Reservoir Simulation Texas A&M
Reservoir Simulation Texas A&M
RESERVOIR SIMULATION
APPLICATION
Class Notes
Spring 2000
R. A. Wattenbarger
Texas A&M University
www.petroman.ir
Contents
Chapter 1
Chapter 1A
Equilibration
Geologic Equilibration
Initialization of Gridblocks
Effect of Heterogeneity
Chapter 2
Chapter 3
History Matching
Technique
Purpose
www.petroman.ir
Quality of Data
Production Allocation
Wrong Rate Problem
Pressure Analysis
Determining OOIP from Depletion History
Continuity Problem
Parameters to be Changed in History Matching
History Matching Procedures
Chapter 4
Coning
Introduction
Analysis of Forces
Critical Coning Rate
Coning Simulation
Addingtons Method
References
Exercises
Appendix
Chapter 5
Compositional Simulation
Introduction
Applications
PVT Behavior
Equations for Compositional Simulation
Regression of EOS Parameters
Modeling of Surface Facilities
Lab Tests
Gas Cycling in Volatile Oil and Gas Condensate Reservoir
Vaporizing Gas Drive (VGD)
Field Examples
Fundamentals
Heavy Fraction Characterization
Pseudoization
Regrouping Scheme of Heavy Components
Compositional Simulation
Heavy Ends Characterization
Default Fluid Predictions
Pre-Regression
Component Pesudoization
Regression
Initialization Data for VIP
Run Data for VIP
Data Files
www.petroman.ir
CHAPTER 1
Introduction
Driving Forces
Vertical Equilibrium
www.petroman.ir
1
Pseudo Relative Permeability Curves
www.petroman.ir
WOR
3D-lab kr
2D-pseudo k r
% PV INJECTED
In this chapter, we present three kinds of static analyses of pseudo relative permeabilities,
each analysis based on the particular driving force or forces dominating in the reservoir. We
begin by discussing the three driving forces which control fluid flow in the reservoir. Then, the
pseudo relative permeabilities for the segregated case for flows dominated by the gravity force is
presented, followed by Hearns method for flows dominated by the viscous force,and finally
presented is the vertical equilibrium method, where capillary and gravity forces are assumed to
be in equilibrium.
2
Pseudo Relative Permeability Curves
www.petroman.ir
Driving Forces
The movement of the fluids in a typical reservoir is controlled by three kinds of forces.
Table 1.1 shows those three forces. The gravity forces tend to segregate the fluid vertically and is
proportional to the difference in densities. Viscous forces, induced by production and injection,
tend to move the fluids in the direction of the induced movement and the forces are proportional
to the Darcy velocity and inversely proportional to mobility. The capillary forces tend to spread
out the saturations according to the slope of the capillary pressure curve and the saturation
gradients. In the segregated flow case, the saturation gradients of interest are vertical.
Table 1
Forces in fluid movement
Forces
Magnitude
144
(a) gravity
(b) viscous
0.00633k
dPC
dS
(c) capillary
S
x
The ratio of gravity forces to viscous forces is known as Gravity Number. This number is
calculated by the following equation for an oil-water system:
N grav =
( w o ) / 144
( q w / A)
(1)
~
0.00633k k rw @Sor
where
k = ( ki hi ) / hi
3
Pseudo Relative Permeability Curves
www.petroman.ir
For an oil/water reservoir under assumptions, the free water segregates to the bottom by
gravity. Above the water/oil contact, we assume that only connate water saturation exists. while
below the water/oil contact, only residual oil saturation exists. Water flows near the bottom and
oil flows near the top.
Given a particular location of the water/oil contact, we can calculate the average
saturation and the average flow coefficient (pseudo relative permeability). Fig. 3 shows the
physical process of segregation for homogeneous reservoirs.
Swc
h
1-Sor
hw
The equations that express the average water saturation, and the oil and water pseudo
relative permeability, are all linear with the water height. Consequently, the pseudo relative
permeabilities are linear with saturation as can be seen in these following equations :
h hw
h
~
S wc
S w = w (1 S or ) +
h
h
(2)
h
~
k rw = w k rw@(1 sor )
h
(3)
h hw
~
k ro =
k ro @(1 swc )
h
(4)
4
If we investigate the "end points" of the flow conditions, we note that when the water/oil
contact is at the bottom,there is no water flow. Water saturation is at the connate water saturation
value and the oil pseudo relative permeability is the value at connate water saturation. When the
water/oil contact is at the top, no oil flows, water flows at the relative permeability at residual oil
saturation and the water saturation, has a value with residual oil saturation present.
Calculation Procedure. The calculation procedures are summarized in the following steps:
(1) Characterize thickness, connate water saturation, and residual oil saturation.
(2) Find end-point water relative permeability and end-point oil relative permeability.
~
~ ~
(3) For a value of h w calculate S w , k rw and k ro
(4) Repeat step 3 for different values of h w .
Application. The .equation that express pseudo relative permeabilities and average water
saturations for the segregated case are linear. If we look at the shape of the pseudo relative
permeability curves, we notice that these are straight lines connecting the end points of the
laboratory relative permeability curves as shown in Fig. 4. If the segregated assumptions supply,
then we can input these straight line as relative permeabilities in a 2-D model and obtain about
the same results as the lab curves produce in a 3-D model.
kr
kr
kr lab
Sw
5
Pseudo Relative Permeability Curves
www.petroman.ir
k1 h1 1
Swc1 Sor1
k2 h2 2
Swc2 Sor2
k3 h3 3 Swc3 Sor3
kn hn n Swcn Sorn
6
Pseudo Relative Permeability Curves
www.petroman.ir
Now we consider a layered case in which water injection has begun and established
different fronts according to the permeability of the layers. The order of breakthrough is
determined by the properties of each layer. After each layer has broken through, we can calculate
the average relative permeabilities and average saturations. These calculations will give us points
on a pseudo relative permeability curve which we can then smooth to obtain a continuous curve.
Calculation Procedures. The calculation procedures are summarized in the following steps.
