. 4.1.3 ............55
Fig.(4.1.15) Relative permeability curves (assumed and calculated) example 4.1.3 .......55
ix
Fig.(4.1.16) Grid size for example 4.1.1 from Eclipse .....................................................57

Fig.(4.1.20) Relative permeability ration (assumed and calculated) for example 4.1.3
after reducing the grid size ................................................................................................59
Fig.(4.1.21) Relative permeability curves (assumed and calculated) for example 4.1.3
after reducing the grid size ................................................................................................59
Fig.(4.2. 1) Production data from Eclipse for example 4.2.1 ...........................................61
Fig.(4.2.4) Relative permeability ration (assumed and calculated) for ex. 4.2.1...............62
Fig.(4.2A.5) Relative permeability curves (assumed and calculated) for ex. 4.2.1...........63
Fig.(4.2.8) Cumulative water volume produced for example 4.2.2 .................................64
Fig.(4.2.9) Relative permeability ration (assumed and calculated) for ex. 4.2.2 ..............65
Fig.(4.2.13) Water fraction for example 4.2.3 after breakthrough time ...........................67
Fig.(4.2.14) Cumulative water volume produced for example 4.2.3 ................................67
Fig.(4.2.15) Relative permeability ratio (assumed and calculated) for ex.4.2.3 ..............68
Fig.(4.2.16) Relative permeability curves (assumed and calculated) for ex. 4.2.3.. ........68
Fig.(4.2.4) Relative permeability ration (assumed and calculated) for ex. 4.2.1 ..............71
Fig.(4.2.5) Relative permeability curves (assumed and calculated) for ex. 4.2.1
........72
Fig.(4.2.6) production data from Eclipse for example 4.2.2 .............................................72

Fig.(4.2.9) Relative permeability ration (assumed and calculated) for ex. 4.2.2 .............74
Fig.(4.2.13) Water fraction for example. 4.2B.3 after breakthrough time .......................76
Fig.(4.2.14) Cumulative water volume produced for example. 4.2.3 ..............................76
Fig.(4.2B.15) Relative permeability ration (assumed and calculated) for ex. 4.2B.3 ......77
Fig.(4.3.3) Water fraction for example 4.3.1 after breakthrough time .............................81
Fig.(4.3.4) Cumulative produced water volume for example 4.3.1 ..................................81
Fig.(4.3. 5) Relative permeability ration (assumed and calculated) for ex. 4.31 ..............82
Fig.(4.3. 6) Relative permeability curves (assumed and calculated) for ex. 4.3.1 ...........82
xi

Fig.(4.3.9) Water fraction for example 4.3.2 after breakthrough time .............................84
Fig.(4.3. 11) Relative permeability ration (assumed and calculated) for ex. 4.3.2 ...........85
Fig.(4.3. 12) Relative permeability curves (assumed and calculated) for ex. 4.3.2 .........85
Fig.(4.3. 16) Relative permeability ration (assumed and calculated) for ex. 4.3.3 ...........87
Fig.(4.3. 17) Relative permeability curves (assumed and calculated) for ex. 4.3.3 .........88
Fig.(4.3.18) Cumulative recovery of oil versus cumulative volume of injected fluid for
different viscosity ratio (Eclipse result data) ....................................................................89
Fig.(4.3.20) Displacement eq. for different examples with different viscosity ratios ......89
Fig.(4.3.21) Cumulative produced water volume for different examples with different
viscosity ratios ..................................................................................................................90
Fig.(4.3.22) Water saturation distribution as a function of distance between injection and
production wells for (a) ideal and (b) non-ideal displacement ........................................90
Fig.(4.3.23) Water cut vs. production rate for heavy and light oil reservoir ....................91
Fig.(4.3.24) Total production vs. water cut for the last examples from Eclipse result data
for different viscosity ratios ..............................................................................................92
Fig.(4.3.25) Effect of viscosity ratio on relative permeability curves ..............................93
Fig.(4.3.26) Viscosity vs. residual oil ...............................................................................93
xii
Fig.(4.3.27) Relative permeability curves by JBN method for different viscosity

ratio....................................................................................................................................94
Fig.(4.3. 30) Relative permeability ratio for example 4.3.4 .............................................97
Fig.(4.3. 31) Relative permeability curves for example 4.3.4 ..........................................97
Fig.(4.3.32) Assumed relative permeability curves as input data in Eclipse for example
4.3.5....................................................................................................................................99
Fig.(4.3.35) Cumulative water volume produced for example 4.3 5 ..............................100
Fig.(4.3.34) Displacement equation for example 4.3.5 ..................................................100
Fig.(4.3. 37) Relative permeability curves (assumed and calculated) for ex. 4.3.5........101
Fig.(4.3. 36) Relative permeability ration (assumed and calculated) for ex. 4.3.5 .........101
Fig.(4.4.1) Displacement equation for field data (well A-225) ......................................103
Fig.(4.4.2) Displacement equation for field data (well A-710) ......................................103
Fig.(4.4.3) Displacement equation for simulated example by use Eclipse .....................104
Fig.(4.4.4) Wedges plot for simulated example by Eclipse ..........................................105
Fig.(4.4.5) Toths plot for simulated example by Eclipse ..............................................105
Fig.(4.4.6) Welges plot for field data for well A-225 ...................................................106
Fig.(4.4.7) Welges plot for field data for well A-710 ...................................................106
Fig.(4.4.8) the value of (b1=1.0068) for simulated example by Eclipse ........................108
Fig.(4.4.9) the value of (b1=1.0106) for the same pervious simulated example by Eclipse
..........................................................................................................................................108
xiii
Fig.(5.1) Skin Zone .........................................................................................................111

Fig. (5.2) Three dimensional shapes by Eclipse .............................................................114
Fig.(5.3) Production data by Eclipse for example 6.1 ....................................................114
Fig.(5.4) Relative permeability curves for different Skin values, example 6.1 ..............115
Fig.(5.5) Relative permeability ratios for different Skin values, example 6.1 ...............115
Fig.(5.6) Capillary pressure vs. water saturation ............................................................116
Fig.(5.7) Relative permeability ratios for different Skin values with capillary pressure
..............................................................................................................................117
Fig.(5.8) Relative permeability curves for different Skin values with capillary pressure
effect ...............................................................................................................................117
Fig.(5.9) Relative permeability with Skin equal to zero with and without capillary
pressure include ..............................................................................................................118
Fig.(5.10) Relative permeability curve with and without capillary pressure by
experimental work. .........................................................................................................119
Fig.(5.11) Cross section for the reservoir with skin zone ...............................................120
Fig.(5.12) Relative permeability curves for different negative Skin values when the
viscosity is 10cp ..............................................................................................................120
Fig.(5.13) The distribution for the permeability by Eclipse (Mean = 100, S.D = 30) ....122
Fig.(5.14) Histogram for random permeability values (Mean = 100, S.D = 30) ............123
Fig.(5.15) Relative permeability curves for homogenous (k=100 md) and heterogeneous
(Mean=100, S.D. =30) ....................................................................................................123
Fig.(5.16) Relative permeability ratios for homogenous (k=100md) and heterogeneous
(Mean=100, S.D. =30) ....................................................................................................124
xiv
Fig.(5.17) The distribution for the permeability by Eclipse (Mean = 100, S.D = 30) with
short channel ...................................................................................................................124
Fig.(5.18) Relative permeability curves for homogenous (k=100 md), heterogeneous
(Mean=100, S.D. =30) and heterogeneous with short channel (k=400 md and 1000 md)
..............................................................................................................................125
Fig.(5.19) 2 D. distribution for the permeability by Eclipse (Mean = 100, S.D = 30) with
short channel ...................................................................................................................125
Fig.(5.20) Relative permeability ratios for homogenous (k=100 md), heterogeneous
(Mean=100, S.D. =30) and heterogeneous with short channel (k=400 md and 1000 md)
..............................................................................................................................126
Fig.(5.21) Histogram for random permeability values (Mean = 500, S.D = 100) ..........126
Fig.(5.22) Relative permeability curves for homogenous (k=500 md), heterogeneous
(Mean=500, S.D. =100) and heterogeneous with short and long channel (k=2000 md and
3000 md) .........................................................................................................................127
Fig.(5.23) The distribution for the permeability by Eclipse (Mean = 500, S.D = 100) with
long channel ....................................................................................................................127
Fig.(5.24) Relative permeability ratios for homogenous (k=500, k=100 md),
heterogeneous (Mean=100, 500, S.D.=30,100) and heterogeneous with short and long
channel (k=2000, 1000, 3000 md) ..................................................................................128
Fig.(5.29) The distribution for the permeability by Eclipse (Mean=100, S.D=30) with
different long channels ....................................................................................................128
Fig.(5.25) Relative permeability curves for homogenous (k=100 md) and different long
channels ...........................................................................................................................129
xv
Fig.(5.26) Relative permeability ratios for homogenous (k=100 md) and different long
channels ...........................................................................................................................129
Fig.(6.1) Pressure versus 1/B ........................................................................................147
Fig.(6.3) Two dimensional shapes by Eclipse ................................................................148
Fig.(6.2) Linear displacement .........................................................................................148
Fig.(6.4) Three dimensional shape by Eclipse ................................................................148
Fig.(6.6) Power low equation for Toth et al. example 5.1 ..............................................149
Fig.(6.5) Displacement equation for Toth et al. example 5.1 ..........................................149
Fig.(6.8) Relative permeability ratios for example 5.1by using different methods ........150
Fig.(6.7) Power low equation for the new method example 5.1 .....................................150
Fig.(6.10) Production data for example 5.2 from Eclipse ..............................................151
Fig.(6.9) Relative permeability for example 5.1by using different methods ..................151
Fig.(6.12) Cumulative injected vs. time for Toth et al. example 5.2 ..............................152
Fig.(6.11) Displacement equation for Toth et al. example 5.2 .......................................152
Fig.(6.14) Relative permeability for ex. 5.2 by using different methods with assumed
values...............................................................................................................................153
Fig.(6.13) Relative permeability ratios for example 5.2 by using different methods with
assumed values ................................................................................................................153
Fig.(6.15) Relative permeability for example 5.2 by using different methods with
assumed values ................................................................................................................154
Fig.(6.18) Relative permeability for exAMPLE 5.3 by using the new method with
assumed values.................................................................................................................155
xvi
Fig.(6.17) Relative permeability ratios for example 5.3 by using different methods with
assumed values ................................................................................................................155
Fig.(6.19) Assumed Relative permeability for example 6.4 ...........................................156
Fig.(6.21) Relative permeability ratios for example 5.4 by using the new method and
Welge method ..................................................................................................................157
xvii
ABSTRACT
Accurate estimation of relative permeability is essential for reliable reservoir history

matching and decision-making, and effective reservoir management. This information is
also critical for the design, implementation, and monitoring of enhanced oil recovery
processes.
This study presents an effective method for acquiring the relative permeability
data directly from the well production data in oil reservoirs where water or gas acts as the
fluid phase displacing oil. The methodology presented in this study provides convenient
interpretation formulae which are applicable to unsteady-state, two-phase, immiscible,
and both compressible and incompressible fluids. The total mobility and the mobility
ratio of the immiscible fluids are related to the characteristic parameters of the
displacement process and the cumulative injected fluid pore volume following Toth et al.
These parameters are then incorporated into a general correlation function which allows
for analytically estimation of the relative permeability functions. The present approach
produces unique estimation of the relative permeability functions and is more practical
than the previous approaches which rely on computationally complicated history
matching procedures, often suffer from the non uniqueness issue of the obtained relative
permeability data.
As an extension of the Toth et al. method, the analytic method developed here
determines the relative permeability functions for compressible and slightly-compressible
fluids uniquely considering the effects of formation heterogeneity and skin factor on the
well production data. This approach provides the estimates of the relative permeability
functions, representative of the macroscopic two-phase flow behavior in the formation
xviii
around the well. The Toth et al. method and its present modification and extension are
evaluated using the simulated well production data generated by means of a commercial
reservoir simulator under various production and reservoir conditions.
It is demonstrated that the drainage area, pressure drop and fluid viscosity have
significant effects on determined relative permeability curves using Toth et al. method for
radial flow. The Toth et al. method works satisfactorily even for heterogeneous reservoirs
and there is skin effect. The skin factor has a significant effect on the relative
permeability curves which increases when the oil viscosity increase or capillary pressure
increase. The effect of heterogeneities on the relative permeability curves is negligible.
However, the effect becomes significant when there are channels. The developed method
is a very simple, general and accurate method that is applicable for both incompressible
and compressible fluids (gas or liquid).
xix
CHAPTER 1
BACKGROUND
Introduction
Relative permeability is essential information required for evaluation, history matching,
effective management, and characterization of multiphase flow in a petroleum reservoir.
Thus, accurate and representative estimates of the relative permeability functions are
critical. Many methods have been proposed for determination of the relative permeability
curves. The goals of this study include the evaluation of the Toth et al. method (1998,
2001, 2005, 2006) and modification for compressible fluids. The Toth et al. method is
one of the simplest direct methods used to estimate relative permeability by processing
production data. It can give accurate and average estimations for relative permeability
compared with other methods that depend on the fluid flow test data obtained with core
samples to estimate the relative permeability curves. However, the Toth et al. method
was derived for incompressible fluids. This limitation is circumvented by developing a
new method for compressible fluids.
Relative permeability is defined as the ratio of the effective permeability to any
specific fluid (oil, water, or gas phases) to the absolute permeability. For example, the oil
relative permeability is given by:
k
kro o - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -1.1
k
Similarity, the water relative permeability is:
k
krw w - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 1.2
k
The gas phase relative permeability is defined in the same manner.
Permeability was defined mathematically for the first time by Darcy (1856) by his
equation known as Darcys law (Ahmed, T. 2001) as shown in Eq.(1.3) below. Darcy
assumed that permeability is a rock property, has a constant value, and does not depend
on the fluid flowing through the rock (Dake L. P., 1978).
k dp
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -1.3
dl
Where v is the flow per unit area per unit time, is fluid viscosity and l is distance.
If there is more than one kind of fluid (oil and water, oil and gas, water and gas or
oil, water, and gas) flowing inside porous rock, each fluid has its own permeability called
effective permeability (ko, kw, kg). The effective permeability depends on the saturation
for each fluid reaching the absolute permeability (k) when there is just one fluid flowing
through the porous medium. Usually, relative permeability is plotted versus the wetting
phase saturation, for example water saturation (Sw) when the rock is water-wet Fig.(1.2).
This study is limited to flow of two phases. We consider the main following fluid
gdisplacement processes. Fluid displacement can be carried out in two ways.
In a drainage displacement process, the non-wetting fluid (oil or gas) displaces the
wetting fluid (water, for example for a water-wet porous media). Therefore, the saturation
of non-wetting phase increases forward as seen in Fig.(1.3), (Patrick W., 2001).
In an imbibition displacement process, the wetting fluid (water, for example in
water-wet porous media) displaces the non-wetting fluid (oil or gas), causing the
saturation of wetting phase to increase as seen in Fig.(1.3).

Many methods are available for measurement of relative permeability. We can
classify them into two types: steady state and unsteady state. We can measure relative
permeability under steady-state conditions when "fixed ratio of fluids is forced through
the test sample until saturation and pressure equilibrium are established, (Alam W. U.,
1988)." Unsteady- state conditions happen when only one phase is injected into the core
to displace the second phase present in the core during the test, causing the saturation to
change continuously, (Alam W. U., 1988) as shown in Fig.(1.4).
Steady- State Methods:

The first study about relative permeability determination was conducted by Wyckoff and
Botest (Caudle, 1951). This was an experimental work for two phase flow in a sand core
sample. In addition, they examined the relations between relative permeability for some
fluids and their saturations. Leverett and Lewis (Levertt, 1941) extended the study by
Wyckoff and Botest to handle the three phase systems (oil, water, and gas). Leverett and
Lewis found that the relative permeability of water was a function of water saturation
only and it was not affected by the presence of other fluids (oil and gas). The relative
permeability of gas was low compared with the two phase system under the same gas
saturation. The oil relative permeability was unstable and more complex because it could
be low or high when compared with the two phase system under the same oil saturation.
Morse et al. (1947) developed a new method for determining relative
permeability. Their method was modified by Osoba et al. (1951). The technique used by
Henderson and Yuster (1948), Caudle et al. (1951), Geffen et al. (1951) , and Morse et
al. (1947) is known as the Penn-State method, illustrated by Fig.(1.5). We can define the
Penn-State method as "forcing fluid mixture through cores mounted in Lucite. The test
core sample is placed between two samples of similar material that are in capillary
contact. Mixing of the two fluids occurs in the first section and the boundary effect of the
wetting phase is confined to the third (outlet) section"(Alam W., 1988). The Penn-State
method applies for liquids or gas-liquid systems at either increasing or decreasing of the
wetting phase (Mehdi H., 1988).
Another technique for estimating relative permeability under steady state
conditions is the Single-Sample Dynamic Method, which was developed by Richardson
et al. (1951), Josendal et al. (1952), and Loomis and Crowell (1962). This technique
differs from the Penn-State method in the handling of the end effects and placement of
the test samples between two core samples and the two phases are injected
simultaneously through a single core, (Mehdi H., 1988).
Unsteady-State Methods:
There are a number of other methods for measuring relative permeability under the
unsteady-state condition, but the most important methods are as follows:
1-Stationary Fluid method developed by Leas et al. (1950).
2- Hassler Method, (1944).
3-Hafford method, (1951).
4- Dispersed Feed Method, (1951).
5- Johnson et al.(JBN), (1959).
6- Jones and Roszelle (JR), (1978).
7- Marle, (1981).
8-Toth et al., (1998).
Most of these methods apply for linear displacement.
The determination of relative permeability under an unsteady-state condition can
be applied faster than the steady-state condition, but the application is mathematically
more complex. Buckley and Leverett, (1942) developed the first displacement theory,
which was later extended by Welge, (1952). Welge was the first to show how to calculate
the relative permeability ratio in case the gravity is neglected. Leverett, (1941) gave the
mathematical basis by combining Darcys law with a definition of capillary pressure to
obtain the following expression:
1
f w2
ko Pc
(
g sin )
qt o x
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -1.4
k
1 o . w
k w o
Where fw2 is the fractional water in the outlet stream; qt is the superficial velocity of total
fluid leaving the core; is the angle between the flow direction x and the horizontal; and
is the density difference between displacing and displaced fluids. Welge, (1952)
showed that if we ignore capillary pressure, assume flow horizontal ( = 0) and after
some mathematical manipulation, we can calculate the relative permeability ratio with the
saturation as shown below:
k
qnw
dQ
f nw
nw 1 w nw
qnw qw
dQ
knw w
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -1.5
Where
Q Q w Qnw - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -1.6
f w 1 f nw - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 1.7
From this definition of fnw. Thus,

kr , nw
kr , w
dQ
nw 1 nw
w
dQ
dQnw nw f nw
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 1.8
dQ
w f w
or in a field expression such as :
f
water water - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -1.9
k r , oil
oil f oil
k r , water
Welge used the following material balance to measure the saturation depending on the
wetting fluid
L
A S nw dx AL(1 S c ) Qnw - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 1.10

o
Then, integratin g the left side by parts yields as :

