Professional Documents
Culture Documents
Three-Dimensional Black-Oil
Reservoir Simulation Problem
Aziz s. Odeh, SPE, Mobil Research and Development Corp.
Summary
A comparison of solutions to a three-dimensional
black-oil reservoir simulation problem is presented.
The test of the problem and a brief description of the
seven simulators used in the study are given.
Introduction
Seven companies participated in a reservoir
simulation project to compare the results obtained by
different black-oil simulators. The companies were
chosen to give a good cross section of the solution
methods used in the industry. The participants were
Amoco Production Co., Computer Modelling Group
of Calgary (CMG), Exxon Production Research Co.,
Intercomp Resource Development and Engineering
Inc., Mobil Research and Development Corp., Shell
Development Co., and Scientific Software Corp.
(SSC). The paper presents the text of the problem, a
comparison of results in graphical form, and a brief
description of each model. The descriptions were
supplied by the participants.
A variety of computers was used. Amoco used
IBM 3033, IBM 370/168, and Amdahl V/6. CMG
used Honeywell 6000 DPS, and Exxon used Amdahl
470/V5 and IBM 370/168. Intercomp used Cray-l
and Harris17. Mobil and SSC used CDC Cyber 175,
and Shell used Univac 111012C Level 36. The
number of time steps and the central processor times
varied considerably. Those interested in the actual
values should contact the individual companies.
Except for Shell, all the participants used singlepoint upstream mobility weighting. Shell used two
points upstream. Constraints and data are given in
the text.
Runs To Be Made
Case 1
Let the bubble-point (saturation) pressure be constant with a value equal to the original value.
Case 2
Let the saturation pressure vary with gas
saturation - i.e., this is a variable saturation-pressure
case. The PVT lines at pressures above the calculated
saturation pressures are parallel to the original line.
Results To Be Reported
01492136/81/00019723$00.25
Copyright 1981 Society of Petroleum Engineers of AIME
JANUARY 1981
13
OIL PRODUCTION
WELL
GAS INJECTION
100 MM SCFI D
LAYER 1
.,.
H. FT.
.3
20
K
Ky (LINK) Sw
KX
500 500
So
8325 FT.
. 12 .88
8335 FT
.88
8380 FT .
. 12 .88
8400 FT
50
.3
LAYER 2
30
. 12
50
50
25
LAYER 3
,.
Ci==ti=.t,=~~_~~~~
I : J: 1
1000 FT.
.3
50
200 200
8425 FT.
4
II
10
3x10- 6
0.25
o
o
200
0.792
FVF
(RB/STB)
Viscosity
(cp)
Density
(Ibm/cu ft)
1.0620
1.1500
1.2070
1.2950
1.4350
1.5000
1.5650
1.6950
1.8270
2.3570
1.0400
0.9750
0.9100
0.8300
0.6950
0.6410
0.5940
0.5100
0.4490
0.2030
46.244
43.544
42.287
41 .004
38.995
38.304
37.781
37.046
36.424
34.482
Solution GOR
(scf/stb)
1.0
90.5
180.0
371 .0
636.0
775.0
930.0
1270.0
1618.0
2984.0
Reservoir
Pressure
(psi a)
FVF
(RB/bbl)
Viscosity
(cp)
Density
(Ibm/cu tt)
14.7
264.7
514.7
1014.7
2014.7
2514.7
3014.7
4014.7
5014.7
9014.7
1.0410
1.0403
1.0395
1.0380
1.0350
1.0335
1.0320
1.0290
1.0258
1.0130
0.3100
0.3100
0.3100
0.3100
0.3100
0.3100
0.3100
0.3100
0.3100
0.3100
62.238
62.283
62.328
62.418
62.599
62.690
62.781
62.964
63.160
63.959
Gas/Water Ratio
(scf/bbl)
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Viscosity
(cp)
0.5100
0.7400
Density
(Ibm/cu ft)
37.046
39.768
FVF
(RB/bbl)
Viscosity
(cp)
4014.7
9014.7
1.0290
1.0130
0.3100
0.3100
14
Density
(Ibm/cu tt)
62.964
63.959
Reservoir
Pressure
(ps ia)
FVF
(RB/bbl)
Viscosity
(cp)
Density
(Ibm/cu tt)
14.7
264.7
514.7
1014.7
2014.7
2514.7
3014.7
4014.7
5014.7
9014.7
0.166666
0.012093
0.006274
0.003197
0.001614
0.001294
0.001080
0.000811
0.000649
0.000386
0.008000
0.009600
0.011200
0.014000
0.018900
0.020800
0.022800
0.026800
0.030900
0.047000
0.0647
0.8916
1.7185
3.3727
6.6806
8.3326
9.9837
13.2952
16.6139
27.9483
M(f)
(psia Icp)
O.