(1) The layers are characterized by thickness, porosity, connate water saturation, and
residual oil saturation.
(2) The layers are arranged in order of decreasing breakthrough of water-oil displacement
front according to the values of the factor k i / i S i ,where S i = (1 S wci S ori )
(3) Calculate the average water saturation at the outflow end at the system before
breakthrough of the first layer, and after breakthrough of each layer
~
~
(4) For each,values of k rw and k ro , the pseudo relative permeabilities to water and to oil,
respectively, are calculated.The resulting equations are summarized as follows:
For n = 0 (before breakthrough):
N
(hSwc )i
~
i
Sw = =1
N
(h )i
i =1
(5)
~
k rw = 0
(6)
~
k ro =
(khk
i =1
ro @ S wc i
(7)
(kh)
i =1
(8)
7
Pseudo Relative Permeability Curves
www.petroman.ir
n
( kh )i k rw @(1Sor )
~
k rw = i =1
N
( kh )i
i =1
(9)
N
( khk ro @Swc )i
~
i
=
k ro = n +1
N
( kh )i
i =1
(10)
N
(h(1 Sor ))i
~
i
SW = =1
N
(h )i
i =1
(11)
~
k rw = k rw@(1 Sor )
(12)
~
k ro = 0
(13)
For n = N
where,
N= total layers
n= layers with breakthrough
Application. Fig. 6 shows the comparison of the pseudo and lab relative permeability curves.
The pseudo relative permeabilities to water are generally very different from lab relative
permeability.
kro
kr
kw
kro
kw
Sw
8
Pseudo Relative Permeability Curves
www.petroman.ir
WOR
High rate
Low rate
% PV INJECTED
9
Pseudo Relative Permeability Curves
www.petroman.ir
Vertical Equilibrium
The vertical equilibrium case (VE) is similar to the segregated case except that it is
modified to include the effects of capillary pressure. The VE concept is the differences between
the pressures in the oil and water(or oil and gas) phases is exactly balanced by capillary pressure.
This is the same as having potentials of each phase constant vertically.6 The factors which favor
VE are:
(1) low resistance to flow normal to the bedding planes,
(2) sands thin in the direction normal to the bedding planes,
(3) low areal rates of fluid movement, and
(4) if there is capillary pressure in transition zone, the capillary and gravity forces are
balanced.
Fig. 8 illustrates equilibrium between capillary forces and gravity forces. The Pc curve
translated to elevation above the free water level (Pc = 0) by Pc = (Z - ZFWL)( w o )/144.
o/144
Pc
w/144
Sw
Sw
~
Sw =
(Z )S
( Z )dZ
(14)
( Z )dZ
0
10
Pseudo Relative Permeability Curves
www.petroman.ir
~
k rw =
k (Z)k
rw
( Z )dZ
(15)
k ( Z )dZ
0
The shape of the pseudo relative permeability and capillary pressure curves are affected by:
(1) density difference,
(2) dip angle,
(3) sand thickness h, and
(4) stratification (k(Z)and p(Z)).
So the resulting pseudo relative permeability curves are not straight lines unless the
assumptions of the homogeneous segregated cases are met. Otherwise, the VE curves will be
somewhere between the laboratory curves and the segregated case curves.
Application. Results summarized in Fig. 9 demonstrate that simulations using correct 1-D, VE
pseudo curves agree closely with the correct behavior.6
Gas Saturation, Sg
11
Pseudo Relative Permeability Curves
www.petroman.ir
~
Sw =
(hS
)j
j =i
JMAX
(h)
(16)
j
j =i
13
Pseudo Relative Permeability Curves
www.petroman.ir
JMAX
~
k rw =
( khk
)j
rw
j =1
JMAX
( kh)
(17)
j
j =1
(8) Calculate dynamic pseudo relative permeability for oil in the oil zone using:
JMAX
~
k ro =
( khk
ro
)j
j =1
(18)
JMAX
( kh)
j =1
(9) The pseudo relative permeabilities to oil and water in the oil zone are then averaged
with values for zones in the remainder of the block thickness. These values are the
dynamic pseudo relative permeabilities.
Fig. 10 illustrates a cross-sectional model which is used to develop dynamic pseudo
relative permeabilities for each column, and Fig. 11 shows the results of the calculation. Higher
rates move the dynamic pseudo relative permeability curves closer to the rock curve.
14
Pseudo Relative Permeability Curves
www.petroman.ir
15
Pseudo Relative Permeability Curves
www.petroman.ir
(a) Darcy's equation is used to calculate dynamic pseudo relative permeabilities of oil and
water.
(b) Pseudo capillary pressure, which is included, accounts for any phenomenon other than
the gravity component in the direction of flow. This gravity component is included in
the areal model flow calculations.
(c) Upstream relative permeabilities will be used in the areal model to calculate flow
between blocks.
(d) Areal model porosity is defined as the ratio of the total void volume of all the crosssectional blocks represented by the given areal block to the total bulk volume of these
same cross-sectional blocks.
(e) Areal block permeability is a type of harmonic average.
(f) Densities and viscosities are evaluated at the arithmetic average pressure.
Calculation Procedure.
Fig. 12 below compares a cross-sectional model with the
corresponding one-dimensional areal model to help define symbols used in the equations.
16
Pseudo Relative Permeability Curves
www.petroman.ir
CROSS SECTION
1-D AREAL
Before developing equations for calculating pseudo functions, we need to define the
relationship between porosities and permeabilities for the cross-sectional and areal models.