S nw(L)
ALSnw(L) A xdS nw AL(1 S c ) Qnw - - - - - - - - - - - - - - - - - - - - - - - - - - - -1.11

S ro
Observing that dSw dS nw and using

AxS w Q
df w(S w )
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 1.12
dS w
results in:
S nw(L)
dQ
1
AL(1 Sc ) Qnw Q nw - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 1.13
AL
dQ
or in field expression
So
1
Vp (1 S wi ) N p (Wi f o ) - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -1.14
Vp
S w 1 So - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -1.15
S w S wi (
Np
W f
i o ) - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 1.16
Vp
Vp
Welge observed the behavior of the production data before and after the
breakthrough time as shown in Fig.(1.6). He also observed that the relationship between
the cumulative recovery of non-wetting fluid (Qnw) versus cumulative volume for injected
wetting fluid (Q) in a linear immiscible displacement experiment is a straight-line (linear)
with slope equal to one that before breakthrough time (Q = QB), but after breakthrough
time (Q > QB) the relation will be as follows, (Collins, 1976):
Qnw a b ln(Q) 1.17

Where (a) and (b) are constants for any small segment:
Q
dQ nw
b 1.18
dQ
Welge also found that the curve for Qnw versus Q functioned strongly for the viscosity
ratio as well as the relative permeability curve, (Collins, 1976).
Note that Welges method can be used for both linear and radial displacements.
The work of Welge was later extended by Johnson et al. (JBN), (1959). They
showed how to calculate the individual relative permeabilites even in the case that the
gravity is not neglected. The equations for JBN method can be summarized (in field
expression) as:
k ro
fo
1.19
1 1
/ d
d
Qw I r Qw
and
f
krw w w kro 1.20
f o o
Where Ir is the relative injectivity, defined as:
Ir
q L
injectivit y
o - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -1.21
int ial injectivit y pk
S w S wi
Qo
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -1.22
AL
Toth et al., (2001) developed a technique for calculating relative permeability

ratio for linear displacement. Later, Toth et al., (2001) extended their method to be
applicable for individual relative permeabilites. The methods of Welge (1952), JBN and
Toth et al. can determine the relative permeability ratio and can be applied with both
laboratory data and field data. In addition, these methods have the following common
assumptions:
(a) Sufficiently high displacement rate so that the effect of capillary end-effect can be
ignored. (b) Incompressible and immiscible fluids. (c) Unsteady-state flow. (d)
Homogenous medium. (e) Constant reservoir properties.
Subsequently, Toth et al. (2005) extended their method for radial displacement
while keeping the same assumptions of their previous work. At the same time they
assumed that there is one production well in the center of the reservoir and there is a
natural water influx or injection wells around the reservoir to cause radial displacement.
First, they started with a small core as a disk with a drill hole in the center for injected
fluid to produce from the surrounding area of that core as shown in Fig.(3.1). Their focus
8
was to check the validity of their equations for their method. Finally, Toth et al. (2005,
2006) tested their method on an actual size reservoir.
Civan and Donaldson, (1989) developed a technique to determine relative
permeability for unsteady-state displacement, which depends on Darcys law. They made
the same assumptions of the Welge and JBN methods, but also included the capillary
pressure effect.
L p
Sw
S wx 0
fw
nw
dx..............................................................................................................A.23
f kA pc S w
1 nw
k nw
nw qnw S w x
dp
f kA pc S w
f w c dS w 1 nw
k nw
dS w
nw qnw S w x
dpc S w
S q L 1
dp
L Q ' w nw nw
p Q '
0.................................A.24
dS w x
kA
k nw
Q '
dQ '
1 k rnw w / k rw nw 1 k rnw / f w .....................................................................A.25
Subsequently, Civan and Evans, (1991) developed a method for estimating

relative permeability for steady-state and unsteady-state displacements based on NonDarcy law. This method was for compressible and immiscible fluids, and included the
capillary pressure effect. Also, they assumed that the viscosity is constant and the density
is variable with pressure. Later on, Civan and Evans, (1993) came up with a technique
for determining the relative permeability for compressible fluids. In addition, they
assumed that viscosity and density are variable with pressure by using a non-Darcy law
with capillary pressure included. However, these methods did not consider the skin factor
effect and heterogeneity effect although we expect a strong relationship between them
because we use the well production data to estimate the relative permeabilities.
PRESENT STUDY
In this study we will focus and extend upon the Toth et al.s (2006) study because it is a
unique, practical and direct method for estimating relative permeabilities in radial
systems and therefore it is applicable for determination of relative permeability from well
production data.
Most of the previous methods available for determining relative permeabilities
relied upon the other methods to check and validate their results. Hence, we cannot be
certain about the accuracy of these estimations because there is no real field data
available for relative permeabilies to compare with.
Our objectives in this study are as the following:
(1) Evaluation and determination of the accuracy and applicability of the Toth et al.
(2006)method. We used simulated data generated by reservoir simulation software for
this purpose.
(2) Determining under what conditions the Toth et al. (2006)method works the best.
(3) Determining the effect of reservoir parameters, essentially controlling the
performance of this method.
(4) Demonstrating these issues by several representative case studies.
(5) Extending the Toth et al. method for application involving the compressible fluids
systems.
(6) Studying the effect of skin factor and reservoir heterogeneity on the relative
permeability curves obtained by using the Toth et al. method.
10
Fig.(1.1) Permeability definition

Source: introduced from (Slatt , 2006)
Fig.(1.2) General relative permeability curves

Source: introduced from (Dake L. P., 1978)
11
* Drainage displacement
Water
+
Oil
Water wet core

At Sw 100%
Oil injection
* Imbibition displacement
Water
+
Oil
Water wet core

At Sor
Water injection
Fig.(1.3) Drainage and Imbibition displacement

Source: the graph was introduced from (Patrick W. 2001)
* Unsteady-State method
Water
Water injection
Oil saturated
Oil
* Steady-state Method
Water
Water
Oil
Oil
Fig.(1.4) Steady state and unsteady state method

For core sample, introduced from (Patrick W. 2001)
12
Fig.(1.5) Three section core for Penn -State method

Source: Introduced from ( Mehdi H., 1988)
Fig.(1.6) General Welges plot

Source: introduced from (Levertt, M.C., 1941)
13
Fig.(1.7) Pressure volume relationship

Source: introduced from (Ahmed, T., 2005)
Fig.(1.8) Fluid density vs. pressure for different fluid types.

14
Fig.(1.9) Flow regimes

15
CHAPTER 2
REVIEW OF THE TOTH ET AL. METHOD FOR
DETERMINATION OF RELATIVE PERMEABILITY FROM WELL
PRODUCTION DATA
_____________________________________________________________
The radial displacement interpretation formulas introduced by Toth et al. (1998, 2001 ,
2005, 2006) for determination of relative permeability from the well production data are
presented in this chapter. This chapter explains and summarizes the various formulations
of Toth et al. presented in different studies here in a consistent manner.
The Toth et al. considered a disk shape porous sample where the displacing fluid
(water) is injected from a small hole in the center of the core to displace the displaced
fluid (oil) towards the surrounding area. Toth et al. assumed one-dimensional radial,
isothermal and unsteady-state flow of two immiscible and incompressible fluids in
homogeneous and isotropic porous media with uniform thickness. Its porosity is and
permeability is k. The thickness of the rock sample is h; the radius of the axial well is rw,
and the external radius is re. The rock sample is saturated with a fluid denoted by a
subscript k. Then, this fluid is displaced by another fluid denoted by a subscript d. The
volumetric rate of the injected fluid is qi. The effect of the capillary force is neglected
(Pc=0) during the displacement processes. The pressure at the inlet face is Pe; the
pressure inside the well (fluid outlet face) is Pw. Thus, the pressure difference between
the outer and inner faces of the disk is P = Pe- Pw. Also, this method assumes that all
reservoir parameters will remain constant during the displacement. In addition, The Toth
16
et al. method is applied at and after the breakthrough time.

2. 1 Flow Equations
2. 1.1 Mantle of Radial Core as an Inlet Face
The total injected rate is assumed equal to the total production rates for phase 1 and phase
2
qi qk qd - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -2.1
The radial Darcy's flow equations are given by:
qd
2rhkkrd dp
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.2
d
dr
qk
2rhkkrk dp
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.3
k
dr
and
k
k
qi 2hk rd rk
d k
dp
dp
2hkY ( S d )
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -2.4
dr
dr
Where the Y(Sd) is the total mobility as shown:
k
k
Y(S d ) rd rk d k - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -2.5
Next, the Levertt functions are introduced for the fractional flow equation as:
q
krd
fd d
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -2.6
qi d Y ( S d )
q
krk
fk k
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -2.7
qi k Y ( S d )
f d f k 1 ,- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -2.8
17
k rd
f
d d - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -2.9
k rk
f k k
After rearranging Eq.(2.4), a differential equation is obtained as:
- dp
qi
dr
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -2.10
2hkY(Sd ) r
The following boundary conditions are applied for the integration:
p pe at r re and p pw at r rw - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.11
Where Pe > Pw then Eq.(2.10) become
p p e p w
qi re
dr
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.12
2hk rw rY ( S d )
Because Y(Sd) is a function of Sd and r is variable in Eq.(2.12); therefore, it should be

transformed as a function of Sd. For this purpose, we consider the Buckley-Leverett,
(1942) solution.
The displacement equation in a radial flow system given as.
V (t)
r 2(S d ) i
f d (S d ) - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -2.13
2h
where
t
Vi(t) qi dt , - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -2.14
0
f d (S d )
df d
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -2.15
dS d
A differentiation of Eq.(2.13) gives the following:
2rdr
Vi
df d - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.16
2h
18
The ratio of Eq.(2.16) and Eq.(2.13) gives:
dr 1 df d
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.17
r
2 f d
So, the new integral boundary conditions become:
f d 1, f d 0, df d 0 at r re and f d f d 2 , f d f d 2 , df d df d 2 at r rw - - - - - -2.18
As a result of Eq.(2.18) and substituting Eq.(2.17) into Eq.(2.12) gives:
p -
qi 0
df d
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.19
4hk f d2
f d Y(S d )
After introducing a special G function and reformulating Eq.(2.19) gives:
qi f d 2 df d
q
i G - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -2.20
4hk 0 f d Y(S d ) 4hk
The time derivative of Eq.(2.20) yields:
d((
1
dt
4hk
dqi
dG
qi dt G dt - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.21
Differentiating Eq.(2.14) twice with respect to time yields:
dVi(t)
dqi d 2Vi(t)
qi
and
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -2.22
dt
dt
dt 2
After rearranging Eq.(2.20), the next two equations can be derived:
4hkp
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.23
qi
And
19
dqi 4hkp d 2Vi (t )
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -2.24
dt
qi
dt 2
The (qi dG/dt) term is interpreted as the following for the outlet face (denoted by
subscript 2):
qi
qi
df d 2
dG
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -2.25
dt Y(S d 2 )f d 2 dt
Then, applying Eq.(2.13) at the outlet face (r = rw), the time derivative is obtained as:
f d 2
2hrw2
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.26
Vi (t )
and
df d 2 2hrw2 dVi (t ) 2hrw2
qi - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.27
dt
Vi (t )
dt
Vi2 (t )
Thus, using Eqs.(2.26), (2.27), and Eq.(2.25) can be reformulated to give:
qi
qi2
dG
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.28
dt
Y ( S d 2 )Vi (t )
Then, using Eqs.(2.24), (2.28), and Eq.(2.21) can be given as:
d((
1
dt
4hk
4hkp d 2Vi (t )
qi2
Y ( S d 2 )V i(t )
dt 2
qi
qi2
p d 2Vi (t )
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -2.29
qi dt 2
4hkY ( S d 2 )V i(t )
We can apply Eq.(2.29) for two types of boundary conditions, (a) P is constant and
20
thus Vi(t) is changing, and (b) qi is constant and P (t) is changing.

Case (a): P is constant
Eq.(2.29) becomes:
qi3
Y(S d 2 )
4hkpVi (t )
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -2.30
d Vi (t )
dt 2
Case (b): qi is constant

This means d2Vi(t)/dt2=0 in Eq.(2.29) and the sum of the fluid mobility is given by
Y(S d 2 )
qi
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.31
d (p)
4hkt
dt
The relative permeability functions can be determined using Eq.(2.29) with Eq.(2.30) or
Eq.(2.31), the last equation is always positive, Y(Sd2) > 0 if Vi(t) is increasing
continuously because:
dVi(t)
d 2Vi(t)
qi 0 and
0 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.32
dt
dt 2
Y(Sd2) > 0 should be positive. At the same time the other parameters are positive in
Eq.(2.31) except d(P)/dt < 0 that because the relative permeability and the phase
saturation of the displacing fluid are increasing forward.
2.1.2 Surface of The Radial Well as an Inlet Face

The boundary conditions are given by:
p pe at r re and p pw at r rw - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.33
If the displacing fluid is taking place inside the radial well, then the solution of the partial
21
differential Eq.(2.12) gives.
p pe pw
qi re
dr
2hk r w rY ( S d )
The difference between this equation and Eq.(2.12) is the minus sign. By applying the
boundary conditions (see Eq.(2.20)); the solution will be as follows:
qi f d 2 df d 2
4hk 0 f d Y(S d )
The minus sign is the difference between this equation and Eq.(2.19). At the same time
this equation is similar to Eq.(2.20). As a result, Eq.(2.21) - (2.31) given in section (2.1.1)
are also applicable for the displacement conditions considered here.
2. 2 Displacement Equations
Vi is referring to the volume of the displacing fluid during (t) time, Vk is the volume of the
displaced fluid during the time and Vd is the amount of displacing fluid. Thus, the
following volumetric expressions can be written.
t
Vi qdi dt
,- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.34
0
t
Vk qk dt - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.35
0
t
Vd qd dt - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.36
0
Vi Vk Vd - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -2.37
Similarly, the following equation can be written for the flow rates:
qdi qk qd - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.38
22
The effluent production rate and cumulative volume of the injected fluid are zero until the
breakthrough. Therefore, Eqs.(2.37) and Eq.(2.38) simplify before and at the
breakthrough time.
Via Vka - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -2.39

qdi qka - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -2.40
The fractional flow of the production fluids after the breakthrough is given by:
q
f k k - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.41
qdi
q
f d a d - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.42
qdi
If capillary effects are negligible, then we can consider the Welge's 6 equations, given by:
df d
1 fd
fk
1
f d
- - - - - - - - - - - - - - - - - -2.43
dS d
S d S d (S d S d 0 ) (S d S d 0 ) Vi /V p
From Eq.(2.43) we can obtain
Vi (S d S d ) (S d S d 0 )
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.44
Vp
fk
Where S d and V p express the average saturation of the injected fluid and pore volume of
the core, respectively.
Substituting Eq.(2.34) and Eq.(2.35) into Eq.(2.41) gives:
fk
dVk
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.45
dVi
The volume balance between the injected and displaced fluids over the core give the
following equations:
23
V
( S d S d 0 ) k - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -2.46
Vp
Hence, Substituting Eqs.(2.45) and (2.46) into Eq.(2.44) gives the following equation:
Vi (Vk / V p ) ( S d S d 0 )
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -2.47
Vp
(dVk / dVi )
2. 3 Distribution of Fluid Saturation along the Core Plug

The distribution of fluid saturation along the core plug with acceptable accuracy depends
on the information of the saturations at the inlet, outlet faces and the average saturation
over the core length, (Toth, 2006).
2.3.1 The Water Saturation Distribution During Water Injection

At the beginning of the water injection process, the water saturation in the core is at least
equal to the irreducible water saturation Swi or the somewhat higher Sw0. After the
breakthrough time (t ta), the water saturation distribution along the core can be
represented as:
2
A
S w ( x) S w0
2.48
x/ L B
The parameters A and B are determined by applying the boundary conditions at a given
time. x 0, S w Smax , and at x l, S w S wf At the breakthrough time t t a . Thus,
(S w, max S w0 )(S wf S wo )
(S w, max S w0 ) (S wf S w0 )
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -2.49
24
(S wf S wo )
(S w, max S w0 ) (S wf S w0 )
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -2.50
S w S w0 (S w, max S w0 )(S wf S wo ) - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.51
After the breakthrough time t ta , S w S w2 ; therefore,

A
(S w2 S w0 )(S w,max S wo )
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.52
(S w,max S w0 ) (S w2 S w0 )
(S w2 S wo )
(S w,max S w0 ) (S w2 S w0 )
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.53
S w S w0 (S w,max S w0 )(S w2 S wo ) - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -2.54
Eqs.(2.51) and (2.54) express the average water saturation in porous media as the
geometric mean of the water saturation increments at the inlet and outlet faces. Generally,
after the breakthrough time, the saturation distribution of the injected fluid along the core
can be represented by:
2
S d ( x) S d 0
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -2.55
x/ L B
Then, the average saturation can be expressed in the following manner:
1 l
A
S d S d 0
L 0
x/L B
dx - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -2.56
Note that the linear flow equations can be transformed to the radial flow equations by
applying the following coordinate transformation inferred by Civan, (2000).
x r 2 rw2
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -2.57
L re2 rw2
25
Such that
x w 0 , r rw - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.58
xe L , r re - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.59
From Eq.(2.56), it can be obtained that
B
S d 2 S d 0 b(S d S d 0 )2 ,where b
A
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -2.60
The relationship given by Eqs.(2.56) and (2.60) can be combined to yield:
V
S d 2 S d 0 b(S d S d 0 )2 b k - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -2.61
Vp