0.777916
0.267580
0.875262
0.270709
0.386910
0.516118
0.803963
0.112256
0.251845
E + 07
E + 08
E-+ 08
E + 09
E + 09
E+09
E + 09
E+10
E + 10
20
18
-MOBIL
16
14
IXI
12
....
SHELL
AMOCO
INTERCOMP
EXXON
----- SSC
(/)
(II)
10
0
....
W- 8
!i
a:
...J
6
EXXON AGREES WITH MOBIL BETWEEN 3-7 YEARS
4
10 TIME, YEARS
Results
A comparison of the results is given in Figs. 3
through 18. No comparison of saturation pressures is
given because the values reported by the seven
companies were within 20 psi of each other.
OilGas
0
0.001
0.02
0.05
0.12
0.2
0.25
0.3
0.4
0.45
0.5
0.6
0.7
0.85
1.0
~ ~
0.0
0.0
0.0
0.005
0.025
0.075
0.125
0.190
0.410
0.60
0.72
0.87
0.94
0.98
1.0
1.0
1.0
0.997
0.980
0.700
0.350
0.200
0.090
0.021
0.010
0.001
0.0001
0.000
0.000
0.000
15
20
,.
...,m
16
-MOBIL
,.
- - SHELL
AMOCO
12
INTERCQMP
'".. 10
- - EXXON
<.>
- .-. ~
'"
"r
"
sse
sse
4
CMG AGREES WITH MOBIL THROUGH 7 YEARS
AND WITH INTERCOMP BETWEEN 7-10 YEARS
0
1
10 TIME , YEARS
INTERCOMP,
5500
SHEll,
AMOCO
-
MOBIL
SHELL
AMOCO
INTERCOMP
_
EXXON
_._ . sse
.'"
4500
sse
ui
II:
::>
'"'"
.
W
II:
3500
MOBil,
INTERCOMI;P:::::::::::;::::~;~-~.~-;;.~_~.;;_;:;.Z
30
- M O BIL
-
SHELL, AMOCO
INTERCOMP
-----SSC
IZ
w 20
a:
Q.
~a:
::::l
ti:
(f)
10
(f)
<C
(!)
o+-----~----~----~----~--~~--~----~----~----_+----~
456
TIME, YEARS
10
Fig. 6 - Case 1 - gas saturation vs. time for producing wel l Cell 10, 10,3.
JANUARY 1981
17
7500
-MOBIL
-
SHELL
INTERCOMP
EXXON
_......... CMG
6500
...0;
a:
:>
~ 5500
a:
..
MOBIL-INTERCOMP~-...
4500
MOBIL-INTERCOMP::;~~~::;;:
3900-__~__~__~__~__~__~__~__~__~__~.
10 TIME , YEARS
4500
_.
.-.
..'"
_._. _.
4000
:!
._.
- .-.-
-MOBIL
W
a:
:>
'"'"a:w
-.
SHELL
AMOCO
INTERCOMP
EXXON
._._. SSC
3500
.-.