Areal model block porosity is defined as shown below :
i =10
II =
(h X )
i =6
(19)
1
(h I + h II ) X II
2
x II
x
hII i
i = 8 hi k i
(20)
i =12
17
Pseudo Relative Permeability Curves
www.petroman.ir
S pwII =
(h xS
w i
i =6
(21)
1
[h I + h II ]x II II
2
(2) Calculate pseudo flow rates for water and oil across the boundry between areal
blocks II and III using Equations 22 and 23 below:
j=9
q pwII = q w10, j
(22)
j =1
j=9
q poII = q o10, j
(23)
j =1
(3) Calculate dynamic pseudo pressures for water and oil using Equations 24 and 25
below:
j =9
p pwII =
[ p
w8 , j
w8, j y 8, j
144
j =1
]k 8, j k rw8, j h8, j
(24)
j =9
8, j
k rw8, j h8, j
j =1
j =9
p poII =
[ p
o8, j
o8, j y8, j
j =1
144
]k 8, j k ro8, j h8, j
j =9
8, j
(25)
k ro8, j h8, j
j =1
(26)
Application. Two sets of example calculations are presented. Case 1 uses cross-sectional
results from an actual reservoir study and illustrates how the dynamic pseudo relative
permeabilities account for differences between cross-sectional and one-dimensional areal block
lengths. Case 2 is a hypothetical example to illustrate the applicability of both the dynamic
18
Pseudo Relative Permeability Curves
www.petroman.ir
pseudo relative permeability and pseudo capillary pressure functions in duplicating crosssectional results.
Fig. 13 shows the cross-sectional and one dimensional areal models for Case 1. The
lengths of the blocks in the areal model are three times the lengths of the blocks in the crosssectional model, and each areal block represents 60 cross-sectional blocks. The resulting
saturation profiles at breakthrough of this case are shown in Fig. 14.
Fig. 15 shows the cross-sectional and one-dimensional areal models for Case 2. There is
limited communication between the upper and lower zones, a high-permeability zone in the
lower half of the model, and an oil-water viscosity ratio of about 10:1. Fig. 16 compares the
reliability of the dynamic pseudos using Kyte and Berry's method with the previous method.
RECOVERY AT
BREAKTHROUGH, % OOIP
o
CROSS SECTION
WATER SATURATION, PV
10. 99
10. 99
7. 30
12. 25
4
3
2
1
0
10
1 - D AREAL BLOCK
19
Pseudo Relative Permeability Curves
www.petroman.ir
20
Pseudo Relative Permeability Curves
www.petroman.ir
WATER SATURATION PV
RECOVERY AT
BREAKTHROUGH
(%OOIP)
0
Cross Section
___ Modified Pseudos W/ 9
Areal Blocks
- - - Conventional Pseudos
w/ 9 Areal Blocks
.. Conventional Pseudos
w/ 27 Areal Blocks
45.2
45.2
34.2
45.6
1 D AREAL BLOCK No
Using this method, the number of computing blocks in the two-dimensional areal model is
reduced by a factor equal to the square of the ratio of the block lengths for the cross-sectional
and areal models.
21
Pseudo Relative Permeability Curves
www.petroman.ir
NOMENCLATURE
dPc
dS
dZ
Ngrav
h
JMAX
k
k(Z)
ko
kr
krw(Z)
kw
kr
kro @1-Swc
krw @1-Sor
p
Pc
S
Sor
Swc
Sw
Swc
u
Z
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
layer
density, lbm/ft3
density difference, lbm/ft3
porosity, fraction
porosity function of depth, fraction
viscosity, cp
=
=
=
=
oil
water
layers with breakthrough
total layers
Symbols
(z)
Subscripts
o
w
n
N
22
Pseudo Relative Permeability Curves
www.petroman.ir
REFERENCES:
1.
Coats, K.H., Neilson, R.L., Terhune, M.H., and Weber, A.G.: "Simulation of ThreeDimensional, Two-Phase Flow in Oil and Gas Reservoirs," SPEJ (Dec. 1967) 377-388.
2.
Martin, J.C.: "Partial Integration of Equations of Multiphase Flow," SPEJ (Dec. 1968)
370-380; Trans., AIME, 243.
3.
Coats, K.H., Dempsey, J.R., and Henderson, J.H.: "The Use of Vertical Equilibrium in TwoDimensional Simulation of Three-Dimensional Reservoir Performance," SPEJ (March
1971) 63-71; Trans., AIME, 251.
4.
5.
Jacks, H.H., Smith, O.J.E., and Mattax, C.C.: "The Modeling of a Three-Dimensional
Reservoir with a Two-Dimensional Reservoir Simulator-The Use of Dynamic Pseudo
Functions," SPEJ (June 1973) 175-85.
6.
Mattax, C.C. and Dalton, R.L.: Reservoir Simulation, Monograph Series, SPE, Richardson,
TX (1990) 20, 13.
7.
Kyte, J.R. and Berry, D.W. : "New Pseudo Functions To Control Numerical Dispersion,"
SPEJ (Aug. 1975) 269-76.
23
Pseudo Relative Permeability Curves
www.petroman.ir
Exercises
Class Problem No. 20 (Pseudo Relative Permeability)
Attached is a plot of a laboratory oil-water relative permeability curve. This relationship applies
to the rock where oil and water both flow. We also have measured that krw = 1.0 at Sw = 1.0.
Our 2-D model includes some parts of a reservoir with an initial water-oil contact. Below this
initial water-oil contact Sw = 1.0.
Construct pseudo relative permeability curves for cells which have 0%, 25%, 50%, and 75% of
their thickness below the initial water-oil contact.
1.0
0.9
0.8
0.7
Kr
0.6
0.5
0.4
0.3
0.2
0.1
0.0
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
Sw
24
Pseudo Relative Permeability Curves
www.petroman.ir
0.9
1.0
k(md)
h(ft)
50
0.2
200
0.2
10
0.2
500
0.2
100
0.2
Assume:
(1) Vertical fluid saturation distribution is controlled by viscous flow forces resulting
from vertical permeability variation (gravity and capillary forces are negligible relative to
viscous effect)
(2) Swc = 0.25
Sor = 0.35
kro = 0.9
krw = 0.3
(a) Calculate the pseudo relative permeability curves using Hearn's approach.
(b) Plot the resulting curves on the attached graph paper.
(c) Plot another pseudo relative permeability curve using the data with an assumption of
gravity segregation. Remember that the system is not homogeneous.