Where (b) is integration constant defined as:
1
(S d, max S di )
1 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -2.62
Where S d,max refers to the maximum saturation that will be reached following an
infinite the displacing fluid throughput, and S di represents the initial displacing fluid
saturation.
By substituting Eq.(2.61) into Eq.(2.47), and then considering that the pore volume Vp
remains constant and separating the variables yields:
d(Vi /V p )
Vi /V p
d(Vk /V p )
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -2.63
Vi /V p 1 b(Vk /V p )
The general solution of Eq.(2.63) yields a linear expression as:
Vi
V
a b i - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.64
Vk
Vp
26
Where a is a integration constant, denotes the fraction of the displaced fluid at the
saturation front with a value less than one ( Toth, 1998,2006). The pore volume is equal
to:
V p (re2 rw2 )h - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.65
So, the average saturation of the displacing fluid in the redial core sample after
breakthrough time is expressed as:
V
Sd S di k - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.A.66
Vp
The saturation of the displacing fluid at the outlet face denoted by a subscript 2 and it can
be estimated by:
Vi
V
p
S d 2 S di b
Vi
a b V
p
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.67
The Leverett-type, (1941) fractional fluid volumes can be determined as following based
on Eqs.(2.37), (2.6), and (2.7):
fk
V
a b i
V p
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -2.68
And
f d 1 f k - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.69
27
Application To Oil and Water System

a) The cumulative oil and water productions and cumulative volume of water influx are
determined respectively by using the following equations:
t
N p qo dt - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.70
o
t
W p qw dt - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.71
o
t
Wi qwi dt - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.72
o
The total production is equal to the water influx, thus:
qwi qo qw - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.73
b) The theoretical displacement equation used to determine the first two constants is
given by:
Wi
W
ab i
Np
Vp
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -2.74
That will be a straight line with slope b > 1 and intercept a < 1, where the constant a is
the oil fraction at the breakthrough time.
Thus, the pore volume for radial system can be estimated by:
V p (re2 rw2 )h - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.75

c) The water and oil fraction at the wellbore is determined by:
fw
qw
a
1
, and f o 1 f w - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.76
2
q w qo
W
a b i
V p
28
d) In the cases that the reservoirs produce under constant pressure, the total mobility can
be determined by:
k
k
a1b12t (b1 1 )
Y(SW ) rw ro
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.77
w
o
4hkp(b1 1 )
Where a1 and b1 are some empirical constant that can be determined by fitting the
empirical power-law function as:
Wi a1t b1 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.78
Note: The value of b1 must be greater than one (b1>1).
e) In the case of the reservoir producing under a constant rate. The total mobility can be
determined by different expression instead of Eq.(2.77):
Y(S w )
qwi
4hka2b2t
b2
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -2.79
Where a2 and b2 are some empirical constants that can be determined by fitting the
empirical power law function as:
p pe pw a2t b2 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -2.80
Note that the value of b2 must be negative (b2 < 0).
Eq.(2.80) can be applied only if the production well works perfectly efficiently so that the
skin factor s is zero. Otherwise, the P value in Eq.(2.77) and (2.80) should be corrected
as (Toth, 2005):
p pmeasured ps - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.81
Where p s is the additional pressure drop due to skin effect, and it is given by:
29
ps
SqBo o
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.82
2kh
f) The relative permeability ratio is determined by:
k rw
f w w
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.83
k ro ( 1 f w )o
g) The individual relative permeability values are determined by:
k rw f w wY(S w ) - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.84
k ro ( 1 f w )oY(S w ) - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -2.85
h) The water saturation is determined by:
Wi
V
p
S w S wi b
Wi
a bV
p
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.86
We apply the Toth et al. method on two examples; one deals with a reservoir under
constant water injection as shown in example (2.1) below, and the other involves a
reservoir under constant pressure as shown in example (2.2).
Example 2.1
This example was introduced from Stiles, (1971) and Toth et al. (2005). This case is
under constant water injection (500 m3/d). The production data is shown in Fig.(2.1). The
reservoir properties are summarized in Table (2.1).
30
Table 2.1 Petrophysical data for example

2.1, (Introduced from Toth et al., 2005)
Parameter
Ex. 2.1
Well radius ,rw , m
0.1
600
Radius of well influence, re , m
155
500
Well head area (A),m2
7477
Pay zone thickness, h.m
29
Pore volume,Vp,m3
478500
Porosity,
0.219
Permeability,K, m2
0.175
P,Pa
Variable
Skin factor, S
Oil formation volume factor, Bo
1.23
Water formation volume

factor, Bw
Oil viscosity, , pa.s
1
0.00132
Water viscosity, , pa.s
0.001
Irreducible water saturation,

(Swi)
0.23
q, m3/d
400
qw
300
200
qo
100
0
0
1000
2000
3000
t, d
Fig.(2.1) Oil and water Producction data for
example 2.1
31
7
6
y = 1.9851x + 0.6718
R = 0.9991
Vi/Np
4
3
2
1
0
0
0.5
1.5
2.5
Vi /Vp
Fig.(2.2) Displacement equation for example 2.1
3.E+5
2.E+5
Np
2.E+5
1.E+5
5.E+4
0.E+0
0.E+0
1.E+6
2.E+6
Wi
Fig.(2.3) Welges plot for example 2.1
32
3.E+6
100
krw/kro
10
Welge-method
TBSC-method
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Sw
Fig.(2.4) Relative permeability ratio curve for example 2.1
0.9
0.8
krw
0.7
kro
0.6
kr
0.5
0.4
0.3
0.2
0.1
0.0
0.2
0.3
0.4
0.5
0.6
0.7
Sw
Fig.(2.5) Relative permeability curves for example 2.1
33
0.8
Example 2.2
This example was introduced from Craig, (1971) and Toth et al. (2005). This example is
under constant water pressure (6800 kpa). The production data shown in Fig. (2.6) and
Fig. (2.7). The reservoir properties are summarized in Table (2.4).
Table 2.2 Petrophysical data for

example 2.2
(Introduced from Toth et al., 2005)
Parameter
Ex. 2.1
Well radius, rw , m
0.1
Pay zone thickness,
15.5
fw
h.m
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
TBSC method
Craig Jr. F.
Pore volume,Vp,m3
16776
Porosity,
Permeability,K, m2
0.0315
P, kpa
6800
Skin factor, S
Sw
Oil formation volume
1.2
Fig.(2.6) Water fractional curve

(after breakthrough time)
for example 2.2
0.1
factor, Bo
water formation
volume factor, Bw
Oil viscosity, , pa.s
0.001
water viscosity, ,
0.0005
pa.s
Irreducible water
0.25
saturation, (Swi)
34
0.2
0.3
0.4
0.5
0.6
0.7
12
10
Wi/Np
8
6
4
2
0
0
10
15
Wi/Np
Fig.(2.7) Displacement equation for example 2.2
3.E+5
y = 7E-05x1.1072
R = 0.9993
2.E+5
2.E+5
Wi
1.E+5
5.E+4
0.E+0
0.E+0
1.E+8
2.E+8
3.E+8
Time, day
Fig.(2.8) Cumulative water influx for example 2.2
35
4.E+8
1000
TBSC
krw/kro
100
Craig Jr. F.
10
1
0.4
0.5
0.6
0.7
0.8
Sw
Fig.(2.9) Relative permeability ratio curve for example 2.2
1.0
0.9
0.8
Craig Jr. F.
0.7
kr
0.6
kro
krw
0.5
0.4
Toth.
0.3
0.2
0.1
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Sw
Fig.(2.10) Relative permeability curves for example 2.2
36
0.8
CHAPTER 3
GENERATION OF SIMULATED WELL PRODUCTION DATA BY
A COMMERCIAL RESERVOIR SIMULATOR
_____________________________________________________________
Eclipse
TM
software developed by Schlumbertger is a reservoir simulator well-known by
the oil and gas industry by over the last 25 years, and it is considered to be the leading
finite difference based reservoir simulator. Eclipse
TM
is a three phase and three
dimensional simulator. It can be used to simulate 1, 2 or 3 phase systems to predict and

manage fluid flow more efficiently. The reservoir simulator has been found to be the
most practical, less expensive, faster, more accurate and adequate when compared with
other methods.
The simulation software is used to generate simulated production data that
substitutes for actual field data. The reason for this is that the actual data is unsuitable for
testing of the method because of noise in the data. However, once the method has been
tested and verified by using simulated production data, this method should be available
for testing with real production data. For this purpose, we assume that the simulation
software represents the real reservoir closely, even though there may be some numerical
solution inaccuracies in the software. Most literature assumes that 10% error is expected
but we will try to avoid any errors when we use the software because the errors in the
numerical solution depend on various factors including the time step size and grid size.
The other main reason for using simulation software is because most of the previous
37
methods available for determining of relative permeability curves relied upon the other
methods to check and compare their results. Consequently, we cannot be sure about their
accuracy. In addition, there is no real field data available for relative permeability to
compare with.
We simulate the radial flow system using the Eclipse
TM
software with a real
reservoir size. The Production well is in the center and the injection wells in the
surrounding areas.
In a radial flow system, there are three main parameters that need to be specified;
(1) re is the blocks outer radius which will divide into several grid blocks in the
simulation software, (2) is the segment angle of the grid block in radians, (3) the
number of layers (we assumed there is one layer in all our examples for simplicity).
Therefore, we started with a simple case and then developed the idea as shown in the
steps described below.
a) We assumed there are six injection wells (the angle is ( =60o, 360/6)) around the
reservoir. We used injection wells instead of aquifer because it is easy to control the
injection wells by constant rate or constant pressure and we do not need to know the
properties for the aquifer.
We can operate the system under unsteady-state by keeping the P constant and
letting the flow rate change or keeping the injection rate constant and letting the pressure
vary. We used the first option P constant for achieving more accuracy with the used the
software. We used water as the injection fluid to displace oil from one production well in
the center. We assumed that the reservoir is saturated by oil before the injection, and we
divided (re) it into five grid blocks as shown in Fig.(3.1) and Fig.(3. 2).
38
Fig.(3.1) Three dimension shape from Eclipse with six injection

wells and one production well in the center (=60o)
Fig.(3.2) Two dimension shape from Eclipse with six injection

wells and one production well in the center (=60o)
39
We can indicate from the two Fig.(3.1) and Fig.(3.2) that the injection wells are not at the
end of the last grid blocks. We even asked the software to do this but we think that
happened because we divided the reservoir into five sections only and we used a large
angle value. That can cause several problems; (1) The software result will not be accurate
because large grid blocks, (2) The injection wells are not at the end of the reservoir so can
displace the entire hydrocarbon, (3) The distance between one injection well to others
sufficiently large. Therefore, we tried another approach as shown in the next step.
b) To avoid the problems in part (a), we increased the number of the injection wells to
50. Thus, we reduced the angle to =7.2o because the angle is equal to 360/ (No. of
injection wells). After we modified the program in the software, we got the result as
shown in Fig.(3.3) and Fig.(3. 4).
Fig.(3.3) Two dimension shape from Eclipse

result
40
We can see from Fig.(3.3) that the 50 injection wells are in the center of the last grid
blocks, and the production well is in the center.
Fig.(3.4) Three dimension shape from Eclipse result
Also, we can see from Fig.(3.4) that the 50 injection wells are in the center of the last
grid blocks and the production well is in the center.
This means that when we divide the external radius (re) to small grid blocks and
decrease the angle, it will give a better result. Therefore, we increased the number of
injection wells again to 100 and the angle became (=3.6o). We expected to get accurate
radial displacement for the reservoir fluid by making the distance between one injection
well to others sufficiently small as shown in Fig.(3.5) below.
41
Fig.(3.5) Three dimensional shape from Eclipse result
We can see from Fig.(3.5) that the (100) injection wells are in the center of the last grids
and the production well is in the center.
c) As we know from parts (a and b) that when we increased the injection wells and
decrease the angle can get accurate results. But at the same time, that makes our program
in the software more complex. For convenience and accuracy, we used a single slice
model as shown in Fig.(3.6) and Fig.(3.7) for most examples in this study because we got
exactly the same result when we used the whole reservoir as a model or when we used
just one slice as shown in Fig.(3.8). We still need to specify the angle value in this case;
therefore, we used (=1o), which means that there are 360 injection wells around the
reservoir.
42
Injection well
Production
well
re
Fig.(3.6) One slice shape for Radial

flow
Fig.(3.7) Radial flow but only for one slice

(From Eclipse result)
12
10
q m3/D
oil rate from 3D shape

water rate from 3D shape
6
oil rate from one silce
water rate from one silce
4
2
0
0
1000
2000
3000
4000
5000
6000
Time,day
Fig.(3.8) Time vs. rate for one slice and the whole reservoir
(The production data is from Eclipse result)
43
7000
Evaluation Technique:
The following approach is used for evaluating the Toth et al., (2006) method. The data
are generated by using the reservoir simulation software in the following manner.
1- Assume the relative permeability curves as input data.
2-Simulate the flow in the radial system by using Eclipse TM to generate the production
data as a result.
3- Recalculate the relative permeability curves by using Toth et al. method using
production data obtained from the software.
4-Compare the calculated relative permeability values with the assumed values to check
the accuracy.
44
CHAPTER 4
EVALUATION OF THE TOTH ET AL. METHOD FOR RADIAL
FLOW USING SIMULATED PRODUCTION DATA
_____________________________________________________________
Several case studies are presented in the following.
4.1 CASE 1: Effect of Reservoir Size on Toth et al. Method

There are some parameters which are affected by longer time periods as well as space
dimensions. Therefore, we have to take them into the consideration during the well
production history.5 In addition, we want to check the applicability of the Toth et al.
equations for different reservoir sizes instead of using only core samples. Also, changing
the reservoir size is changing the location of the water injection wells (the injection wells
located at the outer end of the reservoir) and changing the displaced area around the
wellbore where the radial flow occurs. However, we can control that by changing the
reservoir radius.
We simulated three examples as shown in Table 4.1 and the properties and
relative permeability curves which we assumed as shown in Fig.(4.1). Then, we changed
the radius of well influence re for different values; low, medium, and high. This can
change the pore volume and the size for the reservoir. Then, we ran the software to get
the production data for the three examples as shown in Figs.(4.1.1), (4.1.6), and (4.1.11).
45
We applied the calculation process for the Toth et al. method as described in
chapter two. We determined the constants (a, b, a1, b1) from the production data as shown
in Table 4.4, and then calculated the relative permeability curves for these examples as
shown in Fig.(4.1.5), Fig.(4.1.10), and Fig.(4.1.15).
Table 4.1 Petrophysical parameters for the case studies
Parameter
Example 4.1.1
Example 4.1.2
Example 4.1.3
re , m
75
300
360
h,m
14.76
14.76
14.76
Vp , m3
71103
1137652
1638218
0.26
0.26
0.26
k, md
100
100
100
P, kpa
1300
1300
1300
Bo
1.23
1.23
1.23
Bw
w, pa.s
0.0005
0.0005
0.0005
o, pa.s
0.001
0.001
0.001
46
0.080
Krw assumed
0.070
Kro assumed
0.060
kr
0.050
0.040
0.030
0.020
0.010
0.000
0.28
0.33
0.38
0.43
Sw
Fig.(4.1) Assumed relative permeability curves
for examples 4.1.1, 4.1.2, and 4.1.3
12
qo
qw
10
q m3/d
8
6
4
2
0
1000
2000
3000
4000
5000
Time,day
Fig.(4.1.1) Production data from Eclipse
for example 4.1.1
47
6000
7000
Table 4.2 Production data for example 4.1.1
Time
,day
qo m3/d
qw m3/d
0.1
10.15
0.02
0.4
9.1
1.3
qo
qw
dNP
dWp
0.016
0.3
9.62
0.018
2.88
0.005
8.78
0.012
0.9
8.94
0.014
8.048
0.01
8.71
0.009
2.7
8.75
0.011
23.63
0.03
12.1
8.71
0.006
8.1
8.71
0.008
70.61
0.06
36.4
8.71
0.01
24
8.71
0.011
211.86
0.28
109.3
8.27
0.64
73
8.49
0.32
619.52
23.93
319.66
6.27
3.46
210
7.27
2.05
1531.06
432.25
684.66
3.87
6.38
365
5.07
4.92
1852.63
1799.1
1049.66
3.04
7.39
365
3.45
6.89
1262.61
2515.0
1414.66
2.39
8.19
365
2.72
7.79
993.02
2844.8
1779.66
1.94
8.66
365
2.16
8.42
791.85
3076.42
2144.66
1.66
8.95
365
1.80
8.80
657.83
3215.56
2509.66
1.51
9.16
365
1.58
9.05
579.82
3306.89
2874.66
1.36
9.39
365
1.43
9.27
525.34
3386.77
3239.66
1.19
9.58
365
1.27
9.49
466.66
3464.58
3604.66
1.01
9.74
365
1.10
9.66
403.78
3527.49
3969.66
0.84
9.92
365
0.92
9.83
339.15
3589.25
4334.66
0.72
10.06
365
0.78
9.99
286.14
3648.73
4699.66
0.67
10.16
365
0.69
10.11
255.32
3691.56
5064.66
0.62
10.26
365
0.64
10.21
237.10
3727.58
5429.66
0.58
10.36
365
0.60
10.314
220.56
3764.89
5794.66
0.53
10.45
365
0.56
10.41
204.64
3800.31
6159.66
0.50
10.53
365
0.52
10.49
190.53
3831.13
48
Table 4.3 Np, Wp and Wi for example 4.1.1
Np
2.9
10.9
34.6
105.2
317.0
936.6
2467.6
4320.3
5582.9
6575.9
7367.8
8025.6
8605.4
9130.8
9597.4
10001.2
10340.4
10626.5
10881.8
11118.9
11339.5
11544.2
11734.7
NpBo
3.6
13.5
42.5
129.4
390.0
1152.0
3035.2
5313.9
6866.9
8088.4
9062.3
9871.5
10584.7
11230.8
11804.8
12301.5
12718.6
13070.6
13384.7
13676.3
13947.6
14199.3
14433.7
Wp
Wi
0.005
0.01
0.04
0.11
0.39
24.3
456.5
2255.7
4770.7
7615.5
10692.0
13907.5
17214.4
20601.2
24065.8
27593.3
31182.6
34831.3
38522.9
42250.4
46015.3
49815.6
53646.8
3.6
13.5
42.6
129.5
390.4
1176.3
3491.8
7569.6
11637.7
15703.9
19754.3
23779.0
27799.1
31832.0
35870.6
39894.8
43901.2
47901.9
51907.5
55926.7
59962.9
64015.0
68080.4
49
Wi/NP
1.0015
1.0014
1.0012
1.0009
1.001
1.02
1.1
1.4
1.6
1.9
2.1
2.4
2.6
2.8
3.
3.2
3.4
3.6
3.8
4.
4.2
4.5
4.7
Wi/Vp
0.0001
0.0002
0.0006
0.001
0.005
0.01
0.04
0.1
0.16
0.22
0.27
0.33
0.39
0.4
0.5
0.56
0.61
0.6
0.7
0.7
0.8
0.9
0.95
3.5
y = 4.1661x + 0.9865
R = 0.9985
Wi / Np
2.5
2
1.5
1
0.5
0
0.1
0.2
0.3
0.4
0.5
Wi/Vp
Fig.(4.1.2) Displacement equation for example 4.1.1
80000
70000
y = 10.496x1.0064
R = 1
60000
Wi
50000
40000
30000
20000
10000
0
0
1000
2000
3000
4000
5000
6000
7000
Time, day
Fig.(4.1.3) Cumulative water volume produced for example 4.1.1
50
100
calculated
assumed
krw/kro
10
0.1
0.01
0.3
0.35
0.4
Sw
0.45
0.5
Fig.(4.1.4) Relative permeability ratio (assumed and calculated) for