CMG AGREES WITH AMOCO
3000
1,1
2,2
3 ,3
I~M
4 ,4
5,5
6,6
7,7
GRID POINT LOCATION
8,8
9,9
10,10
18
S= M
. .1.... sf
1
50
S
A'I ' E= M
-MOBIL
SHELL
AMOCO
INTERCOMP
EXXON
_ _ SSC
........... CMG
SSC = E
SSC , E=A
....
z
w
0
.
0:
W
E=I
CMG~ A
45
0:
:1
::>
E= M
t(
<II
<II
sse=
""
CMG ~ 1
'1'
CMG= A
_._._.
40 ~~-+~~~~~~~-r~=-~~~-=~+-~~~~~~~
1,1
2,2
3 ,3
4,4
5,5
6,6
7,7
8 ,8
9,9
10,10
GRID POINT LOCATION
Fig. 9 - Case 1 - gas saturation vs. grid-point location , lime =8 years, top layer.
50
I, E:A
CMG: M
40
-MOBIL
SHELL
AMOCO
INTERCOMP
EXXON
__ SSC
.......... CMG
SSC = A
....
Z
0:
W 30
o
~
0:
'c::>
CMG=
GAS SAT. < .005 AT:
1) 4 ,4- SHELL, EXXON, AMOCO
20
2) 5 ,5 -ALL PARTICIPANTS
3) 6 ,6- ALL PARTICIPANTS EXCEPT MOBIL,
CMG
4 ) 7 .7_SHELL, AMOCO
<II
<II
""
slc,
E= CMG
5} 8,8- SHELL
10
A, E, CMG= M
I, Ef A
CjG= M
E= CjG
r SSC, CMG: M
0 ~~~~'~~"'~
" ""~"~'~1~~
1,1
2,2
3 ,3
4,4
5,5
6,6
7,7
GRID POINT LOCATION
8,8
9 ,9
10,10
Fig. 10 - Case 1 _ gas saturation vs. grid-point location, time=8 years, middle layer.
JANU ARY 1981
19
water. Less common features it can model are vaporization of oil into the gas phase and hysteresis in the
capillary pressure and relative permeability data.
GPSIM has only minor restrictions on the number
of grid blocks that it can use; large problems can be
run using only relatively modest amounts of central
memory. This desirable feature is accomplished by
using disks to store data temporarily by planes for
three-dimensional problems or by rows for twodimensional ones. If the central memory made
available is sufficiently large, the program
automatically will eliminate the temporary data
storage, keeping all data within core.
Intercomp's Black-Oil Simulator
20
EXXON,AMOCO
18
SHELL
= MOBIL
- M O BIL
16
SHELL
AMOCO
_._ . SSC
14
0
"-
a:I
~
(JJ
12
(\')
... 10
0
~
II:
-I
(5
6
INTERCOMP AGREES WITH MOBIL UP T O 7 YEARS,
AND WITH AMOCO BETWEEN 7-10 YEARS
2
0
10 TIME, YEARS
20
20
18
-MOBIL
16
AMOCO
EXXON
12
,
U.
(,)
10
III
'"
ri
0
SHELL
_. _ . SSC
14
III
tIII
"
,
,
,,
4
2
0
1
10 TIME. YEARS
Fig. 12-Case2-GORvs.lime.
6000
-MOBil
SHELL
AMOCO
EXXON
SHELL-
AMOCO
UP TO
4 YEARS
_. - . SSC
........... CMG
~
5000
INTERCOMP AGREES WITH SHELL UP TO
4 YEARS, AND WITH MOBIL AFTER THAT
III
0.
rr
::>
III
III
W
rr
0.
4000
"-'-.
3000 +----+--~~--~--~----~---T--~~--~--~~~
o
1
2
3
4
5
6
7
8
9
10
TIME, YEARS
Fig. 13-Case 2 - pressure vs. time for producing well Cell 10, 10,3.
JANUARY 1981
21
30
-MOBIL
SHELL, AMOCO
INTERCOMP
_ ._ . SSC
IZ
a:
20
Q.
fi
a:
::>
t(
UJ
UJ
10
<
CJ
O +----+----+----+----~--_r----~---r----~--~--~
5
6
TIME, YEARS
10
Fig. 14 - Case 2 - gas saturation vs. ti me for produci ng well Ceil 10, 10, 3.