(d)
1.0
0.9
0.8
0.7
Kr
0.6
0.5
0.4
0.3
0.2
0.1
0.0
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
Sw
25
Pseudo Relative Permeability Curves
www.petroman.ir
1.0
40
20
(psia)
30
10
-10
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Sw
k and k
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
Sw
26
Pseudo Relative Permeability Curves
www.petroman.ir
1.0
We are modelling a waterflood in which we observe a sharp oil bank. The formation is
homogeneous and there is no gas saturation.
Our model runs are exhibiting numerical dispersion, which is showing early breakthrough
and giving the wrong shape to WOR curves.
Modify the attached laboratory relative permeability curve to give a Buckley-Leverett
front as water leaves a cell. Construct a fractional flow curve and calculate the proper relative
permeabilities at breakthrough. Sketch the rest of the relative permeability curves.
Water viscosity = 0.6 cp
Oil viscosity = 2.5 cp
1.0
0.9
0.8
0.7
Kr
0.6
0.5
0.4
0.3
0.2
0.1
0.0
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
Sw
27
Pseudo Relative Permeability Curves
www.petroman.ir
0.9
1.0
RESERVOIR SIMULATION
CLASS PROBLEM No. 28 (SOLUTION)
(a) First we change Pc vs Sw curve to h vs Sw curve. Given a capillary pressure Pc, the relevent
water column height 'h' could be calculated by the following formula:
( w - o )h
144
or in the form:
144 Pc
h =
( w - o )
where:
r = lb/cu ft
Pc = psi
h = ft
pc =
1
0.9
40
0.7
0.6
0.5
20
0.4
(psia)
30
10
0.3
0.2
0.1
-10
0
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
Sw
(b) Three 100 ft intervals are chosen from different capillary pressures as:
Case 1: Pc= 0.0 - 10.0 psi.
Case 2: Pc=10.0 - 20.0 psi.
Case 3: Pc=20.0 - 30.0 psi.
28
Pseudo Relative Permeability Curves
www.petroman.ir
0.8
(c) Saturation vs height profiles for the above three cases are shown in Fig. 2:
100
80
h (ft)
60
40
20
0
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Sw (fraction)
(d) krw vs height profiles for the three cases are shown in Fig. 3:
0.4
k (fraction)
0.3
0.2
0.1
0
0
10
20
30
40
50
60
70
80
h (ft)
(e) The pseudo krw and Sw for the three cases are calculated by the formulas:
h
(Z) S
Sw =
dZ
0
h
(Z)dZ
0
29
Pseudo Relative Permeability Curves
www.petroman.ir
90
100
k(Z) k
k rw =
rw
(Z)dZ
0
h
k(Z)dZ
0
In this problem, h = 100 ft and we assume k(Z) = const., p(Z) = const. so the formulas are
simplified:
h
S w (Z)dZ
Sw =
100
h
k rw (Z)dZ
k rw =
100
The trapezoidal rule is used for the above integration and the results are shown in the following
table:
Case 1
Case 2
Case 3
Sw
0.6121
0.4084
0.2884
krw
0.2612
0.06538
0.01348
The pseudo-curve from the above results is plotted and shown in Fig. 4:
1.0
k (fraction)
0.8
0.6
0.4
0.2
0.0
0.0
0.1
0.2
0.3
0.4
0.5
0.6
Sw (fraction)
30
Pseudo Relative Permeability Curves
www.petroman.ir
0.7
0.8
0.9
1.0
CHAPTER 1A
Equilibration
Geologic Equilibration
Initialization of Gridblocks
Effect of Heterogeneity
www.petroman.ir
Equilibration
Geologic Equilibration
Equilibration is the initialization of fluid saturations for all gridblocks in a numerical
reservoir simulator such that the fluid potential gradients are equal to zero.
Fig. 1 shows a schematic of a reservoir along with the corresponding capillary pressure
curves, depth and water saturation relationships.
cow =
Oil
Z FWL FWL
po- pw
144
144
Z
FWL
Pcow
0
FWL
cow = 0
Water
0 S wc
p
(a)
(b)
Sw
(c)
100 %
0 S wc
100 %
(d)
Fig. 1 - (a) Schematic of an oil-water reservoir system. (b) Oil and water phase pressures versus
depth. (c) Oil-water capillary pressure as a function of water saturation. (d) Water saturation versus
depth.
Consider an oil-water reservoir system, Fig. 1.a, with a maximum water saturation at ZFWL
at which Pcow = 0. If the imbibition capillary pressure curve has a distinct capillary transition zone,
as shown in Fig. 1.c, then above ZFWL, the water saturation will be distributed in accordance with the
capillary pressure-saturation relationship. In particular, the capillary pressure at a distance h above
the free water level (FWL) can be calculated as,
Pow(Sw)=Po-Pw
= 144 ( Z Z FWL )
(1)
or,
Equilibration
www.petroman.ir
144
Z = ZWFL
PCO W ( SW )
(2)
where,
Po = PFWL
o
144
PW = PFWL
( Z Z FWL )
W
144
(3)
( Z Z FWL )
(4)
= 0
Equilibration
www.petroman.ir
Initialization of Gridblocks
The most commonly used initialization method "initializes on the oil phase." Oil pressure is
calculated for each gridblock. The saturations and other phase pressures are derived from capillary
pressures and phase pressure gradients. The following steps are involved.
1.
Select an oil-phase reference pressure, p1, a reference water/oil capillary pressure, PC2,
a reference gas/oil capillary pressure PC3, and the depths at which these reference
pressures apply (D1, D2 and D3).
2.
3.
Calculate water pressures at the centers of all blocks with the oil pressure at D2,
capillary pressure PC3, and the water pressure gradient.
4.
Calculate gas pressures with the oil pressure at D3, capillary pressure PC3, and the gas
pressure gradient.
5.
Having established block-centered pressures for oil, gas, and water, read the oil, gas, and
water saturations at block centers from the appropriate capillary-pressure curves.
Initialization methods become more complicated for reservoirs with multiple WOC's or
GOC's and rock/fluid properties that vary with depth. The J function can be applied for initialization
of saturations for heterogeneous reservoirs. Application of the J function is discussed in the
following section.