example 4.1.1
0.08
Krw calculated
kr
0.07
Kro calculated
0.06
Krw assumed
0.05
Kro assumed
0.04
0.03
0.02
0.01
0
0.29
0.34
0.39
0.44
0.49
Sw
Fig.(4.1.5) Relative permeability ratio (assumed and calculated) for
example 4.1.1
51
Example 4.1.2
10
9
qo
qw
8
7
q m3/d
6
5
4
3
2
1
0
0
5000
10000
15000
20000
25000
30000
Time,day
Fig.(4.1.6) Production data from Eclipse for example 4.1.2
2.3
2.1
y = 5.3059x + 0.9381
R = 0.9953
Wi/Np
1.9
1.7
1.5
1.3
1.1
0.9
0
0.05
0.1
0.15
0.2
Wi/Vp
52
0.25
300000
250000
y = 8.351x1.0108
R = 1
Wi
200000
150000
100000
50000
0
0
5000
10000
15000
20000
25000
30000
Time, day
10
calculated
assumed
krw/kro
0.1
0.01
0.001
0.28
0.33
0.38
Sw
Fig.(4.1.9) Relative permeability ration (assumed and calculated)
for example 4.1. 2
53
0.43
0.08
kw cal.
0.07
ko cal.
0.06
kw assumed
ko assumed
kr
0.05
0.04
0.03
0.02
0.01
0
0.29
0.34
0.39
0.44
Sw
Fig.(4.1.10) Relative permeability curves (assumed and calculated)
for example 4.1.2
Example (4.1.3)
10
qo
qw
q m3/d
7
6
5
4
3
2
1
0
0
10000
20000
30000
Time, day
40000

54
50000
2.4
2.2
y = 5.2803x + 0.9417
R = 0.9957
Wi/Np
2
1.8
1.6
1.4
1.2
1
0.8
0
0.05
0.1
0.15
0.2
0.25
0.3
W/i/Vp
250000
230000
y = 8.0823x1.0118
R = 1
210000
190000
Wi
170000
150000
130000
110000
90000
70000
50000
8000
13000
18000
23000
28000
Time, day
55
10
calculated
assumed
Welge
krw/kro
0.1
0.01
0.28
0.3
0.32
0.34
0.36
0.38
0.4
0.42
Sw
Fig.(4.1.14) Relative permeability ration (assumed and calculated) for
example 4.1.3
0.08
Krw calculated
0.07
Kro calculated
Krw assumed
0.06
Kro assumed
kr
0.05
0.04
0.03
0.02
0.01
0
0.29
0.34
0.39
0.44
Sw
example 4.1.3
56
Table 4.4 Constant parameters for examples

(4.1.1), (4.1.2), and (4.1.3)
Parameter
a
b
a1 m3/ sec
b1
Example 4.1.1
Example 4.1.2
0.98
4.16
1.21 e -4
1.0064
0.93
5.3
9.6 e -5
1.0108
Example 4.1.3
0.94
5.28
8.3 e -5
1.0118
We can conclude from this case study that the Toth et al. method can give consistent
results when the size of the reservoir is small ( re < 150 m) because we can see a good
match between assumed relative permeability curves and calculated curves by the Toth et
al. method as shown in Fig.(4.1.5) and a poor match when the reservoir size is large ( re
> 300 m) as shown in Fig.(4.1.15). When we increase the distance for injection wells
from the production well, the amount of water production will decrease unless we do not
increase the pressure drop as shown in Fig.(4.1.11).
To check further the accuracy of the simulation software. Numerical solutions give
good results when the grid size is small; therefore, we may be need to change the grid
size for the previous examples, especially example 4.1.3, because when we assumed the
reservoir radius large we kept the grid blocks size the same as shown in Figs. (4.1.16),
(4.1.17) and (4.1.18).
We can see from the Figs. (4.1.16) that the grid block length for the first example is
1.5 m, for the second example is 6 m and for the third example is 7.2 m. Therefore, we
decreased the grid block length for the third example to 1.2 m and then studied the effects
on relative permeability curves.
57
re = 75 m
re = 250 ft
Grids
is is
Gridslength
length
1.55mft
Fig.(4.1.16) Grid size for example 4.1.1 from Eclipse
re = 300 m
Grids
length
is is
Grids
length
6 m20 ft
58
Grids
length
Grids
length
is is
24ft
7.2 m
re = 360 m
re = 360 m
Gridslength
length
Grids
is is
1.2 m4ft
59
After we reduced the grid block size and ran the software, we still get a poor match as
shown in Fig.(4.1.20) and Fig.(4.1.21). Therefore, we concluded that the error was not
caused by the software and our program is accurate, but caused by the reservoir size.
100
calculated
assumed
Welge
krw/kro
10
0.1
0.01
0.3
0.35
0.4
0.45
0.5
Sw
for example 4.1.3 after reducing the grid size
0.08
Krw calculated
kr
0.07
Kro calculated
0.06
Krw assumed
0.05
Kro assumed
0.04
0.03
0.02
0.01
0
0.29
0.34
0.39
Sw
0.44
0.49
0.54
Fig.(4.1.21) Relative permeability curves (assumed and

calculated) for example 4.1.3 after reducing the grid size
60
4.2.1 The Effect of P Value on the Toth et al. Method

The reason for study of the pressure drops effect on the relative permeability curves is
because Toth et al. assumed high pressure drop. However, the real reservoirs produce
under low pressure drop. Also, we observe a direct proportion between the reservoir size
and pressure drop as we will approve later.
To study the P effect we used the same examples as in case 1, but we increased
the P value from 1300 kpa to the 3500 kpa. Then, we applied the Toth et al. method to
recalculate relative permeability curves.

Parameter
Example 4.2.1
Example 4.2.2
Example 4.2.3
rw, m
0.0435
0.0435
0.0435
re , m
75
300
360
h,m
14.76
14.76
14.76
Vp ,m3
71103
1137652
1638218
0.26
0.26
0.26
k, md
100
100
100
P, kpa
3500
3500
3500
Bo
1.23
1.23
1.23
Bw
w , pa.s
0.0005
0.0005
0.0005
o , pa.s
0.001
0.001
0.001
61
Example (4.2.1):
35
qo
qw
30
q m3/d
25
20
15
10
5
0
0
2000
4000
6000
time
4
3.5
y = 3.7853x + 0.9972
R = 0.9982
Wi/Np
3
2.5
2
1.5
1
0.5
0
0.2
0.4
Wi/Vp
0.6
62
140000
y = 27.81x1.0069
R = 1
120000
100000
60000
40000
20000
0
0
1000
2000
3000
4000
5000
time
100
calculated
assumed
10
krw/kro
Wi
80000
0.1
0.01
0.3
0.35
0.4
0.45
0.5
Sw
Fig.(4.2.4) Relative permeability ration (assumed and
calculated) for example 4.2.1
63
0.55
0.08
Krw calculated
Kro calculated
Krw assumed
Kro assumed
0.07
0.06
kr
0.05
0.04
0.03
0.02
0.01
0
0.29
0.34
0.39
0.44
0.49
0.54
Sw
for example 4.1
Example 4.2.2
30
qo
qw
25
q m3/d
20
15
10
5
0
0
5000
10000
15000
time,day
20000
25000
Fig.(4.2.6) Production data from Eclipse for example 4.2
64
30000
3.5
3
y = 4.8655x + 0.9849
R = 0.9973
Wi/Np
2.5
2
1.5
1
0.5
0
0
0.1
0.2
0.3
0.4
0.5
Wi/Vp
700000
600000
y = 23.178x1.0062
R = 1
500000
Wi
400000
300000
200000
100000
0
0
5000
10000
15000
20000
25000
30000
Time
65
10
calculated
assumed
Welge
krw/kro
0.1
0.01
0.28
0.3
0.32
0.34
0.36
Sw
0.38
0.4
0.42

for example 4.22
0.08
krw calculated
0.07
kro calculated
0.06
krw,assumed
kro,assumed
kr
0.05
0.04
0.03
0.02
0.01
0
0.29
0.34
0.39
Sw

for example 4.2.2
66
0.44
Example (4.2.3):
30
qo
qw
25
q m3/d
20
15
10
5
0
0
10000
20000
30000
40000
50000
60000
Time
4.5
4
y = 4.1044x + 0.9833
R = 0.9977
3.5
Wi/Np
3
2.5
2
1.5
1
0.5
0
0.2
0.4
Wi/Vp
0.6
67
0.8
1
Welge
0.9
Toth et al.
0.8
fw
0.7
0.6
0.5
0.4
0.3
0.2
0.29
0.34
0.39
0.44
Sw
Fig.(4.2.13) Water fraction for example 4.2.3 after breakthrough time
1600000
1400000
1200000
y = 22.379x1.0071
R = 1
Wi
1000000
800000
600000
400000
200000
0
0
10000
20000
30000
40000
50000
60000
Time
68
100
calculated
assumed
Welge
krw/kro
10
0.1
0.01
0.3
0.32
0.34
0.36
0.38
0.4
0.42
0.44
Sw
Fig.(4.2.15) Relative permeability ratio (assumed and calculated)
for example 4.2.3
0.08
Krw calculated
0.07
Kro calculated
0.06
Krw assumed
Kro assumed
kr
0.05
0.04
0.03
0.02
0.01
0
0.28
0.33
0.38
0.43
Sw
for example 4.2.3
69
4.2.2 Case 2: For High P Value

To check further the effect of pressure drop, we studied the P effect for the same
examples as in case 2, but we increased the P value from the 1300 to the 7500 kpa.
Then we ran the software to generate the well production data. Finally, we applied the
Toth et al. method to recalculate the relative permeability curves.
Parameter
Example 4.2.1
Example 4.2.2
Example 4.2.3
rw, m
0.0435
0.0435
0.0435
re , m
75
300
360
h,m
14.76
14.76
14.76
Vp , m3
71103
1137652
1638218
0.26
0.26
0.26
K, md
100
100
100
P,kpa
7500
7500
7500
Bo
1.23
1.23
1.23
Bw
w , pa.s
0.0005
0.0005
0.0005
o , pa.s
0.001
0.001
0.001
Swi
0.3
0.3
0.3
70
Example (4.2.1):
70
60
q m3/d
50
qo
qw
40
30
20
10
0
0
1000
2000
3000
4000
Time
5000
6000
7000
25
20
Wi/Np
15
10
0
0
Wi/Vp
71
500000
450000
400000
y = 60.115x1.0075
R = 1
350000
Wi
300000
250000
200000
150000
100000
50000
0
0
1000
2000
3000
4000
Time
5000
6000
7000
1000
calculated
assumed
100
krw/kro
10
1
0.1
0.01
0.001
0.3
0.35
0.4
Sw
0.45
0.5

for example 4.2.1
72
0.55
0.08
Krw calculated
0.07
Kro calculated
0.06
Krw assumed
Kro assumed
kr
0.05
0.04
0.03
0.02
0.01
0
0.29
0.34
0.39
Sw
0.44
0.49
0.54

for example 4.2.1
Example (4.2.2):
70
qo
qw
60
q m3/d
50
40
30
20
10
0
0
5000
10000
15000
20000
25000
Time,day
Fig.(4.2.6) production data from Eclipse for example 4.2.2
73
30000
4
3.5
y = 4.5929x + 0.9994
R = 0.996
Wi/Np
3
2.5
2
1.5
1
0.5
0
0.1
0.2
0.3
0.4
0.5
0.6
Wi/Vp
1400000
1200000
1000000
y = 50.063x1.007
R = 1
Wi
800000
600000
400000
200000
0
0
5000
10000
15000
20000
Time
74
25000
100
calculated
assumed
Welge
krw/kro
10
0.1
0.01
0.29
0.34
0.39
0.44
0.49
Sw
Fig.(4.2.9) Relative permeability ration (assumed and calculated) for
Example 4.2.2
0.08
EclipseKrw
0.07
EclipseKro
krw,assumed
0.06
kro,assumed
kr
0.05
0.04
0.03
0.02
0.01
0
0.29
0.34
Sw
0.39
0.44

for example 4.2.2
75
Example (4.2.3):
70
qo
60
qw
q m3/d
50
40
30
20
10
0
10000
20000
30000
Time
40000
50000
60000
4
3.5
y = 4.5929x + 0.9994
R = 0.996
Wi/Np
3
2.5
2
1.5
1
0.5
0
0.1
0.2
0.3
Wi/Vp
0.4
0.5
76
0.6
1
Welge
0.9
Toth et al.
0.8
fw
0.7
0.6
0.5
0.4
0.3
0.2
0.25
0.3
0.35
Sw
0.4
0.45
0.5
3500000
3000000
y = 48.748x1.0072
R = 1
2500000
Wi
2000000
1500000
1000000
500000
0
0
10000
20000
30000
Time
40000
50000
60000
77
100
calculated
assumed
Welge
krw/kro
10
0.1
0.01
0.29
0.34
0.39
Sw
0.44
0.49

for example 4.2.3
0.08
Krw calculated
0.07
Kro calculated
Krw assumed
0.06
Kro assumed
kr
0.05
0.04
0.03
0.02
0.01
0
0.29
0.34
0.39
Sw
0.44
0.49

for example 4.2.3
78
Table 4.7 Constant parameters for P = 3500 kpa

Parameter
Example(4.2.1)
Example(4.2.2)
Example(4.2.3)
0.99
0.98
0.98
b
a1, m3/sec
b1
3.78
0.00032
1.0069
4.86
2.68 e -4
1.0062
4.1
2.58 e -4
1.0071
Table 4.8 Constant parameters for examples P = 7500 kpa

Parameter
Example(4.2.1)
Example(4.2.2)
Example(4.2.3)
0.98
0.99
0.99
b
a1, m3/sec
4
6.9 e -4
4.59
5.7 e -4
3.96
5.6 e -4
b1
1.0075
1.007
1.0072
We observe that there is a direct correlation between the pressure drop and reservoir
size which effects the estimation of relative permeability curves. Also, we determined
from these examples that the Toth et al. method yield good results when the value of
pressure drop is high as shown in Fig.(4.2.5) even with large reservoir size. That is
because the main assumptions for Toth et al. are the rock and fluid properties are constant
during the test and the capillary pressure is negligible (for real reservoir). We can reach
these assumptions with high pressure as shown in Fig.(6.1).
4.3 CASE 3: Effect of Viscosity on The Toth et al. Method

The purpose for studying the viscosity effect is to determine the applicability of the Toth
et al. method to gas/oil or water/gas systems because the viscosity ratio is very high when
79
there is gas in the fluid system. The viscosity ratio should have a significant effect on
determining the relative permeability curves. However, many literatures (Mehdi H., 1988
and Johnson, 1959) claim insignificant viscosity ration effect on relative permeability.
To study this issue, we used the best example that we used in other cases (1 and 2).
Therefore, we chose the example with re = 75 m, and P = 7500 kpa. Then, we changed
the viscosity ratio to study the effects of viscosity on the Toth et al. method.

