0% found this document useful (0 votes)
39 views

Capillary Pressure Modeling

The document discusses different methods for modeling capillary pressure in reservoir simulators, including scaling capillary pressure curves based on water saturation endpoints for each grid block, scaling based on rock properties using a J-function, and scaling curves horizontally based on a given initial water saturation array.
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
39 views

Capillary Pressure Modeling

The document discusses different methods for modeling capillary pressure in reservoir simulators, including scaling capillary pressure curves based on water saturation endpoints for each grid block, scaling based on rock properties using a J-function, and scaling curves horizontally based on a given initial water saturation array.
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 9

Capillary Pressure modeling

Alina Suieshova
End-Point scaling
Input: Result:
Swc array and Sgmax array
Sgmax = 1-Swc Simulator creates new saturation functions
0.2 0.22 0.25 0.8 0.78 0.75 for each grid block, by scaling (shifting)
original curves according to new saturation
0.1 0.3 0.23 0.9 0.7 0.77 end-points
0.15 0.24 0.35 0.85 0.76 0.65
Example: new scaled curves for grid block
with Swc=0.2 and Sgmax=0.8
(Assume other saturation end-points are
consistent for the whole model)

One set of RelPerm curves + one Pc curve


(or J-function)
Water saturation initialization in the simulator
1. Assigns phase saturations in gas, oil, water zones based on saturation
endpoints ( given explicitly/taken as first-last values from the table)
2. Calculates phase saturations at transition zones:
a) Calculates Pc= (pw-po)*g*HAFWL
b) Inverse-lookup Sw from Pc input tables and assign the value to cell
centers
t=0, Pc= (pw-po)*g*HAFWL (pw, po initial)
H= 5010m, HAFWL= 155m, Pc= (975-601)*9,8*155=568106 Pa=5.68 Bar

H= 5130m, HAFWL= 35m, Pc= (976-686)*9,8*35=99470 Pa=0.99 Bar

FWL=5165 m, HAFWL=0, Pc=0


t=0, Pc= (pw-po)*g*HAFWL (pw, po initial)
H= 5010m, HAFWL= 155m, Pc= (975-601)*9,8*155=568106 Pa=5.68 Bar

Pc of the cell> max Pc


Sw=Swc=0.22

H= 5130m, HAFWL= 35m, Pc= (976-686)*9,8*35=99470 Pa=0.99 Bar


Pc of the cell=0.99, look up from the table, Sw=0.34
WaterSaturation CapPressure
FWL=5165 m, HAFWL=0, Pc=0 0.22 2.0776
0.25 1.807
0.26 1.5716
0.29 1.1889
0.33 1.034
0.37 0.7822
0.45 0.5146
0.54 0.3893
0.62 0.2561
0.7 0.1938
1 0.0015
3 ways of Capillary Pressure scaling

1st step: vertical scaling

Pcmax array SWATINIT J-function


scales Pc curve scales input Pc curve to honor scales input J curve according
To new Pcmax value initial water saturation array to rock properties (porosity and permeability)
Also assigns new Pcmax
J-function scaling
1st step: vertical scaling of Pc based on Por and K
WaterSaturation WaterRelPerm OilRelPerm Pc
por=0.05 K=5 mD 0.1 0 0.9 2.48
0.2 0.0007 0.7897 2.16
input: 0.3 0.0018 0.7209 1.88
0.4 0.0057 0.5976 1.42
WaterSaturation WaterRelPerm OilRelPerm Jfunction 0.5 0.0125 0.4742 1.24
0.1 0 0.9 2.72 0.6 0.0242 0.3472 0.93
0.2 0.0007 0.7897 2.37 0.7 0.057 0.1671 0.62
0.3 0.0018 0.7209 2.06 0.75 0.1036 0.0589 0.47
0.4 0.0057 0.5976 1.56 0.85 0.1629 0.0101 0.31
0.5 0.0125 0.4742 1.35 0.95 0.2352 0 0.23
0.6 0.0242 0.3472 1.02 1 1 0 0.00
0.7 0.057 0.1671 0.67
0.75 0.1036 0.0589 0.51
0.85 0.1629 0.0101 0.34
0.95 0.2352 0 0.25
1 1 0 0.00 WaterSaturation WaterRelPerm OilRelPerm Pc
0.1 0 0.9 2.08
0.2 0.0007 0.7897 1.81
0.3 0.0018 0.7209 1.57
0.4 0.0057 0.5976 1.19
0.5 0.0125 0.4742 1.03
0.6 0.0242 0.3472 0.78
0.7 0.057 0.1671 0.51
por=0.1 K=10 mD 0.75 0.1036 0.0589 0.39
0.85 0.1629 0.0101 0.26
Rock with better rock properties has lower Pcmax 0.95 0.2352 0 0.19
1 1 0 0.00
2st step: horizontal scaling of Pc based on Swc array WaterSaturation WaterRelPerm OilRelPerm Pc
Swc=0.2 0.2 0 0.9 2.48
0.25 0.0007 0.7897 2.16
0.36 0.0018 0.7209 1.88
0.43 0.0057 0.5976 1.42
0.54 0.0125 0.4742 1.24
0.66 0.0242 0.3472 0.93
0.73 0.057 0.1671 0.62
WaterSaturation WaterRelPerm OilRelPerm Pc 0.78 0.1036 0.0589 0.47
0.1 0 0.9 2.48 0.87 0.1629 0.0101 0.31
0.2 0.0007 0.7897 2.16 0.95 0.2352 0 0.23
0.3 0.0018 0.7209 1.88 1 1 0 0.00
0.4 0.0057 0.5976 1.42
0.5 0.0125 0.4742 1.24
0.6 0.0242 0.3472 0.93
0.7 0.057 0.1671 0.62
0.75 0.1036 0.0589 0.47 WaterSaturation WaterRelPerm OilRelPerm Pc
0.85 0.1629 0.0101 0.31 0.3 0 0.9 2.48
0.95 0.2352 0 0.23 0.32 0.0007 0.7897 2.16
1 1 0 0.00 0.36 0.0018 0.7209 1.88
0.44 0.0057 0.5976 1.42
Swc=0.3 0.57 0.0125 0.4742 1.24
0.69 0.0242 0.3472 0.93
0.75 0.057 0.1671 0.62
0.78 0.1036 0.0589 0.47
0.89 0.1629 0.0101 0.31
0.95 0.2352 0 0.23
1 1 0 0.00
Comparison of different ways of Pc modelling
Pros Cons
1.The Pc curve shape is the same for all rock
types.
Pc max array Easy implementation. 2.Values of Pcmax has to be provided for
each grid block.
1. Initial water saturation distribution,
including distribution in the capillary
fringe, has to be provided.
SWATINIT The best way to correct 2. Double check of unphysically high
and adapt Fluid in Place Pcmax has to be done.
volumes.

Physically, the most If the range of connate water saturation is


correct approach. Each big, the simulator does significant horizontal
J-function rock type has its own scaling, resulting in losing data from J-
capillary pressure shape. function

You might also like