Equilibration
www.petroman.ir
Effect of Heterogeneity
J Function
The J function is a term for correlating capillary-pressure data of rock formations of similar
geological characteristics. The correlating procedure is a dimensionless grouping of the physical
properties of the rock and saturating fluids. The J function can be expressed as,
J ( SW ) =
PCOW
cos
(5)
where,
Sw = Water saturation, fraction of pore volume
Pcow = Capillary pressure, dyne/cm2
cow
250
500
1000
100 %
Equilibration
www.petroman.ir
k
= 1000
100 %
It is often difficult to obtain all capillary pressure-water saturation data from direct laboratory
measurements for all k/f values of a heterogeneous reservoir due to data availability and costs. The
following example illustrates the use of the J function to obtain a vertical water saturation profile for
a heterogeneous reservoir with laboratory measurements available for only one layer.
Fig. 4 shows a three layered reservoir with k/f values of 1000, 250 and 1000. Laboratory
capillary pressure-water saturation measurements are only available for the top layer. To obtain a
vertical water saturation profile for the reservoir, we first convert the top layer's capillary pressure
data to the J function for various water saturations to obtain a J function-water saturation curve as
shown in Fig. 3.
= 1000
= 250
= 1000
FWL
cow
FWL
0
0
(a)
Sw
100 %
(b)
Sw
100 %
(c)
Fig. 4 - (a) Three layered reservoir. (b) Profile of capillary pressure (Pcow) versus water saturation
(Sw). (c) Profile of elevation (Z) versus water saturation (Sw).
We can calculate J function values for the lower layers since we know the k/f values and Pcow
from the capillary pressure equation. With this information we can find the water saturation values
from the J function-water saturation correlation curve obtained from the top layer. The results can
be plotted as shown in Fig. 4.b. Depth versus water saturation can also be plotted as shown in Fig.
4.c since the capillary pressure and depth are linearly related. If all three layers of the reservoirs are
fully perforated, we can see from the vertical saturation profile, Fig. 4.c, that the top layer will
produce oil, the middle layer will produce water and the bottom layer will produce oil and water.
Equilibration
www.petroman.ir
References:
1. Mattax,C.C.,and Dalton, R.L.:"Reservoir Simulation," Monograph Vol. 13 SPE. Chap.4, p. 41.
Equilibration
www.petroman.ir
CHAPTER 2
Introduction
Laboratory Behavior
Reservoir Behavior
Diffusion Effect
www.petroman.ir
Stock Tank
Gas
Phases
Vapor
Liquid
Stock Tank
Oil
Rs
P
Fig. 2(a) - Typical Rs curves.
Rs
P
Fig. 2(b) - Possible Rs curves.
2
Repressuring with Variable Bubble Point
www.petroman.ir
Rs
P
Fig. 2(c) - Rs curves used in simulation.
In a reservoir, gravity segregation causes the gas move to the top and oil to settle down to the
bottom of the reservoir. At the gas-oil contact (GOC), there is a sharp change in the pressure
gradient depending on the densities of the fluids. Throughout the gas cap, the pressure is the same
(because of the low density). In the oil zone, an increase in pressure is expected with depth. This is
because the heavier ends settle to the bottom. At the GOC, the fluids are in equilibrium and the oil is
saturated. This is the true bubble point of the reservoir, and is less than the average value of bubble
point determined using a tank model. Fig. 3 shows the possible variation in pressure in the oil zone.
Contrary to popular belief, it has been observed for many field cases that the oil zone pressure could
see a decrease with depth. This was reported by Standing1 from his study of oil expansion due to
solution gas.
g
144
GOC
Z
b.p.
144
p
Fig. 3 - Bubble point curves.
3
Repressuring with Variable Bubble Point
www.petroman.ir
For a constant API and temperature oil reservoir, we know that the structurally higher wells
will produce with a higher GOR, as illustrated in Fig.4.
Low
GOR
High
GOR
Gas
Oil
In the above diagram, the well near the GOC produces with a high GOR and the well
completed at a much lower depth produces with a low GOR. Using these values of produced GOR,
we can find the corresponding bubble point pressures by plotting GOR produced vs. pressure.
Naturally it follows that the bubble point pressure at the GOC depth lies somewhere above the two
values. This is illustrated in Fig. 5.
High
Low
GOR
pb
pb
p
GOC
1
2
Undersaturated
Oil
3
p
5
Initial Pressure,
p
i
p
Fig. 6 - Oil becomes more undersaturated at lower depths.
1
2
3
4
5
Rs
Initial
conditions
p
Fig. 7 - Varying Rs with depth.
5
Repressuring with Variable Bubble Point
www.petroman.ir
Laboratory Behavior
To describe the two processes, a PVT cell will be considered as laboratory model of what
happens in the reservoir.
Gas Liberation. First, reservoir oil above the bubble point is placed into a PVT cell. A pressure
decrease above the bubble point results in an expansion of the oil which is controlled by the oil
compressibility. After the bubble point is reached, any further decrease in pressure will cause gas
to evolve from solution. As the pressure declines, the gas volume keeps increasing and the oil
volume keeps decreasing. Fig. 8 presents the PVT cell behavior during depletion process.
p < pb
GAS
p = pb
p < pb
GAS
OIL
OIL
OIL
OIL
4500
4000
3500
3000
p > pb
6
Repressuring with Variable Bubble Point
www.petroman.ir
Reservoir Behavior
The process of redissolving gas in the field is more complicated. As a solution gas drive
reservoir below the bubble point is produced, it is found that the gas/oil ratio increases. While this is
happening, the ratio of total gas and oil remaining in the reservoir is changed. In addition to the
amount of gas available, the characteristics of the porous media and the reservoir geometry (dip
angle etc.), make the resolution process more complicate than it was in the laboratory.
Properties Behavior. During the production period the gas/oil ratio of the produced fluids increases
because of the increasing gas saturations and relative permeabilities. Later, after the gas/oil ratio has
peaked, the gas/oil ratio decreases because of the increase of formation volume factor.