Parameter
Example 4.3.1
Example 4.3.2
Example 4.3.3
re , m
75
75
75
h,m
14.76
14.76
14.76
Vp , m3
71103
71103
71103
0.26
0.26
0.26
K , md
100
100
100
P, kpa
7500
7500
7500
Bo
1.23
1.23
1.23
Bw
w , pa.s
0.0005
0.0005
0.0005
o , pa.s
0.001
0.002
0.004
w /o
80
Example 4.3.1
70
60
50
q m3/d
qo
qw
40
30
20
10
0
0
1000
2000
3000
4000
5000
6000
7000
Time
3
2.5
y = 4.0813x + 0.9806
R = 0.9978
Wi/Np
2
1.5
1
0.5
0
0
0.1
0.2
0.3
Wi/Vp
81
0.4
1
0.95
0.9
fw
0.85
0.8
0.75
0.7
0.65
0.6
0.3
0.35
0.4
Sw
0.45
0.5
0.55
300000
250000
y = 60.05x1.0076
R = 1
Wi
200000
150000
100000
50000
0
0
1000
2000
3000
4000
Time
Fig.(4.3.4) Cumulative produced water volume for example 4.3.1
82
5000
1000
calculated
assumed
krw/kro
100
10
0.1
0.01
0.3
0.35
0.4
Sw
0.45
0.5
0.55
Fig.(4.3. 5) Relative permeability ration (assumed and calculated)

for example 4.3.1
0.08
Krw calculated
Kro calculated
Krw assumed
Kro assumed
0.07
0.06
kr
0.05
0.04
0.03
0.02
0.01
0
0.29
0.34
0.39
0.44
0.49
0.54
Sw
Fig.(4.3. 6) Relative permeability curves (assumed and calculated)
for example 4.3.1
83
Example 4.3.2
70
qo
qw
60
q m3/d
50
40
30
20
10
0
0
1000
2000
3000
4000
5000
6000
7000
Time
1.5
y = 7.3392x + 0.9899
R = 0.9894
1.4
Wi/Np
1.3
1.2
1.1
1
0.9
0
0.01
0.02
0.03
0.04
0.05
Wi/Vp
84
0.06
1
0.9
fw
0.8
0.7
0.6
0.5
0.4
0.3
0.32
0.34
0.36
0.38
0.4
0.42
0.44
Sw
400000
350000
y = 23.601x1.1115
R = 0.9998
300000
Wi
250000
200000
150000
100000
50000
0
0
1000
2000
3000
4000
5000
6000
7000
Time
85
1000
calculated
assumed
krw/kro
100
10
0.1
0.01
0.3
0.32
0.34
0.36
0.38
0.4
0.42
0.44
Sw
for example 4.3.2
0.08
Krw calculated
Kro calculated
Krw assumed
Kro assumed
0.07
0.06
kr
0.05
0.04
0.03
0.02
0.01
0
0.29
0.34
0.39
0.44
Sw
for example 4.3.2
86
Example 4.3.3
70
qo
qw
60
50
q m3/d
40
30
20
10
0
0
1000
2000
3000
4000
5000
6000
7000
Time
1.5
1.4
y = 14.078x + 0.9975
R = 0.9906
Wi/Np
1.3
1.2
1.1
1
0.9
0
0.005
0.01
0.015
0.02
0.025
0.03
Wi/Vp
87
0.035
140000
y = 10.085x1.1963
R = 0.9983
120000
100000
60000
40000
20000
0
0
500
1000
1500
2000
2500
3000
Time
1000
calculated
assumed
100
10
krw/kro
Wi
80000
0.1
0.01
0.3
0.35
0.4
Sw

for example 4.3.3
88
0.45
0.08
Krw calculated
Kro calculated
Krw assumed
Kro assumed
0.07
0.06
kr
0.05
0.04
0.03
0.02
0.01
0
0.29
0.31
0.33
0.35
0.37
0.39
0.41
0.43
0.45
Sw
for example 4.3.3
We can conclude from this case study that the Toth et al. method gives good results when
the viscosity ratio is close to one. The results become worse when the viscosity ratio
increases as shown in Fig. (4.3.6) and Fig.(4.3.17). The viscosity ratio has a significant
effect on determining relative permeability curves.
According to Marle, (1981) relative permeability is a function of the density and
viscosity ratio. Thus, the relative permeability curves must change when the viscosity
ratio changes.
In fact, Welge, (1952) demonstrated strong dependency of relative permeability
on the viscosity ratio as shown Fig.(4.3.18). Similar result is observed with the Toth et al.
method because their graphs change with the change of the viscosity ratio as shown in
Fig. (4.3.20) and Fig. (4.3.21):
89
25000
mo/mw=2
mo/mw=8
mo/mw=0.5
NP
20000
mo/mw=4
mo/mw=1
15000
10000
5000
0
0
20000
40000
60000
80000
100000
Wi
Fig.(4.3.18) Cumulative recovery of oil versus cumulative volume of
injected fluid for different viscosity ratio (Eclipse result data)
9
mo/mw=8
mo/mw=4
mo/mw=2
mo/mw=1
mo/mw=0.5
Wi/Np
6
5
4
3
2
1
0
0
0.2
0.4
0.6
0.8
1.2
Wi/Vp
Fig.(4.3.20) Displacement equation for different examples
With different viscosity ratios
90
1.4
100000
mo/mw=2
mo/mw=8
mo/mw=0.5
90000
80000
mo/mw=4
mo/mw=1
70000
Wi
60000
50000
40000
30000
20000
10000
0
0
2000
4000
time
6000
Fig.(4.3.21) Cumulative produced water volume for different

examples with different viscosity ratios
The main reason for changing Wedges and Toths graph with the viscosity ratio is
because the unfavorable mobility ratio (M). If M 1 oil is capable of traveling with a
velocity equal to or greater than the water and if M > 1 the water is capable of traveling
faster than the oil and as the water pushes the oil through the reservoir, the latter will be
by passed, (Dake L. P., 1978) as shown in Fig.(4.3.22).
The concepts mentioned above are defined as:
Water M obility
Kw
w
Oil M obility
Ko
o
Mobility Ratio
M 1
2
Water production increases as the viscosity ratio increases. Thus, oil production
decreases. In addition, Fig.(4.3.23) by Qin et al., (2009) shown that the water cut
increases when the oil become viscous. Therefore, Qin et al. tried to control the water cut
by decreasing the mobility ratio to achieve effective displacement.
91
M<1
M>1
Fig.(4.3.22) water saturation distribution as a function of distance

between injection and production wells for (a) ideal and (b) nonideal displacement.
Source: the graph introduced from (Dake L. P. 1978).
Fundamentals of Res. Eng. (p121)
Fig.(4.3.23) Water cut vs. production rate for heavy and

light oil reservoir
Source: Introduced from ( Qin et al., (2009))
Permission of the Society of Petroleum Engineers
92
water cut
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
VR=2
VR=8
VR=4
20
40
60
80
Total production
Fig.(4.3.24) Total production vs. water cut for the last examples
from Eclipse result data for different viscosity ratios
In another study, Wang et al., (2006) found that the oil relative permeability decreases
and the residual oil saturation increases as the viscosity ratio increases (oil viscosity
increase) as shown in Fig.(4.3.25) and Fig.(4.3.26).
Thus, we conclude that it is inadequate to keep the same relative permeability
curves and only change the viscosity ratio, because the relative permeability curves are a
function of the viscosity ratio. Therefore, to study the effect of the viscosity ratio we must
assume relative permeability curves that match with the viscosity ratio, and then study the
accuracy for the Toth et al. method even though some literature said there is no
significant effect on relative permeability curves when the viscosity ratio increases or
changes as Johnson et al. (1959) said has shown in Fig.(4.3.27).
93
Fig.(4.3.25) Effect of viscosity ratio on relative

permeability curves
Source: The graph introduced from (Wang et al., 2006).
Fig.(4.3.26) Viscosity vs. residual oil

Source: the graph introduced from (Wang et al., 2006)
94
Fig.(4.3.27) Relative permeability curves by JBN method

for different viscosity ratio
Source: The graph introduced (Johnson, 1959).
.
We tested the effect of viscosity ratio on the Toth et al. method and studied the
accuracy by using also the core sample data obtained from another study. This example
has a viscosity ratio equal to 10.7. First, we determined the relative permeability curves
by using the Toth et al. method and then tried to simulate another example by using the
simulation software assuming the same relative permeability curves for the core sample
as shown below.
95
Example 4.3.4
Laboratory example for core sample was introduced from Jones and Roszelle (1978).
Vp=31.13 ml
w=0.97 cp
o=10.45cp
K=35.4 md
Swi=0.35
p=100 psi
=0.215
Table 4.10 Production data for the core sample
Wi (m3)
Time (day)
Wp(m3)
Np (m3)
0.002
3.09E-06
0.004
0.000007
3.09E-12
3.09E-06
0.006
1.09E-05
7E-12
0.000007
0.008
1.53E-05
1.09E-11
7.8E-06
0.01
1.99E-05
1.528E-11
8.33E-06
0.01
2.79E-05
1.989E-11
8.7E-06
0.01
3.78E-05
2.79E-11
9.01E-06
0.04
9.95E-05
3.78E-11
9.32E-06
0.06
0.000177
9.95E-11
9.9E-06
0.10
0.000277
1.766E-10
1.01E-05
Table 4.11 Use Toth et al. method to recalculate relative permeability curves
Wi/Np
Wi/Vp
fo
fw
Sw
krw/kro
Y(S)w
krw
kro
0.09
0.81
0.18
0.4
0.02
2.40
0.0004
0.02
0.2
0.35
0.64
0.47
0.16
2.68
0.0016
0.01
1.39
0.3
0.2
0.79
0.51
0.36
2.84
0.002
0.0059
1.83
0.4
0.12
0.87
0.54
0.66
2.97
0.0025 0.0038
2.28
0.6
0.08
0.91
0.57
1.05
3.08
0.0027 0.0025
3.09
0.8
0.045
0.95
0.59
1.931
3.22
0.0029 0.0015
4.05
1.2
0.02
0.97
0.61
3.35
3.35
0.0031 0.0009
10
3.1
0.004
0.99
0.65
20.64
3.82
0.0036 0.0001
17.5
5.6
0.0014
0.998
0.66
62.70
4.13
0.004 6.39E-05
26.8
8.8
0.00061
0.999
0.67
151.425
4.40
0.0042 2.82E-05
96
30
y = 2.9811x + 0.4455
R = 0.9998
25
Wi/Np
20
15
10
5
0
0
Wi/Vp
10
0.0003
y = 0.0038x1.1546
R = 0.9999
0.00025
Wi
0.0002
0.00015
0.0001
0.00005
0
0
0.02
0.04
0.06
0.08
0.1
0.12
Time, Day
97
1000
Toth et al.
100
krw/kro
10
0.1
0.01
0.4
0.5
Sw
0.6
0.7
Fig.(4.3. 30) Relative permeability ratio for example 4.3.4
0.025
krw Toth et al
kro Toth et al
0.02
0.015
kr
0.01
0.005
0
0.35
0.45
0.55
0.65
Sw
Fig.(4.3. 31) Relative permeability curves for example 4.3.4
98
Example 4.3.5
This example is simulated by the software with the oil to water viscosity ratio=10.77.
Also, the assumed relative permeability curves introduced from Jones and Roszelle,
(1978) when the viscosity ratio is10.77.
Table 4.12 Petrophysical parameters for example 4.3.5
Parameter
Example1
rw, m
0.0435
re , m
75
h,m
14.76
Vp , m3
71103
0.26
K , md
100
P, kpa
7500
Bo
1.23
Bw
w , pa.s
0.00097
o , pa.s
0.01045
o / w
10.77
Swi
0.35
99
0.06
Krw assumed
0.05
Kro assumed
0.03
0.02
0.01
0.00
0.35
0.45
0.55
Sw
0.65
0.75
Fig.(4.3.32) Assumed relative permeability curves as input data in

Eclipse for 4.3.5 example
120
qo
qw
1000
2000
100
80
m3/d
kr
0.04
60
40
20
0
0
3000
4000
5000
6000
7000
Time, day
100
18
16
y = 2.0447x + 0.9797
R = 0.9999
14
Wi/Np
12
10
8
6
4
2
0
0
4
Wi/Vp
600000
y = 54.012x1.0365
R = 0.9961
500000
Wi
400000
300000
200000
100000
0
0
2000
4000
6000
8000
Time, day
Fig.(4.3.35) Cumulative water volume produced for example 4.3 5
101
100
calculated
assumed
krw/kro
10
0.1
0.01
0.3
0.4
0.5
0.6
0.7
0.8
Sw
for example 4.3.5
0.07
Krw calculated
Kro calculated
Krw assumed
Kro assumed
0.06
0.05
kr
0.04
0.03
0.02
0.01
0
0.3
0.4
0.5
0.6
0.7
Sw
Fig.(4.3.37) Relative permeability curves (assumed and
calculated) for example 4.3.5
102
0.8
We can conclude from the previous examples that the Toth et al. method is very
sensitive to viscosity ratio and accurate for the viscosity ratio close to one, even if the
value of the viscosity ratio is not significantly high (o / w =10.77) as shown in
Fig.(4.3.37). However, this is because we changed the viscosity ratio without changing
the assumed relative permeability curves as input data in Eclipse
TM
as shown in
examples 4.3.2 and 4.3.3.
4.4 CASE 4: Evaluating the two critical equations for the Toth et al. method
The first important equation is the displacement equation, which should be a straight line.
The intercept a must be less than one because it is the oil fraction at breakthrough time
and the slope b must be greater than one. Also, we can see from the Toth et al.
formulation (see chapter 2) that the saturation of displaced fluid and relative permeability
ratio depends on these constants.
Wi
W
ab i
Np
Vp
(Displacement equation)
Toth et al.5 state that this equation is valid when the reservoir is homogenous and if it is
not we will get different slope values over different ranges for this equation and the slope
will be constant for each section as shown in Fig.(4.4.1) and Fig.(4.4.2).
.
103
3.0
A-225.
2.5
Wi/Np
2.0
Slope 1
1.5
Slope 2
1.0
0.5
0.0
0
50000
0.2
100000
150000
0.4
200000
250000
0.6
W
Wi/V
,m
i
p3
Fig.(4.4.1) Displacement equation for field data (well A-225)

Source: The data introduced from ( Toth, 2006)
4.5
A-710.
4.0
3.5
Wi/Np
3.0
2.5
Slope 1
Slope 2
2.0
1.5
1.0
0.5
0.0
0
100000
200000
0.4
300000
400000
0.8
3
WW
i/Vip, m
Fig.(4.4.2)
equation
data A-710,
(well A-710)
Fig.Displacement
(4.4.2), Source:
Field for
datafield
for well
Ref. (19)
Source: The data introduced from (Toth, 2006)
104
We used Eclipse
TM
software to evaluate the Toth et al. method and we assumed that
the reservoir is homogenous for all previous examples. However, we still have the same
problem (more than one slope) as shown below.
2.1
1.9
Wi/Np
1.7
slope 1
1.5
slope 2
1.3
1.1
0.9
0
0.05
0.1
Wi/Vp
0.15
0.2
0.25
Fig.(4.4.3) Displacement equation for simulated example by use

Eclipse
This signals that there is another reason for that issue. In fact, Huppler, (1970) state that
the heterogeneities have insignificant effects on relative permeability curves. However,
the effect become significant when there are channels.
We observed that the reservoir performance some time behaves the same after a
short period following the breakthrough time as before the breakthrough time. Therefore,
the trend will be very different (other slope appears). Also, we can prove Welge's
method, (Collins, 1976) because the plot cumulative water (wetting fluid) influx (Wi)
versus cumulative oil (non-wetting fluid) production (Np), yields a straight line with slope
(45o) before the breakthrough time (ta) and the slope will change and remain constant for
each small segment after the breakthrough time as shown in following equation:
105
Wi a b ln N p for t ta
Welge said the graph will change however if the viscosity ratio changes.
180000
160000
After tB
140000
120000
Np
100000
80000
60000
40000
before tB
20000
=45o
0
0
50000
100000
150000
200000
250000
300000
350000
400000
450000
Wi
Fig.(4.4.4) Wedges plot for simulated Example by Eclipse

2.1
1.9
Wi/Np
1.7
1.5
1.3
1.1
0.9
0
0.05
0.1
0.15
0.2
0.25
Wi/Vp
Fig.(4.4.5) Toths plot for simulated example by Eclipse
106
80000
all data
after tb
70000
A-225
60000
40000
30000
20000
B
t
10000
0
0
50000
100000
150000
200000
250000
Wi
Fig.(4.4.6) Welges plot for field data for well A-225.

Source: the data introduced from (Toth, 2005)
100000
90000
all data
after tb
80000
A-710
70000
60000
Np
Np
50000
50000
40000
30000
20000
tB
10000
0
0
50000
100000
150000
200000
250000
300000
350000
Wi
Fig.(4.4.7) Welges plot for field data for well A-710

Source: the data introduced from (Toth, 2005)
107
400000
We can see from Fig.(4.4.6) and Fig.(4.4.7) that there are three separate regions A, B and
C. Region (A) expresses the data before the breakthrough time. Region (B) expresses the
data after the breakthrough time, but the data behaves the same as before breakthrough
time (the same slope) because of the few amount of water production . This especially
happens when the viscosity ratio is low or close to one as shown in Fig.(4.4.3) but that
does not happen when the viscosity ratio increases. When the viscosity ratio increases,
the water cut increase because the mobility ratio also increases as we discussed before as
shown in Fig.(4.3.28) and Fig.( 4.3.34). Region (C) expresses the data after the
breakthrough time with different slopes as Welge theoretically explained.
Toth et al. depended on Welge's method when they derived their equations. In
addition, Toth et al. theoretically proved that the data after the breakthrough time can fit
with a straight line if they plot Wi /Np vs. Wi /Vp . Therefore, we think the data after the
breakthrough time behave the same as before breakthrough time and that caused the
second slope in the displacement equation for Toth et al. method. Also, the low pressure
drop, large reservoir size (long production time) or viscosity ratio greater or less than one
can cause the varies slopes as approved before (see Fig.(4.1.7), Fig.(4.1.12)) and
Fig.(4.1.20).
The second important equation of the Toth et al. method is an empirical equation
for cumulative water influx versus time.
Wi a1t b
where a1 and b1 are empirical constants. Also, b2 should be greater than one to make
Eq.(A.8) (see chapter 2) work. However, we observed that 1 < b1< 1.5 can give the best
results with the power- law equation. Therefore, it is very sensitive for the power
108
exponent value (b1). This means a little change in (b1) value will cause a significant effect
on the calculated relative permeability curves. Fig.(4.4.8) show the relative permeability
curves when the value of b1 is 1.007 (good matching), but if we change the value of b1 to
1.0127, we get poor matching as shown in Fig.(4.4.8).We can conclude from these results
and the Toth et al. formulation that individual relative permeabilities depend on the value
of these constants (a1 and b1) appreciably.
0.08
Krw calculated
Kro calculated
Krw assumed
Kro assumed
0.07
0.06
kr
0.05
0.04
0.03
0.02
0.01
0
0.29
0.34
0.39
Sw
0.44
0.49
Fig.(4.4.8) the value of (b1=1.007) for simulated example by Eclipse

0.08
Krw calculated
Kro calculated
Krw assumed
Kro assumed
0.07
0.06
kr
0.05
0.04
0.03
0.02
0.01
0
0.28
0.33
Sw
0.38
0.43
Fig.(4.4.9) the value of (b1=1.0127) for the same pervious simulated example
by Eclipse
109
4.5: Conclusions
The following conclusions are made based on the studies carried out in this thesis:
The Toth et al. method gives very good results for cases involving high pressure
drop and small reservoir size.
The Toth et al. method gives poor results for large reservoir size (long production
time) unless the pressure drop is increased.
The Toth et al. method gives accurate results when the fluid viscosity varies.
The different slope regions occurring in the displacement equation for the Toth et
al. are not caused by the heterogeneity . However, The Toth et al. method does
not yield a straight line when the viscosity ratio is less or greater than one, large
reservoir (long production time) size, and low pressure drop (less than 1300 kpa)
conditions. In addition, we can take the average straight line for these slopes or
the straight line for the first slope only.
The power-law equation of Toth et al. is very sensitive for the exponent value. A
small change can give different relative permeability curves. Also, we observed
that the best value of the b1 is 1< b1< 1.5.
The fluid saturation and relative permeability ratio mostly depend on a and b
constants.
110
CHAPTER 5
CONSIDERATION OF SKIN FACTOR AND HETEROGENEITY
DURING THE ESTIMATION OF RELATIVE PERMEABILITY
CURVES FOR UNSTEADY STATE RADIAL DISPLACEMENT
___________________________________________________________
The focus of this chapter is to test the applicability of the Toth et al. method when there is
heterogeneity or skin factor effect. Also, we want to study the effect of heterogeneity or
skin factor on determination of relative permeability.
5.1 Effect of Skin Factor.