22
7500
-MOBIL
SHELL
AMOCO
INTERCOMP
EXXON
_ . _ . SSC
.......... CMG
6500
...
'"
:i
MOBIL-SHELL-EXXON'.
\
5500
::>
'"'"a:w
..
4500
3900' ~--+---~--~--~--~--~--~--~--~---:;
10 TIME, YEARS
L
4500
...'"
4000
-MOBIL
-
SHELL
ui
a:
AMOCO
::>
'"'"a:w
INTERCOMP
EXXON
_._" SSC
3500
.......... CMG
1,1
2 ,2
3 ,3
4,4
5,5
6,6
7,7
GRID POINT LOCATION
8,8
9,9
10,10
2l
References
1. Behie, A. and Vinsome, P.K.W.: "Block Iterative Methods for
Fully Implicit Reservoir Simulation," paper SPE 9303
presented at the SPE 55th Annual Technical Conference and
Exhibition, Dallas, Sept. 21-24, 1980.
2. Au, A.D.K., Behie, A., Rubin, B., and Vinsome, P.K.W.:
"Techniques for Fully Implicit Reservoir Simulation," paper
SPE 9302 presented at the SPE 55th Annual Technical Conference and Exhibition, Dallas, Sept. 21-24, 1980.
3. Spillette, A.G., Hillestad, J.G., and Stone, H.L.: "A HighStability Sequential Solution Approach to Reservoir
Simulation," paper SPE 4542 presented at the SPE 48th
Annual Meeting, Las Vegas, Sept. 30-0ct. 3, 1973.
SSC:I
.............
A:~
55
1~
E: S ....
S7 E
1-
-MOBIL
CMG:A
I-
_....... ......
Q.
~.-
Z 50
SHELL
AMOCO
INTERCOMP
EXXON
_ _SSC
IX:
CMG
A:S
fi
IX:
::J
t:c
f/)
<
_.. .,
CMG.L S
_ . .-
f/)
45
CMG:M
I
SSC, CMG: A
. .-
A, CMG : M
1, 1
2,2
3,3
4 ,4
5 ,5
6 ,6
ssc: E
I
'f
CMG~ A
40
SSC : E
7 ,7
.............
8 ,8
............
9 ,9
10,1 0
24
..
1= M
50
CMG= A
'.
-MOBIL
CMG= E
... 40
zw
o
a:
SHELL
AMOCO
INTERCOMP
E,CMG=M
-.~.J
EXXON
_._.- SSC
Q.
........... CMG
30
A=I
a:
::)
ti
rJ)
20
rJ)
ct
3) 6,6-SHELL,
10
ssc
S, I, E, CMG= M
SSC,S =E
rSSC,CMG=M
1,1
2,2
3,3
Nomenclature
h
I
J
k
k rg
k ro
kx
ky
M(p)
=
=
=
=
thickness
number of grid points in the x direction
number of grid points in the y direction
permeability
relative permeability to gas
relative permeability to oil
permeability in the x direction
permeability in the y direction
pseudo gas potential
JANUARY 1981
Sg
So
Sw
ill
<I>
gas saturation
oil saturation
water saturation
time step
porosity
m3
Pas
m3
C
m
kg
kPa
stdm 3
JPT
25
(SPE 9741)
conformal mapping to simulate reservoir performance. This was the first thing that occurred to
me to test the validity of the method.
The results have been excellent for the transient
treatment of fluid flow for a well in an enclosed
circle, either at the center of the circle or offset in the
enclosure. In Table 4-2 of the text, I start with the
conformal rectangle, whereby I can reproduce
Muskat's solution that applies to the offset well in a
circle.
No one has made these comparisons, which should
be made available if we are to place validity on the
known methods of reservoir simulation published in
the literature.
'
01492136/81/00039741$00.25
Copyright 1981 Society of Petroleum Engineers of AIME
552
JPT