Sg > 0
Saturation Curve
Sg = 0
Rs
7
Repressuring with Variable Bubble Point
www.petroman.ir
Calculation procedure:
Oil in place:
OIP
n+1
(1)
where,
OIP =
Vp
So
(2)
Bo
Gas in place:
GIP
n+1
(3)
where,
Vp Sg
Bg
V p S o Rs
(4)
5.615Bo
Solubility:
Rcell =
GIP(S.T. )
OIP(S.T. )
(5)
With this new solubility in the cell go to the curve Rs vs. P and find the pressure that
corresponds to this new solubility. This is the new bubble point pressure of the cell.
To find the other properties above the new bubble point (oil volumetric factor and viscosity
of oil) draw a line from the new bubble point pressure using the same slope that the properties
followed with the original bubble point.
Rs
Bo
8
Repressuring with Variable Bubble Point
www.petroman.ir
(6)
Rs
Bo
co = -
Bg dR s
1 dBo
+
Bo dp
Bo dp
(7)
1 dBo
Bo dp
(8)8
cw = -
1 dBw
Bw dp
(9)9
and
cg = -
1 dBg
Bg dp
(10)10
9
There are three different ways in which the resolution of gas could occur in the reservoir.
Case A (Complete Gas Re-solution). This case is in agreement with the Odeh's repressuring case
(Case 2). If a reservoir is depleted to a low pressure and then the reservoir is repressured by fluid
injection, it can be expected that the gas/oil ratio remaining in the reservoir will be considerably less
than the initial gas/oil ratio. For this case, it will be assumed that gas redissolves along the original
curve present.
GOR
Therefore, gas redissolves just like PVT cell until all the free gas is back in solution. Since
the gas/oil ratio is lower than original, all of the remaining gas redissolves before the original bubble
point is reached. When this occurs, a new bubble point is reached which depends on the gas/oil ratio
remaining in the reservoir. As pressure is increased from this new bubble point, the fluid properties
reflect the new bubble point. The reservoir may be pressured to the original pressure above the
original bubble point. In this case it will be found that the oil properties are different than the original
oil properties. The oil viscosity is higher and the formation volume factor is lower.
Case B (Gas does not Redissolve). Following Odeh's notation this corresponds to Case 1. In this
case the lowest pressure which is achieved in the reservoir becomes the bubble point during
repressuring. Even though free gas is available, it is considered not to redissolve. If this is the case,
the oil properties resemble a "dead" oil even at higher pressures. The oil compressibility is liquidlike. The oil viscosity also resembles a "dead" oil at higher pressures.
One reason for having this case in the field is that gas might freely migrate to a secondary
gas cap, or at least to the top of the layer from which it evolves. We can readily see that the gas is no
longer in intimate contact with the oil in this case. For that reason there is very little resolution of the
gas during repressuring. The gas has been physically removed from the oil. To achieve this scenario
in a simulator requires that the user specify that gas does not redissolve. It may be necessary to
change the fluid properties during a re-start run input the fluid properties at a lower bubble point. To
calculate the new properties (oil FVF and viscosity), using the lowest pressure reached in the
10
Repressuring with Variable Bubble Point
www.petroman.ir
reservoir as new bubble point pressure. From this point draw lines following the same slope that they
have above the original bubble point.
Rs
Bo
It can be seen that the solubility curve is no longer a fundamental relationship between
reservoir oil and gas. Instead the solubility curve reflects some combination of fluid behavior and the
actual mechanics of gas going into solution. Once it is decided how this curve should be constructed,
it is important that the formation volume curve and the oil viscosity curve be constructed, not just
drawn, to be consistent with the amount of gas in solution at various pressures.
CASE A
CASE B
Rs
Rs
11
Repressuring with Variable Bubble Point
www.petroman.ir
Primary
Waterflooding
No resolution
Partial Resolution
(Actual Case)
p
All Gas
Redissolving
t
Fig. 15 - History matching with partial resolution
12
Repressuring with Variable Bubble Point
www.petroman.ir
Sg > 0
Sg > 0
Rs
pi
pi
The same general process occurs even if pressure is not held constant. The usual case will be
that pressure is either increasing or decreasing during the injection period. The trace of the points on
the gas/oil ratio versus pressure curve for different timesteps depends on the history of pressure and
gas injection into the cell. It is important that the simulator be programmed to properly account for
this gas injection above the bubble point. It is also important that user understand how the simulator
behaves and input the proper solubility and fluid property data into the simulator.
13
Repressuring with Variable Bubble Point
www.petroman.ir
Diffusion Effect
The movement of gas in porous media during reservoir repressurizing is explained by
molecular diffusion. Molecular diffusion is effective in miscible displacement of oil by an enriched
gas or in gas-cycling operation.
The rate of molecular diffusion in porous media is very slow. This may explain why the gas
does not redissolve in the oil when the reservoir is repressurized. An estimate of the time needed for
the movement of the gas is determined by the following equation,
t =
13.1 D
Lc
(11)
where,
Lc
D
=
=
=
=
For the gas to move 10 ft in porous media using the above equation, it needs 9.33 years, but
for 20 ft, it needs 37.3 years.
14
Repressuring with Variable Bubble Point
www.petroman.ir
=
=
=
2.316 MMRB
1.797 MMRB
1.620 MMSTB ( Boi = 1.12 RB/STB)
Production Well
-9000
Injection Well
100
-9100
2000
15
Repressuring with Variable Bubble Point
www.petroman.ir
Fig.17 .
Grid. Grid dimensions, well locations and rate schedules used in simulation are shown below.