Frequently, materials such as mud filtrate, cement slurry, clay particles during, water
flooding, completion or workover and other factors can reduce the permeability in the
formation around a wellbore (Ahmed, T., 2001) and the well production. The wellbore
region can be stimulated by acidizing or hydraulic fracturing, which increases the
permeability around the wellbore. This effect is considered a near-wellbore issue and the
region is called the skin zone, which can extend from a few inches to several feet
(Ahmed, T., 2001) as shown in Fig.(5.1).
Skin factor is a numerical value ranging from -6 for an infinite-conductivity
massive hydraulic fracture to more than 100 for a poorly executed gravel pack. This
value
is
highly
dependent
on
the
value
of
the
formation
capacity
kh
(www.glossary.oilfield.slb.com). We can determine the skin factor S from the following

equation:
111
k
r
S
1 ln skin - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 5.1
k skin rw
pressure profile
Improved
K
Damaged
Zone
Kskin
P
<0
P
>0
Undamaged
Zone
K
Reduced
K
rw
rskin
Fig.(5.1) Skin Zone

Source: Introduced from (Ahmed, T., 2001)
Positive Skin Factor, S > 0 happens when the kskin is less than k, which indicates that a
damaged zone near the wellbore exists (Ahmed, T., 2001). The value of the skin factor
increases as kskin decreases and as the radius of the damage rskin increases.
Negative Skin Factor, S < 0 happens when the kskin is higher than k, which indicates an
improved wellbore condition (Ahmed, T., 2001).
Zero Skin Factor, S = 0 occurs when no alternation in the permeability around the
wellbore is observed, i.e., kskin = k (Ahmed, T., 2001).
To study the effect of the skin factor on the relative permeability curves, we
simulated examples using a reservoir simulator. The rock and fluids properties for the
simulated examples are presented in Table 5.1. We assumed that the reservoir is
homogenous, unsteady, two-phase, immiscible, incompressible, and radial flow. We
started with the skin factor value equal to zero, and then assumed different positive skin
112
values directly by using specific keywords in the completion part in the Eclipse TM. Then,
we ran the software to generate the production data. From this production data we
estimated the relative permeability curves for different positive skin values as shown in
Fig.(5.4) and Fig.(5.5).
Because Toth et al. ( 2005, 2006) depends on well production data, we expect a
strong effect when we consider the positive skin (permeability damage). That effect
happened because we determined that the relative permeability curves depend on
production data. We know that any damage for the permeability around the wellbore will
reduce the production data. Therefore, we can see from Fig.(5.4), there are significant
reductions for the relative permeability curves. We can conclude from these figures that
there is a significant effect of the positive skin factor on the estimation of relative
permeability curves. In addition, we conclude that a significant change in the relative
permeability ratios occurs when the water saturation increases. However, these changes
were not drastic when compared with individual relative permeabilities as shown in
Fig.(5.5).
Reservoir Simulation Approach

Because it is difficult to collect actual production data for the wells with prescribed skin
factor values or with given permeability distribution, we used one of the most popular
reservoir simulators in the oil and gas industry, which is Eclipse
The purpose of using the Eclipse
TM
TM
by Schlumberger.
software was to generate production data that
substitutes for the actual field data. The reason for this is that the actual data is usually
noisy and therefore, unsuitable for testing of most methods. We evaluated the simulated
113
example using our new technique as explained before in chapter (3), then we applied the
Toth et al. (2006) method to determine the relative permeabilities for radial flow, as
shown in example 5.1.
The formulations that Eclipse
TM
used to consider the skin factor are the
following.
In radial geometry, the formula used to calculate the corresponding transmissibility factor
is (Eclipse manual, 2009):
cKh
Twj
r12
r22 r12
5.2
ln( r1 / r2 ) 0.5 S
Where, Twj is the corresponding transmissibility factor. r1 and r2 are the inner and outer
radii of the block. is the segment angle of the grid block.
In addition, the Eclipse TM software defines the steady-state productivity index (J)
as:
Qp
Pd Pw
5.3
Where, Qp is the production rate of the chosen phase and Pd is the pressure at the
drainage radius. Also, Eclipse
TM
derived the relationship between the productivity index
and transmissibility factor, assuming a steady-state radial Darcy flow with uniform
mobility throughout the region bounded by the drainage radius as (Eclipse manual,
2009):
ln (r /r ) S
5.4
J Twj M pj o w
j
ln (rd /rw ) S
Where
denotes the summation over all the connections j belonging to the well.
114
The Eclipse
TM
software always assumes the steady-state condition between the time
steps as a numerical approximation to solve the equations even under the unsteady-state
conditions.
Example 5.1
This example was simulated by Eclipse TM. The rock and fluid properties are presented in
Table 5.1. We assumed that the reservoir is homogenous, unsteady-state, two-phase,
immiscible, incompressible, and radial flow. We started with (S=0), then we assumed
different skin factor values.
Table 5.1 Petrophysical parameters for
Example 5.1
q m3/d
Fig. (5.2) Three dimensional shapes by Eclipse

10
9
8
7
6
5
4
3
2
1
0
qo
qw
10000
20000
30000
time,day
Fig.(5.3) Production data by Eclipse for
example 5.1
115
Parameter
Example
rw, m
0.043
re, m
75
H, m
14.76
Vp
71103
0.26
K,
100
P, kpa
1300
Bo
Bw
0.001
0.10
Swi
0.3
0.09
0.08
Krw,S=0
krw, S=0
Kro,S=0
kro, S=0
0.07
krw,S=1
kro,S=1
0.06
krw,S=2
kro,S=2
krw,S=4
kro,S=4
krw,S=6
kro,S=6
0.34
0.36
kr
0.05
0.04
0.03
0.02
0.01
0
0.3
0.32
Sw
Fig.(5.4) Relative permeability curves for different Skin values,

example 5.1
100
krw/kro
10
S=0
S=1
S=2
S=4
S=6
0.1
0.01
0.001
0.3
0.32
Sw
0.34
0.36
Fig.(5.5) Relative permeability ratios for different Skin values,

example 5.1
116
Case 5.1.1: Skin Factor and Capillary pressure effect

Relative permeability and capillary pressure are important properties for reservoir
engineering in that they have a major influence on the performance of two phase fluid
reservoirs under development. Also, all numerical simulations of the reservoir
performance require the input of relative permeability and capillary pressure values.
However, most relative permeability methods did not include the capillary pressure
effect.
Because the Toth et al. method ignored the capillary pressure effect, we expect
additional effects on relative permeabilites with skin effects included. However, Toth et
al. (2006) method can give good estimation for the relative permeability in case there is
capillary pressure effect also because the capillary pressure can effect on the well
production data that used for determination of the relative permeability.
14
12
Pc ,psi
10
8
6
4
2
0.3
0.35
0.4
Sw
Fig.(5.6) Capillary pressure
vs. water saturation
117
0.1
0.09
0.08
0.07
kr
0.06
Krw,S=0
krw,S=0
krw,S=0
krw,S=2
Kro,S=0
kro,S=0
kro,S=2
krw,S=4
kro,S=4
krw,S=6
kro,S=6
krw,S=1
kro,S=1
0.05
0.04
0.03
0.02
0.01
0
0.3
0.32
0.34
Sw
0.36
Fig.(5.7) Relative permeability curves for different Skin

values with capillary pressure effect
100
S=0
S=1
S=2
S=4
S=6
10
krw/kro
0.1
0.01
0.001
0.3
0.32
Sw
0.34
0.36
Fig.(5.8) Relative permeability ratios for different Skin

values with capillary pressure
118
0.14
krw,S=0,with pc
0.12
kro,S=0,without pc (pc=0)
0.1
krw,S=0 without pc (pc=0)
0.08
kr
kro,S=0 with pc
0.06
0.04
0.02
0
0.3
0.32
0.34
0.36
Sw
Fig.(5.9) Relative permeability with Skin equal to zero

with and without capillary pressure include
The capillary pressure is the difference between the non-wetting phase pressure and the
wetting-phase pressure (Pc = Pnw - Pw). We can conclude from Fig.(5.7) and Fig.(5.9)
that capillary pressure can have positive effect on the oil relative permeability curves and
negative effect on the water relative permeability when the capillary pressure is positive.
In contrast, the capillary pressure can have a negative effect on the oil relative
permeability and positive effect on the water relative permeability when the capillary
pressure is negative. Therefore, we can see in Fig.(5.9) that the oil relative permeability
increases more because the capillary pressure increased the oil recovery (P0 > Pw) for
positive capillary pressure values. We found the same result in a previous experimental
study (Alam,1980) but without including the skin effect. However, the effect will
decrease or increase depending on the capillary pressure values.
119
Fig.(5.10) Relative permeability curve with and without

capillary pressure by experimental work.
Source: introduced from (Alam,1980)
Case 5.1.2: Negative Skin Factor

Wellbores intersecting fractures may exhibit enhanced permeabilities as the fractures
offer much greater conductive paths to the fluids around the wellbore, thus enhancing the
permeability. This situation may also be required as part of effective reservoir
management. Therefore, hydraulic fractures or acidizing workovers can improve the
permeability around the wellbores.
In Eclipse
TM
software we can apply the positive skin factor directly in the
completion part using specific keywords, but Eclipse TM does not run with a negative skin
value. Therefore, we simulated negative skin by assuming high permeability with
different radii around the wellbore manually to form a permeability enhanced zone as
shown in Fig.(5.11). Then, we determined the skin using Eq.(5.1). After we simulated
different negative skin values, we ran the software to generate the production data. From
120
this production data we estimated the relative permeability curves for different negative
skin values using the Toth et al. method as shown in Fig.(5.12). Therefore, we can see
from Fig.(5.12) that there are significant improvements for the relative permeability
curves. The improvement for the relative permeability curves came from increasing well
production because the permeability increased around the near-wellbore zone. We can
conclude from this figure that there is a significant effect of the negative skin factor on
the estimated of relative permeability curves. Also, we can see that the negative skin
factor increases the relative permeabilities curves.
ra
ka
r
w
re
k
Fig.(5.11) Cross section for the reservoir with skin zone

0.16
krw,S=0
kro,S=0
krw,S=-2
kro,S=-2
krw,S=-2.78
kro,S=-2.78
krw,S=-2.4
kro,S=-2.4
0.14
0.12
kr
0.1
0.08
0.06
0.04
0.02
0
0.3
0.32
0.34
Sw
0.36
0.38
Fig.(5.12) Relative permeability curves for different negative

Skin values when the viscosity is 10cp
121
5.2 Effect of Heterogeneity on Relative Permeability

Most reservoirs involve varying degrees of heterogeneity. However, all the methods for
estimating relative permeability assumed homogenous reservoirs.
There are many studies about the heterogeneity effect but the most important
study is by Huppler (1970). He used a numerical technique to investigate the effects of
heterogeneity on core samples and found that the heterogeneities have an insignificant
effect on the relative permeability curves and as the heterogeneities become channel their
influence on relative permeability curves becomes pronounced (Huppler, 1970).
However, Huppler did not show the distribution for the permeabilities and the channels.
The second important study is by Darryl et al. (Fenwick, 2001). They observed a
significant effect of heterogeneity on relative permeability using the JBN method. Darryl
et al. did not consider channels in the reservoir. Also, the Huppler and Darryl et al.(2001)
studies were for linear displacement. In contrast, we were interested in investigating the
heterogeneity effect using the Toth et al. radial method including channels by using a
reservoir simulator with real reservoir size. We used the Toth et al. method because it is a
unique direct method for radial flow. We used the same example 6.1 to study the
heterogeneity. Concurrently, we generated different random values for the permeabilities.
We started with mean permeability equal to 100 md and standard deviation equal to 30 as
shown in Fig.(5.13) and Fig.(5.14). Then, we tested another example with the mean
permeability equal to 500 md and standard deviation equal to 100 as shown in Fig.(5.21).
We assumed that the permeability is a normal distribution. Also, we assumed that there is
a short or a long channel in the reservoir as shown in Fig.(5.17) and Fig.(5.22).
After generating the production data from the software, we determined the relative
permeability curves including the heterogeneity effect with or without channels as shown
in Fig.(5.15), Fig.(5.19), Fig.(5.23) and Fig.(5.26). Also, we can see that the
heterogeneity reduced the relative permeability curves because the permeabilities
distributed randomly, which indicates non-uniform displacement as shown in Fig.(5.14).
In addition, there are increases in the relative permeabilities when there is a channel with
the heterogeneity because channel means high permeability and an increase the
production data. At the same time, there is an effect on the relative permeability curves
122
by the dimension and permeability of the channel as shown in Fig.(5.23) and Fig.(5.26).
We can see from this figure that the dimension and permeability of the channel are
directly proportional to the relative permeability curves.
We observed from this study that the heterogeneities have an insignificant effect on
the relative permeability curves. However, the effect becomes significant when there are
channels.
Note that these examples, we assumed the permeabilities in and Z directions are
zero or very low to investigate radial flow even with the heterogeneity effect.
Fig.(5.13) The distribution for the permeability by Eclipse

(Mean=100, S.D=30)
123
Fig.(5.14) Histogram for random permeability values

(Mean=100, S.D=30)
krw, k=100,homogenous
0.1
kro,K=100 homogenous
0.08
krw, mean k=100, St.D=30

kro, mean k=100, St.D.=30
kr
0.06
0.04
0.02
0
0.3
0.32
Sw
0.34
0.36
Fig.(5.15) Relative permeability curves for homogenous

(k=100 md) and heterogeneous (Mean=100, S.D. =30)
124
100
k=100 ,homogenous
mean k=100, St.D.=30
krw/kro
10
0.1
0.01
0.3
0.31
0.32
0.33
0.34
0.35
0.36
0.37
Sw
Fig.(5.16) Relative permeability ratios for homogenous (k=100md)

and heterogeneous (Mean=100, S.D. =30)

(Mean=100, S.D=30) with short channel
125
Fig.(5.18) 2 D. distribution for the permeability by Eclipse

(Mean=100, S.D=30) with short channel
0.12
krw,k=100 homogenous
kro,k=100 homogenous
0.1
krw,Mean=100,S.D=30
kro,Mean=100,S.D=30
0.08
kr
krw,short cha.=400
kro,short cha.=400
0.06
krw,short cha. 1000

kro,short cha. 1000
0.04
0.02
0
0.3
0.31
0.32
0.33
0.34
0.35
0.36
Sw
Fig.(5.19) Relative permeability curves for homogenous (k=100 md),

heterogeneous (Mean=100, S.D. =30) and heterogeneous with short
channel (k=400 md and 1000 md)
126
0.37
100
k=100 homogenous
Mean=100,S.D=30
10
short cha. 400
krw/kro
short cha. 1000
0.1
0.01
0.3
0.31
0.32
0.33
0.34
0.35
0.36
0.37
Sw
Fig.(5.20) Relative permeability ratios for homogenous (k=100

md), heterogeneous (Mean=100, S.D. =30) and heterogeneous
with short channel (k=400 md and 1000 md)
Fig.(5.21) Histogram for random permeability values

(Mean=500, S.D=100)
127

(Mean=500, S.D=100) with long channel
0.12
krw,M=500,S.D=100
kro,M=500,S.D=100
krw,one cha.2000
kro,one cha. 2000
krw,k=500,homogenous
kro,k=500,homogenous
krw,long cha.3000
kro,long cha.3000
krw,mean=500,long cha.=2000
kro,mean=500,long cha.=2000
0.1
kr
0.08
0.06
0.04
0.02
0
0.3
0.32
0.34
0.36
0.38
Sw
Fig.(5.23) Relative permeability curves for homogenous (k=500

md), heterogeneous (Mean=500, S.D. =100) and heterogeneous
with short and long channel (k=2000 md and 3000 md)
128
10000
k=100
M=100,S.D=30
short cha. 1000 with mean 100
M=500,S.D=100
mean=500,one cha.2000
k=500,homogenous
mean=500,long cha.=3000
mean=500,long cha.=2000
1000
krw/kro
100
10
0.1
0.01
0.3
0.32
0.34
0.36
0.38
0.4
Sw
Fig.(5.24) Relative permeability ratios for homogenous (k=500,

k=100 md), heterogeneous (Mean=100, 500, S.D.=30,100) and
heterogeneous with short and long channel (k=2000, 1000, 3000 md)

(Mean=100, S.D=30) with different long channels
129
0.14
krw,100,100
kro,100,100
krw,100,200
kro,100,200
krw,100,50
kro,100,50
krw,100,30
kro,100,30
krw,100,10
kro,100,10
0.12
0.1
kr
0.08
0.06
0.04
0.02
0
0.3
0.32
0.34
0.36
0.38
Sw
Fig.(5.26) Relative permeability curves for homogenous