Grid Dimensions and Well Locations
x
Feet
y
Feet
z
Feet
Number of
Producer
Injector
Cells
Location
Location
10x5x1
200
500
20
50
8,1,1
4,1,1
25x10x1
80
500
10
250
20,1,1
7,1,1
50x20x1
50x40x1
40
40
500
500
5
2.5
1000
2000
40,1,1
40,1,1
14,1,1
14,1,1
Grid Size
Rate Schedule
Time
(Days)
qo
STB/D
qwi
STB/D
Np
MSTB
W pi
MSTB
Stage
0.0
0.0
500
Depletion Phase
0.0
182.499
365
250
0.0
Repressuring Phase
600
303.749
850
291.000
Procedure. Producing approximately 10% of the OOIP/Year resulted in a daily flowrate of 500
STB/D. Production remained constant for one year in order to form a secondary gas cap which
occupied approximately one third of the total reservoir (Fig. 2). In order to increase the average
reservoir pressure, water was injected at a rate of 600 STB/D while production was reduced to 250
STB/D. As the pressure increased, a fraction of the gas redissolved into the oil, but the major portion
of gas become compressed at pressures in excess of the bubble point pressure (Fig. 3). Rates were
held constant throughout the simulation of the reservoir, which ended when reservoir pressures
surpassed the available data (out of table).
16
Repressuring with Variable Bubble Point
www.petroman.ir
Production
Well
Injection
Well
New
Gas/Oil
C t t
Cas/Oil
Contact
Production
Well
Results and Conclusions. Fig. 20 illustrates the pressure and depth of the gas/oil contact as the
reservoir was depleted and repressured. It is apparent that during repressuring, the gas/oil contact
only slightly rises while the pressure at the contact increases dramatically. During the depletion
phase, the pressure at the GOC dropped from 3584 psi to 2923 psi (P = -661 psi), while the
elevation of the GOC fell from -9000 ft to -9049 ft (E = -49 ft). At 850 days, after repressuring to
3855 psi (P = +929 psi), the elevation of the GOC was still located at -9046.4 ft (E = +3.4 ft).
Therefore, regardless of the grid adaptation, a gas cap remains as the reservoir is repressured.
17
Repressuring with Variable Bubble Point
www.petroman.ir
Elevation of GOC,Feet
----
Fig. 20
18
Repressuring with Variable Bubble Point
www.petroman.ir
Pressure
Elevation
X - Direction
21
Well 2
GOR
Y - Direction
Starting Waterflood
21
qo & GOR
qo
Well 1
Production Time
Production and Injection Schedule. We set the bottomhole pressure at 100 psia and restricted the
maximum production rate of 2000 STB per day per well. We first depleted reservoir to an average
pressure of 200 psia then started water injection by converting the Well 1 into an injector and at the
same time kept the Well 2 producing at a constant bottomhole pressure of 100 psia. For comparative
purpose, we ran another case in which the reservoir was depleted in the same way as the previous
case to an average pressure of 200 pisa. Then, instead of injecting water, we shutin Well 1 and
allowed Well 2 to continue production until the average reservoir pressure reached 100 psia.
Results and Discussions. Fig. 23 shows the GOR and qo vs. production time from simulation runs.
The solid line represents the case without water injection; the circles represent the case with water
19
Repressuring with Variable Bubble Point
www.petroman.ir
injection. For the reference purpose, we also include the oil production rate from water injection case
in which it clearly shows the time the oil bank reached the production well. For water injection case,
we started water injection after 400 days of primary depletion. After 120 days of water injection, the
oil bank reached Well 2. It can be seen from the figure that before the oil bank reached Well 2, the
GOR performances with water injection and without water injection agree very well. For water
injection case, after the oil bank reached the production well, the GOR dropped dramatically because
of substantial oil coming out with little free gas. The GOR behavior was not affected by the water
injection until the oil bank reached the production well. Therefore it is clear that no gas bank was
created by water injection. This can be explained as follows: because the mobility of gas is very high
compared to that of the oil behind, the free gas keeps flowing until it stop moving without
establishing a gas bank. Fig. 4 shows the behavior of reservoir pressure and wellbore pressure for
water injection case. This plot provides more information about how the reservoir was depleted and
then repressurized. The solid line represents the reservoir pressure and the dash line represents the
wellbore pressure.
To study how the oil viscosity affects the reservoir performance, we ran another two similar
cases by changing the oil viscosity only to 10 times of the original values. The reason for changing
viscosity of oil is that we expected that the oil viscosity would play an important role in reservoir
performance. Fig. 25 and Fig. 26 present the similar information as Fig. 23 and Fig. 24. It also shows
that no gas bank was created by the water injection.
In addition to the above simulation runs, we also varied some other reservoir properties and
process parameters attempting to simulate the GOR behavior observed in Tibu Field. The results we
obtained are very similar to those of above cases without finding Tibu field phenomena. The
following parameters and reservoir properties were changed:
1) wellbore pressures (for constant pressure production);
2) maximum injection rate and maximum production rate;
3) reservoir pressures ( depleted to different reservoir pressure before water injection);
4) permeability;
5) gas viscosity,
6) pseudo relative permeability curve for segregation.
Conclusion and Recommendation. Using SPE 2 data and 2-D homogeneous model, we did not
see the similar phenomena encountered in the Tibu field. Since the mobility of gas is much high than
that of oil, water injection can not create a gas bank before oil bank. We conclude that the Tibu field
phenomena is not a simple phenomena which can be simulated using above 2-D model.
Since Tibu field is a multilayered reservoir with a high degree of heterogeneity, we would
expect that the phenomena observed in Tibu Field might be caused by reservoir heterogeneity and
multilayer system. Our 2-D, homogeneous reservoir simulation model could not model
thiwaterflood performance for such complicated reservoir. A 3-D model is recommended to simulate
the Tibu field phenomena.
20
Repressuring with Variable Bubble Point
www.petroman.ir
water injection
begins
oil bank
reaches Well 2
Fig. 23 - Comparison of GOR: with and without water injection (SPE2 data).
water injection
begins
oil bank
reaches Well 2
Fig. 24 - Reservoir pressure performance: for the case with water injection.
21
Repressuring with Variable Bubble Point
www.petroman.ir
GOR, scf/stb
water injection
begins
oil bank
reaches Well 2
oil bank
reaches Well 2
water injection
begins
Fig. 26 - Reservoir pressure performance: for the case with water injection
(SPE2 data with oil viscosity increased 10 times).