(k=100 md) and different long channels
100
10
krw/kro
1
k=100,100
k=100,10
0.1
k=100,30
0.01
k=100,50
k=100,200
0.001
0.3
0.31
0.32
0.33
0.34
0.35
0.36
0.37
Sw
Fig.(5.27) Relative permeability ratios for homogenous
(k=100 md) and different long channels
130
0.38
5.3: Conclusions
We make the following conclusions:
We observed that the Toth et al. method works satisfactorily even for
heterogeneous reservoirs or when there is skin effect.
There is a significant effect of the skin factor on the relative permeability

curves. Such effect increases when the capillary pressure increases.
The field data can give a better estimation for the relative permeabilities than
core data when there is skin because the rock properties change during core
preparation. Therefore, we recommend modifying the relative permeability
methods for the skin factor when we use core samples.
The heterogeneities have an insignificant effect on the relative permeability

curves. However, the effect becomes significant when there are channels.
131
CHAPTER 6
MODIFICATION AND GENERALIZATION OF THE TOTH ET AL.
METHOD FOR COMPRESSIBLE AND INCOMPRESSIBLE FLUIDS
AND EVALUATION BY MEANS OF A RESERVOIR SIMULATOR
_____________________________________________________________
Relative permeability for gas/oil and gas/water systems under reservoir conditions
requires the consideration of the effect of fluid compressibility. However, most previous
studies assumed incompressible fluid formulations to determine the relative permeability
curves by assuming that compressible fluids behave like the incompressible fluids at high
pressures (Johnson, 1959 adn Welge, 1952). Our objective is to remove the
incompressible fluid assumption.
There are many methods to determine relative permeability. We reviewed some of
these methods that helped us to develop our technique. One of these methods is the Civan
and Donaldson 31 method. This method is a semi-analytic method for two immiscible and
incompressible fluids with unsteady state displacement and including the capillary
pressure. Civan and Donaldson (1989) relied on the Darcy equation and the material
balance, and simultaneous solution of the differential equations to derive their method.
Civan and Evans (1991) extended the Civan and Donaldson method by including the
non-Darcy effect and the interfacial drag force when there are gas and liquid phases.
Also, they used the pseudo-pressure formulation to convert the compressible and slightly
132
compressible flow equations to incompressible fluid type equations. Initially, they

assumed that the viscosity and formation volume factor are constant and the density is
variable with changing pressure. Later on, Civan and Evans (1993) came up with a
technique to determine relative permeability for compressible fluids without such
limitations of the previous work. This method is now applicable for unsteady-state nonDarcy displacement of variable property fluids.
In this study, we developed a direct method, which depends on Darcys law to
estimate the relative permeability for compressible and incompressible fluids. Also, we
applied the pseudo-pressure formulation to convert the compressible fluid equations to
incompressible fluid type equations. Direct methods are based on the integral
formulations of the equations describing fluid pressure and saturation for interpretation of
the tests with laboratory cores under unsteady-state conditions. Consequently, these
methods do not require numerical solutions for differential equations involved in indirect
methods or reservoir core history matching. At the same time, we present four examples
taken from literature under different conditions to study the accuracy for the new method
by using a commercial reservoir simulator. In addition, we check the accuracy of the new
method depending on our simulator technique and compare the results with other relative
permeability methods.
Formulation
The formulations required for processing of the displacement data and determination of
relative permeability and relative permeability ratio from radial and linear, two-phase
fluid displacement in laboratory core sample is derived below. Also, we assumed that the
133
reservoir undergoes a one-dimensional, isothermal, and unsteady-state flow of two

immiscible and compressible or incompressible fluids, in horizontal homogeneous and
isotropic porous media. We also assumed that the capillary pressure is negligible
justifiable under the elevated pressure conditions of reservoirs. Thus, viscosity, density
and the formation volume factor change with pressure change. This leads to a
generalization of the Toth et al. method for variable property fluid system
We used pseudo-pressure instead of regular pressure to convert the equations
from variable property fluids to constant variable property fluids type. For this purpose
we applied the definition for pseudo-pressure by Civan and Evans (1991) as shown in
Eq.( 6.1) instead of the Al-Hussainy et al. (1966) definition as shown in Eq.( 6.2)
because equation (6.1) is more general which is applicable for gas or slightly
compressible liquids than equation (6.2).
p2
1
dp 6.1
p1 B
m( p )
p2
p
dp 6.2
p1 Z
m( p ) 2
There are many studies describing the relationship between pressure and 1/B, as
we summarize in Fig.(6.1). We can conclude from this figure that the relationship
between the pressure and 1/B is approximately a straight line when the average pressure
is less than 2000 psi (Ahmed, T., 2005). In addition, Fetkovich (1973) found that the
relationship is almost constant when the average pressure is greater than 3000 psi. Also,
we can indicate from Fig.(6.1) that we cannot make any such simplifying assumptions for
the region located between 2000 to 3000 psi.
134
6.1: Formulations of Relative Permeabilities under Linear Flow

The linear Darcy's flow equations are given by:
qd
Akkrd p1
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -6.3
d x
qk
Akkrk p2
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -6.4
k x
Where, we can define the pseudo pressure as:
m( p )
P2
p1
1
dp - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 6.5
B
and
m(P)
1
p - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -6.6
B
By substituting Eq.(6.6) into Eq.(5.3) and Eq.(6.4) gives:

q d Akkrd Bd
m( p d )
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 6.7
x
q k Akkrk Bk
m(pk )
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 6.8
x
m( p d )
m(pk )
q d q k AK k rd Bd
k rk Bk
- - - - - - - - - - - - - - - - - - - - - - - -6.9
x
x
Let Y(S d ) be equal to

m( p d )
m(pk )
Y(S d ) k rd Bd
k rk Bk
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - 6.10
x
x
fd
qd
qd qk
m( p d )
x
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -6.11
Y(S d )
k rd Bd
135
Similarly for fk
fk
qk
qd qk
m( p k )
x
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 6.12
Y (S d )
k rk
f d f k 1 , - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 6.13
By dividing Eq.(6.11) on Eq.(6.12) and rearranging, we get:
f d k r d Bd m(pd )
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -6.14
fk
k r k Bk m(pk )
By substituting Eq.(6.6) into Eq.(6.14), we get:
1
f d k r d Bd d Bd p d
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 6.15
1 p k
fk
k r k Bk
k Bk
p1 p 2 Because p c 0 ; therefore, Eq.(6.15) yields:

k rd d
k rk
k
fd
fk
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 6.16
u ub B - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -6.17
q qb B - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -6.18
b /B - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 6.19
q
ub b - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 6.20
A
Where (b) indicates the standard base conditions (1 atm and 60 Fo)
By substituting Eqs.(6.17), (6.18) and (6.20) into Eq.(6.16), we get:
u B
bd d d - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 6.21
k rk
ubk Bk k
kr d
136
or
k rd qbd Bd d
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 6.22
k rk
qbk Bk k
Where
qnw Akkrnw Bnw
m( pnw )
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 6.23
x
Where (nw) denotes the non-wetting phase.
p
We concluded from Civan and Donaldson (1991) that nw is equal to :
x
pnw
1
dp
p Qbo
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 6.24
x
L
dQbo
(o) refer to the inlet face
Note
m( pnw )
1 pnw
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 6.25
x
nwBnw x
By substituting Eq.(6.25) into Eq.(6.24), we get:
m( pnw )
1
o dp
p Qb
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 6.26
x
Bnw nw L
dQbo
Substituting Eq.(6.26) into Eq.(6.23), we get:
qnw Akkrnw
o dp
p Qb
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - -6.27
nw L
dQbo
1
Then, substituting Eq.(6.18) into Eq.(6.27) gives:
137
o dp
qbnw Akkrnw
p Qb
- - - - - - - - - - - - - - - - - - - - - - - - - - - -6.28
o
Bnw nw L
dQ
b
By rearranging Eq.(6.28), we get:
k rnw
qbnw Bnw nw L
dp
Ak p Qbo
dQbo
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 6.29
p1 p2 p - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 6.30
From Eq.(6.22) we obtain:
k B q
k rw rnw w w bw - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -6.31
Bnw nw qbnw
Where (w) is noted as the wetting phase.
The relationship between pressure and time for the unsteady-state can fit with the powerlaw equation as shown in Fig.(1.9), Toth et al. (1998, 2001, 2006) also found that:
p a2 t b2 (For Radial displacement), Toth et al. (2005)

V (t )
p a1 i
Vp
b1
(For Linear displacement) Toth et al. (1998), where the Vp is constant.
Thus, we can conclude:
p Qbo - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 6.32
Note that the p should modify to pseudo-pressure for compressible fluid.
Where < 0. Therefore,
Qbo qiot - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -6.33
138
and
dp
dQbo
Q 1 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -6.34
dp
0
Where
dQ o
b
Thus, Eq.(6.29) becomes

k rnw
q bnw Bnw nw L
( 1)
Ak p Qbo Qbo

6.2: Formulation for Relative permeabilities under Radial Flow

The radial Darcy's flow equations are given by:
qd
2rhkkrd p1
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -6.35
d
r
qk
2rhkkrk p2
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -6.36
k
r
By substituting Eq.(6.6) into Eq.(6.35) and Eq.(6.36), we get:
qd 2rhkkrd Bd
m( pd )
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -6.37
r
qk 2rhkkrk Bk
m(pk )
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -6.38
r
qd qk 2rhk
1
k rd Bd m(pd ) k rk Bk m(pk ) - - - - - - - - - - - - - - - - - - - - - - - - - 6.39
r
Let Y(S d ) be equal to
139
Y(S d ) k rd Bd m(pd ) k rk Bk m(pk ) - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 6.40

Thus, Eq.(6.40) becomes
q d q k 2rhk
fd
1
Y(S d ) - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -6.41
r
qd
k B m(pd )
rd d
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -6.42
qd qk
Y(S d )
Similarly for fk
fk
qk
k B m(pk )
rk k
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 6.43
qd qk
Y(S d )
By dividing Eq.(6.42) on Eq.(6.43) and rearranging gives:
f d k rd Bd m(pd )
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 6.44
f k k rk Bk m(pk )
By substituting Eq.(6.6) into Eq.(6.21) and rearranging, we get:
k rd d f d
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 6.45
k rk k f k
Or by using the same procedures for the linear flow Eq.(6.45), it becomes:
krd qbd Bd d
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 6.46
k rk
qbk Bk k
We can conclude from Eq.(6.22) and Eq.(6.46) that the relative permeability ratio does
not depend on reservoir shape and the type of flow.
6.3: Formulation for Fluid Saturations

The formulations below are introduced from Civan and Donaldson (1989) and Civan and
Evans (1991).
140
The material balance equations for phase d and phase k are:
( d ud )
S
d d 0 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 6.47
x
t
( k uk )
( k S k )
0 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -6.48
x
t
Where
S d S k 1 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 6.49
S S b B - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 6.50
Q Qb B - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 6.51
By substituting Eqs.(6.19) and (6.50) into Eqs.(6.47) and (6.48), we get:
ubd
S
bd 0 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -6.52
x
t
ubk
S
bk 0 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -6.53
x
t
By integrating Eqs.(6.52) and (6.53) for x distance at inlet face, we get:
ubd ubd
x
S bd dx - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 6.54
t 0
u bk u bk
x
S bk dx - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 6.55
t 0
u bi
1 dQbi
, i d , k - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 6.56
A dt
ub
1 dQb
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -6.57
A dt
o
ubi ubi
dQbi
dQbo
,i d,k - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -6.58
141
The average saturation face over x distance from the inlet can be presented as:
Sbd
1x
Sbd dx - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -6.59
x0
By differentiating Eq.(6.59), we get:

Sbd Sbd x
Sbd
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -6.60
x
Sbd
S
2 Sbd
2 bd x
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 6.61
x
x
x 2
Civan and Donaldson (1991) expressed Eqs.(6.60) and (6.61) in terms of cumulative
injection of the displacing phase as:
Sbd Sbd Qbo
dSb1
dQbo
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 6.62
Sbd
Qbo dSbd Qbok d 2 Sbd
2
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 6.63
x
L dQbo
L dQbok
By applying Eq.(6.54) over the core length and using Eqs.(6.56) and (6.59), we get:
o
ubd ubd
ubo A
dS bd
dQbo
0 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 6.64
By substituting Eq.(6.58) into Eq.(6.64) with rearranging, we get:
dS bd
dQbo
ubd
/ ubo dQbd / dQbo
AL
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -6.65
The integration of Eq.(6.65) gives:

0 o
1 Qb ubd

Sbd Sbd (t 0)
dQbo Qbd - - - - - - - - - - - - - - - - - - - - - - - - - - - 6.66
AL 0 u o
The differentiation of Eq.(6.65) gives:
142
d 2 S bd
dQbok
1 d
A dQ o
b
uo
bd
uo
b
d 2Q
bd
dQ ok
b
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -6.67
If the core sample is initially saturated by phase k and only phase d is injected, this leads
o
) and Sbd(t 0) so we can simplify Eqs.(6.66) and (6.67) to:
to (ubo ubd
S bd S bdi
Qbo Qbd
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -6.68
V p A - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 6.69
So Eq.(6.68) becomes:
Sbd Sbdi
Qbo Qbd
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 6.70
Vp
Application To Oil - Water System

Let us assume that we inject water to displace oil, so the equations became:
S w S wi
Wbio W p
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -6.71
Vp
k B q
k rw ro w w bw - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -6.72
Bo o qbo
or
k f
k rw ro w w - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 6.73
o f o
143
For constant injection rate
k ro
qbo Bo o L
( 1 )
Ak P Qbo Wbio

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -6.74
We can get and by empirical fitting of the following equation:

P Wbio
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -6.75
Where 0
For constant P, Eq.(6.74) becomes

q B L
k ro bo o o - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 6.76
Akp
Note that P should modify to pseudo-pressure for compressible fluid. Also, the B and
will be at average pressure when the P is constant.
6.4: Evaluation of the new Technique Using a Reservoir Simulator

To evaluate this method, we simulated different examples using a reservoir simulator
(Eclipse
TM
software). Also, we used some examples from the literature. Then, we used
our technique to evaluate the new method. We can summarize our cases by: (1)
Simulating the linear and radial system by using Eclipse
TM
as shown in Figs.(6.2), (6.3),
and Fig.(6.4), (2) Assuming relative permeability curves as input data, (3) Running
Eclipse
TM
software to get the production data as a result, (4) Recalculating the relative
permeability curves by using the new method, and (5) Comparing the calculated values
with the assumed values to check the accuracy as shown in examples below. In addition,
144
we compared the new method with other methods ( JBN (1959), Toth et al. (1998), and
Jones (1978)).
We can make the system unsteady by keeping the P constant and letting the flow
rate change or by keeping the injection rate constant and letting the pressure change. In
addition, we used injection wells because they are easy to control by constant pressure
drop or by constant rate. We used water as the injection fluid to displace oil or gas
towards one production well as shown in Figs.(6.2), (6.3), and Fig.(6.4).
6.5: Application and Verification

The applicability of the above equations is demonstrated by four examples. First,
Example 6.1 is introduced from literature to check the applicability for the new method
when the fluids are incompressible. The other Examples were simulated by the simulator
software to examine the accuracy for the new method when there are incompressible
fluids under constant pressure drop as in Example 6.2, or when there are compressible
fluids under linear or radial displacement as in Examples 6.3 and 6.4.
Example 6.1: This example was introduced by Jones and Roszelle (1978), concerning
with a laboratory core test under constant injection rate. The rock and fluid properties are
presented in Table 6.1. In addition, this example is applied for incompressible fluids (oilwater system). For convenience, we started with incompressible fluids to test the new
method. At the same time, there are many methods available for incompressible fluids to
compare with. Also, we wanted to check the new method for constant injection rate.
We can see from Fig.(6.7) that the power-law equation for the new method
(Eq.6.75) fits well. Also, we found that the results for new method are close to the Toth
145
and Jones 24 methods as shown in Fig.(6.8) and Fig.(6.9).

Example 6.2: The rock and fluid properties for this example are presented in
Table 6.1. This example was simulated by Eclipse
TM
for constant pressure drop,
incompressible fluids and linear flow. Also, we assumed that there is one layer as shown
in Fig.(6.2). The objective of this test is to study the accuracy of the new method under
constant pressure drop by our simulation technique.
We can note from Fig.(6.13) and Fig.(6.15) that the result for the new method
gave good matching with assumed values. Also, the results are close to the Toth et al.
method.
Example 6.3: This example was simulated by Eclipse
TM
for constant pressure
drop, compressible fluids (oil by water) and linear flow. The objective of this test is to
check the applicability of the new method for compressible fluids. The rock and fluid
properties used for this example are presented in Table 6.1.
We can see from Fig.(6.17) and Fig.(6.18) that the new method is very accurate
because of the good matching obtained between the assumed and calculated values of
relative permeability.
Example 6.4: The rock and fluid properties for this example are presented in Table 6.1.
This example was simulated by Eclipse
TM
for constant pressure drop and compressible
fluids (oil by gas). The objective of this test is to check the applicability of the new
method when we have compressible fluids and radial flow. In addition, we can see from
Figs.(6.21) that the new method gave good results because of the good matching obtained
between the assumed and calculated values of the relative permeability ratio.
146
6.6: Conclusions
The following discussions and conclusions are provided based on the present studies:
The new method is very accurate because of the good matching obtained between
the assumed and calculated values of relative permeability.
The result for the relative permeabilities and the relative permeability ratios of the
new method are very close to the results of the Jones (1978) and Toth et al. (1998)
methods for the incompressible fluids.
The new method is a general method for compressible fluids (gas or liquid) and
very simple to apply.
The relative permeability ratio does not depend on the reservoir shape or the type
of flow.
147
Table 6.1 Petrophysical parameters for examples 6.1, 6.2, 6.3

and 6.4
Parameter Example 5.1 Example 5.2 Example 5.3 Example 5.4
L, m
A, m2
Vp, m3
D
qi, m3/sec
K, m2
w
o
Swi
P
Bo at Pi
Bw at Pi
rw
re
h
Bg
g
0.0225
1.82 e -4
0.000031
0.215
1.5 in
2.2 e -8
3.49 e -14
0.001 pa.s
0.01045 pa.s
0.35
--------------------------------------------------------------
15.24
0.304
1.204
0.26
----------------4.93 e -14
0.0005 pa.s
0.001 pa.s
0.30
6900 kpa
------------------------------------------------------
15.24
0.304
1.204
0.26
-------------4.93 e -14
0.0005 pa.s
0.012 pa.s
0.30
6900 kpa
1.5
1.3
---------------------------------------
------------------71103
0.26
------------2.96 e -13
-------0.0132 pa.s
0
10300 kpa
1.23
-------0.044m
76.2m
15 m
0.81
0.0000134
1/B
2000
3000
Pressure,psi
Fig.(6.1) Pressure versus 1/B

Source: This figure introduced from (Ahmed, T., 2005)
148
Injection
well
Production
well
Fig.(6.2) Linear displacement

Fig.(6.3) Two dimensional shapes by
Eclipse
Fig.(6.4) Three dimensional shape by Eclipse
149
28
y = 2.9806x + 0.4523
R = 0.9999
14
0
0
10
Wi/Vp
Fig.(6.5) Displacement equation for Toth et al. example 6.1
8.5
8
7.5
y = 5.584x-0.137
R = 0.9673
Wi/Np
21
6.5
6
5.5
5
4.5
4
0
Wi/Vp
Fig.(6.6) Power low equation for Toth et al. example 6.1
150
8.5
8
7.5
y = 8.6999x-0.127
R = 0.958
6.5
6
5.5
5
4.5
4
0
50
100
150
200
250
300
Wi
Fig.(6.7) Power low equation for the new method example 6.1
1000
Toth et al.
Jones
Corey
100
krw/kro
new method
10
1
0.5
0.55
Sw
0.6
0.65
Fig.(6.8) Relative permeability ratios for example 6.1 by using

different methods
151
1
kro Jones
kro corey
krw Toth
krw,new method
0.9
0.8
krw Jones
krw corey
kro Toth
kro,new method
0.7
kr
0.6
0.5
0.4
0.3
0.2
0.1
0
0.35
0.4
0.45
0.5
Sw
0.55
0.6
0.65
0.7
Fig.(6.9) Relative permeability for example 6.1 by using different

methods
0.2
0.18
q m3/d
0.16
0.14
qo
0.12
qw
0.1
0.08
0.06
0.04
0.02
0
0
10
15
20
25
30
35
time,day
Fig.(6.10) Production data for example 6.2 from Eclipse
152
40
35
30
y = 3.58x + 0.9966
R = 0.9998
25
Wi/Np
20
15
10
5
0
0
Wi/Vp
Fig.(6.11) Displacement equation for Toth et al. example 6.2
4.E+6
3.E+6
y = 0.1222x1.0255
R = 0.9998
3.E+6
Wi
2.E+6
2.E+6
1.E+6
5.E+5
0.E+0
0.E+0
5.E+6
1.E+7
2.E+7
Time, sec
Fig.(6.12) Cumulative injected vs. time for Toth et al. example 6.2
153
1000000
Toth et al.
100000
assumed
10000
New method
krw/kro
1000
Corey
100
10
1
0.1
0.01
0.25
0.3
0.35
0.4
0.45
0.5
0.55
0.6
Sw
different methods with assumed values
1
Krw assumed
kro assumed
krw Toth
kro Toth
kro new
krw new
krw Corey
kro Corey
0.9
0.8
0.7
kr
0.6
0.5
0.4
0.3
0.2
0.1
0
0.3
0.35
0.4
0.45
0.5
0.55
0.6
Sw

methods with assumed values
154
0.07
0.06
0.05
krw
Krwassumed
assumed
kro assumed
krw Toth
kro Toth
kro new Method
krw new method
kr
0.04
0.03
0.02
0.01
0
0.3
0.35
0.4
0.45
0.5
0.55
0.6
Sw
methods with assumed values
0.35
qo
0.3
qw
0.25
q m3/d
0.2
0.15
0.1
0.05
0
0
10
15
Time, day
155
20
1000
assumed
New method
100
Corey
Welge
krw/kro
10
0.1
0.01
0.001
0.3
0.35
0.4
0.45
Sw
different methods with assumed values
0.18
Krw
krwassumed
assumed
0.16
kro assumed
0.14
kro new Method
0.12
krw new method
kr
0.1
0.08
0.06
0.04
0.02
0
0.3
0.35
Sw
0.4
0.45
Fig.(6.18) Relative permeability for example 6.3 by using the new

method with assumed values
156
100
krg/kro
10
0.1
0.25
0.3
0.35
0.4
0.45
0.5
Sg
Fig.(6.19) Assumed relative permeability ratio for example 6.4
5
0.01
qo
qg
q m3/d
4.5
0.009
0.008
3.5
0.007
0.006
2.5
0.005
0.004
1.5
0.003
0.002
0.5
0.001
0
0
200
400
600
800
1000
Time, day
157
0
1200
1.E+7
1.E+6
assumed
1.E+5
welge
1.E+4
new method
krg/kro
1.E+3
1.E+2
1.E+1
1.E+0
1.E-1
1.E-2
1.E-3
0
0.2
0.4
0.6
0.8
Sg
Fig.(6.21) Relative permeability ratios for example 6.4 by using the

CHAPTER
6 method
new method
and Welge
158
REFERENCES
Ahmed T., and P. D. McKinney Advanced Reservoir Engineering Gulf Publishing
Burlington, MA 01803, USA , 2005, 401 pages.
Ahmed, T. Reservoir Engineering Handbook Second Edition, Gulf Pablishing Co.

Boston, USA, 2001, 1185 pages.
Alam W.U. , Relative Permeability Measurement at Simulated Reservoir Conditions,

Dissertation, Norman, Oklahoma (1988) pp. 1-13.
Al-Hussainy, R., Ramey, H.J., and Crawford, P.B. (1966), The Flow of Real Gases
Through Porous Media, Trans. AIME, 237, 624.
Amyx, J.W., M.B. Daniel, and L.W. Robert, Petroleum Reservoir Engineering,
McGraw-Hill Book Company, 1960.
Buckley, S, E. and Levertt, M.C., Mechanism of fluid displacement in sands,

Trans.AIME, 146, pages 107, 1942.
Caudle, B. H., Slobod, R. L., and Brownscombe, E. R. W. , Further developments in the

laboratory determination of Relative permeability , Trans. AIME,192, pages145,1951.
Civan F. and Donaldson E.C., Relative permeability From Unsteady-State
159
Displacements with Capillary pressure Included, SPE Formation Evaluation, pp.189193, June 1989.
Civan F. and Evans R.D, Relative permeability and Capillary pressure From Non-Darcy
Flow of Gas/Brine Systems in Laboratory Cores, SPE 26151, Gas Technology
Symposium, Alberta, Canada, June 1993.
Civan F. and Evans R.D,Non-Darcy Flow Coefficient and Relative permeabilities For
Gas/Brine Systems, SPE 21516, Gas Technology Symposium, Houston, TX. Jan. 1991.
Civan, F., Reservoir Formation Damage-Fundamentals, Modeling, Assessment, and

Mitigation, Gulf Pub. Co. Houston, TX, 2000,742 p.
Collins, R.E. Flow of Fluids through Porous Materials Houston, Texas (1976) 270
pages.
Craig, F.F.Jr,"The reservoir engineering aspects of waterflooding", Monograph Series,

SPE, Richardson,TX 1971,Vol.3. pp. 112-14.
Dake L.P., Fundamental of reservoir engineering, Elsevier, Amsterdam, The Hague,

The Netherlands, 1978, 437 pages.
Eclipse TM manual 2009.
160
Fenwick D. N., and Lenormand, R. "The effect of heterogeneity on unsteady-state

displacements" Institute Franais du Ptrole, 2001.
Fetkovich, M.J. The isochronal testing of oil wells SPE 4529, Fall Meeting of the
Society of Petroleum Engineers of AIME, 30 September-3 October 1973, Las Vegas,
Nevada.
Geffen, T. M., Owens, W. W., Parrish, D. R., and Morse, R. A., Experimental
investigation of factors affecting laboratory relative permeability measurements,
Trans.AIME, 192, pages 99,1951.
Hassler, G. L., U. S. Patent 2, 345, 935, 1944.
Henderson, J. H. and Yuster, S. T., Relative permeability study, World Oil, 3, pages
139, 1948.
http://www.glossary.oilfield.slb.com/search.cfm.
John D.Huppler, Numerical investigation of the effects of core heterogeneities on

waterflood relative permeabilities, SPE 2874, ESSO Production Reserach Co., Houston,
Texas, Dec.1970.
Johnson, E.F., Bossler, D.P., and Naumann, V.O., Calculated of Relative Permeability
161
from Displacement Experiments, Trans.AIME, 216. pages 370,1959.
Jones, S.C., Roszelle, W.O.,Graphical Techniques for Determining Relative

Permeability from Displacement Experiments, JPT, May 1978, pp. 807- 817.
Josendal, V. A., Sandiford, B. B. , Wilson, J. W., Improved multiphase flow studies

emplying radioactive tracers, Trans.AIME,195, pages 65,1952.
Korn, G.A. and Korn, T.M.: Mathematical Handbook for Scientists and Engineers,
Second edition, McGraw-Hill book Co., New York City (1968).
Leas, W. J., Jenks, L. H., and Russell, Charles D., Relative permeability to gas,
Levertt, M.C., Capillary behavior in porous solids, Trans. AIME, 142, pages 152, 1941.
Levertt, M.C., Flow of oil-water mixtures through unconsolidated sands, Trans.AIME,

132, 149, 1939.
Levertt, M.C.and Lewis, W.B., Steady flow of gas liquid mixtures through sands,
Trans.AIME, 142, pages 107, 1941.
Loomis, A. G. and Crowell, D. C., Relative permeability studies for Gas-Oil and water-
162
oil system, U.S. Bureau of Mines Bulletin Barttesivlle, Okla., 1962, pages 599.
Marle, C.M.,Multiphase Flow in Porous Media, Gulf Publishing Co. Houston (1981)
257 pages.
Mehdi H., Leonard k., Herbert H., Relative permeability of petroleum reservoir, Boca
Raton, Florida, 1988, 137 pages.
Morse, R. A., Terwilliger, P. L., and Yuster, S. T., Relative permeability measurements
on small sample, Oil & Gas J., 46, pages 109, 1947.
Osoba, J. S., Richardson, J. G., Kerver, J. K. Hafford ,J.A., and Blair , P. M., Laboratory
relative permeability measurements, Trans.AIME,192, pages 47,1951.
Patrick W. M. "Reservoir Simulation Series", Heriot Watt University, Glossary of terms

chapter, 41 pages, 2001.
Qin and Wojtanowicz, Water problems and control techniques in heavy oils with bottom
aquifers, SPE 125414, Americas E&P Environmental and Safety Conference San
Antonio, Tx, USA, March 2009.
Richardson, J.G, calculation of waterflood recovery from steady-state relative

permeability date, Trans, AIME, 201, 373, 1957.
163
Slatt RM., Stratigraphic Reservoir Characterization for Petroleum Geologists,

Geophysicists Volume (6), USA, 2006 p.93.
Stiles, W.E.,"Use of Permeability Distribution in Waterflood Calculations", Trans., 1949,

Vol.186, pp.9-13.
Toth, J., Bodi, T., Szucs, P., and Civan, F., Determination of relative permeability from
unsteady- state radial fluid displacements Paper SPE94994. SPE Annual Technical
conference and exhibition, Dallas 9-12 October 2005.
Toth, J., Bodi, T., Szucs, P., and Civan, F., Near-well bore water-oil relative
permeability inferred from production with increase water- cut. SPE 102312.SPE
Annual Technical conference and exhibition, San Antonio, 24-27 September 2006.
Toth, J., Bodi, T., Szucs, P., and Civan, F., Practical method for analysis of immiscible
displacement in laboratory core test Transport in porous media 31:347-363, (1998).
Toth, J., Bodi, T., Szucs, P., and Civan, F., Well drainage area diagnostics at
waterflooding, University of Miskolc, Hungary, 2007.
Toth, J., Szucs, P., Bodi, T., and Civan, F., Direct determination of relative permeability
from non-steady state constant pressure and rate displacements, paper SPE 67318, SPE
Mid-continent operations symposium held in Oklahoma City, ok, March 2001.
164
W. Dong and Asghari, Effect of oil viscosity on heavy oil/water relative permeability
curves, SPE 99763, Symposium on Improved Oil Recovery, 22-26 April 2006, Tulsa,
Oklahoma, USA.
Welge, H, J.,A simplified method for computing recovery by gas or water drive,
Wyckoff, R.D. and Botest, H.G, Flow of gas oil-water mixtures through sands,
Physics, 7, 325, 1936.
165
APPENDIX A: NOMENCLATURE
P
- pressure difference, Pa
-empirical constant, dimensionless
-cross section area m2
a1
empirical constant, m3/ sec
a2
empirical constant, Pa
- Formation volume factor
b, b1, b2
- Constant parameters, dimensionless
- Fractional fluid volume, dimensionless
- Layer thickness, m
- Permeability, m2
kg
-effective gas permeability, m2
ko
- effective oil permeability, m2
kr
-Relative permeability, dimensionless
kw
-effective water permeability, m2
- Mobility ratio, dimensionless
Np
-cumulative produce oil volume, m3

- Porosity, fraction
- Pressure, Pa
- Well discharge, m3/sec
Qb
-cumulative injection, m3
re
- Radius of well influence, m
rw
- well radius, m
166
- Skin factor, dimensionless
Sw
- water saturation, dimensionless
Swf
- water saturation at breakthrough time, dimensionless
Swi
- Initial water saturation, dimensionless
- Time, sec
ta
- Breakthrough time, sec
Vp
- pore volume, m3
Wi
- cumulative displacing water volume, m3
Wp
-cumulative produce water volume, m3
Y(Sd)
- supplementary function, sec/Pa.s
- Constant parameters
-Fluid Mobility, m2/Pa.s
- Fluid viscosity, Pa.s
-apparent fluid flowing velocity, m/sec
Subscripts
a
-breakthrough
-standard base condition
-displacing
d0
-initial displacing
df
-displacing front
-front
- Influx or injected
167
-displaced
nw
- non-wetting phase
- Oil
-inlet face
-pore or produced
- Relative
- Water or wetting phase
168

Generating RP PDF

Transféré par

Informations du document

Description originale:

Titre original

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

Generating RP PDF

Transféré par

Droits d'auteur :

Formats disponibles

UNIVERSITY OF OKLAHOMA

DETERMINATION OF RELATIVE PERMEABILITY FROM WELL PRODUCTION BY

DETERMINATION OF RELATIVE PERMEABILITY FROM WELL PRODUCTION BY

A THESIS APPROVED FOR THE

Copyright by AHMED Z. HUSSIEN Al-YASERI 2010

I love you mom!

DURING THE ESTIMATION OF RELATIVE PERMEABILITY

5.1.2 NEGATIVE SKIN FACTOR ................................................................................119

AL. METHOD FOR COMPRESSIBLE AND INCOMPRESSIBLE

well in the center (=60o) .................................................................................................39

Fig.(4.1.16) Grid size for example 4.1.1 from Eclipse .....................................................57

Fig.(4.2.6) production data from Eclipse for example 4.2.2 .............................................72

Fig.(4.3.7) Production data from Eclipse for example 4.3.2 ............................................83

Fig.(4.3.27) Relative permeability curves by JBN method for different viscosity

Fig.(5.1) Skin Zone .........................................................................................................111

Accurate estimation of relative permeability is essential for reliable reservoir history

Similarity, the water relative permeability is:

saturation of wetting phase to increase as seen in Fig.(1.3).

Steady- State Methods:

From this definition of fnw. Thus,

or in a field expression such as :

A S nw dx AL(1 S c ) Qnw - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 1.10

Then, integratin g the left side by parts yields as :

ALSnw(L) A xdS nw AL(1 S c ) Qnw - - - - - - - - - - - - - - - - - - - - - - - - - - - -1.11

Observing that dSw dS nw and using

Qnw a b ln(Q) 1.17

Where Ir is the relative injectivity, defined as:

Toth et al., (2001) developed a technique for calculating relative permeability

1 k rnw w / k rw nw 1 k rnw / f w .....................................................................A.25

Subsequently, Civan and Evans, (1991) developed a method for estimating

Fig.(1.1) Permeability definition

Fig.(1.2) General relative permeability curves

Water wet core

Water wet core

Fig.(1.3) Drainage and Imbibition displacement

Fig.(1.4) Steady state and unsteady state method

Fig.(1.5) Three section core for Penn -State method

Fig.(1.6) General Welges plot

Fig.(1.7) Pressure volume relationship

Fig.(1.8) Fluid density vs. pressure for different fluid types.

Fig.(1.9) Flow regimes

et al. method is applied at and after the breakthrough time.

Where the Y(Sd) is the total mobility as shown:

The following boundary conditions are applied for the integration:

Because Y(Sd) is a function of Sd and r is variable in Eq.(2.12); therefore, it should be

A differentiation of Eq.(2.13) gives the following:

The ratio of Eq.(2.16) and Eq.(2.13) gives:

After introducing a special G function and reformulating Eq.(2.19) gives:

4hk 0 f d Y(S d ) 4hk

The time derivative of Eq.(2.20) yields:

Differentiating Eq.(2.14) twice with respect to time yields:

dqi 4hkp d 2Vi (t )

Thus, using Eqs.(2.26), (2.27), and Eq.(2.25) can be reformulated to give:

Then, using Eqs.(2.24), (2.28), and Eq.(2.21) can be given as:

thus Vi(t) is changing, and (b) qi is constant and P (t) is changing.

Case (b): qi is constant

2.1.2 Surface of The Radial Well as an Inlet Face

differential Eq.(2.12) gives.

Via Vka - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -2.39

2. 3 Distribution of Fluid Saturation along the Core Plug

2.3.1 The Water Saturation Distribution During Water Injection

S w S w0 (S w, max S w0 )(S wf S wo ) - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.51