22
Repressuring with Variable Bubble Point
www.petroman.ir
Exercises
Class Problem No. 90 (Repressuring)
A cell of a "black oil" simulator is being repressured following the original Rs curve. Consider the
cell to be isolated from other cells. Use Odeh's data to solve the following problems.
The following conditions were observed for the cell at a lower pressure:
Sg
So
p
=
=
=
0.10
0.50
1,000 psia
(1) Find the new bubble point that will be obtained during repressuring.
(2) Find the correct values of o and Bo at a pressure of 2,000 psia.
23
Repressuring with Variable Bubble Point
www.petroman.ir
CHAPTER 3
History Matching
Technique
Purpose
Quality of Data
Production Allocation
Pressure Analysis
Continuity Problem
www.petroman.ir
History Matching
History matching is the part of reservoir simulation which often requires most of the time
and money. This is a process of trying to reconstruct a field's past performance with model runs. An
example of a history matching comparison is shown below, in which the model pressures are
compared with actual pressures from the field build-up tests. After the history match is completed
within the desired tolerances, or within the time and money constraints allowed for history matching,
then prediction runs are performed. In this case three prediction runs are made (Cases A, B, and C),
all beginning at the end of the history matching period.
Field data
Simulation
Field data
Simulation
Pdatum
WOR
time
time
1
History Matching
www.petroman.ir
Technique
The usual technique is to input actual oil production and compare the model values of gas
and water production against the field values. When simulating the behavior of oil reservoirs, oil
production data will always be the most accurate because it is the value that must be accurately
measured for sales purposes. This comparison is also made for bottom-hole flowing pressure,
individual cell pressures, and average reservoir pressure. The oil production is usually available in
monthly periods, but for the purposes of data input, the monthly production is allocated on a daily
basis using a simple average unless detailed well records are available.
Field data
Simulation
Predictions
Pwell
Case A
Case B
Case C
Time
2
History Matching
www.petroman.ir
Purpose
The purpose of history matching is to validate the model for use in predictions and also to
understand the reservoir mechanisms better. We will have much more confidence in the validity of
our model if past reservoir performance data is available. In some cases, the complications that arise
from history matching might make it necessary to acquire additional field data or to analyze the
existing data differently. Very often the information learned during the history match process leads
to a re-interpretation of the geologic study. Another purpose of the history matching procedure is to
simply provide initial reservoir conditions for the predictive runs.
Purpose of history matching:
1) Calibration of model.
a) To make predictions
b) To understand reservoir mechanisms and geology.
2) Provide initial conditions for predictions.
Many simulation projects have been completed without field performance history. These
projects are much easier to perform because the tedious history matching procedure is eliminated. In
these cases the simulator should be updated as field data becomes available, otherwise the actual
performance of the reservoir may not be modeled accurately. Reservoir simulation is usually used in
these cases because it is capable of providing more accurate results and including more aspects of
reservoir description and operating conditions than could be done by hand methods, and minor
adjustments may be done easily.
3
History Matching
www.petroman.ir
Quality of Data
Accuracy of Model vs. Its Importance. An interesting comparison can be made between the
accuracy of a model, as determined by history matching, and the importance of having a good
reservoir simulation study. At the time of abandonment of a field, it is possible to get the most
complete description of a reservoir from performance history matching. However, at that time a
reservoir study has little value. At the beginning of production, the need for accurate predictions and
reservoir description is the most important. It is at this time that important capital expenditures are
made and perhaps irreversible decisions are made about how the field should be produced. At the
beginning of production we do not have the benefit of performance history matching so we can say
that the model has the least accuracy at this time.
Pressure Data. It is important to realize that the performance data that is being matched might have
significant errors in it. The quantity and accuracy of the pressure data, for example, is limited by
money available for testing and the sensitivity and the accuracy of the pressure measuring device.
Pressure errors can also be made because of poor calibration of the tool or incorrect analysis of test
results.
Accuracy
Importance
Decision
Value
of
Simulation
Discovery Delineation
Development
Secondary Infill
EOR
Abandonment
Life of a Reservoir
Fig. 4 - Relative importance and accuracy of a simulation study during the life of a producing well.
Pressure Data:
(a) Quantity of data.
(b) Sensitivity/Accuracy of tool.
(c) Calibration of tool.
(d) Analysis of test.
Production Data. It is also important to realize that production data, even oil production data, is not
always accurate. The overall lease or tract production might be considered to be accurate, but
4
History Matching
www.petroman.ir
individual well production rates are usually allocated from field tests that are done periodically, if at
all. Sometimes testing procedures are not frequent enough or accurate enough. For oil reservoirs,
the gas and water production rates are likely to be in greater error than the oil production rate. The
engineer should know whether these rates were metered during tests and whether they were metered
monthly at the tank battery. In many cases water production is not needed but is estimated from
production tests of brief duration and sometimes questionable accuracy. If wrong rates are input into
the simulator, then the simulator rates and the resulting pressures will be of questionable accuracy.
Production data:
(a) Frequency of test.
(b) Accuracy of test.
(c) Available equipment/meters.
(d) Allocation of rates.
5
History Matching
www.petroman.ir
Production Allocation
The sketch below shows how production might be allocated in a typical production setup.
Each of the wells will have an individual flow line that goes to the manifold at a common point.
Each well can then be directed towards the production separator, along with the other wells, or to the
test separator by itself. The normal procedure is to test a well for 24 hours at a frequency of,
perhaps, once a month. The monthly production from all of the wells will be the most accurate
measured value but there are other complications such as wells being down during periods of the
month. The monthly production rates at the sales point are allocated to individual wells based on
their test rates and the producing time of each well. The allocated rates would not usually be exactly
the same as the test production rates. Then it is necessary to apply some correction factor to make
the monthly well production add up to that measured at the sales point.
The same procedure is, more or less, followed for the gas and water phases. These values
will be more accurate if the rates are metered at the tank battery during normal operation and
adjusted with test production data.
Well 1
Test Separator
Tank
Battery
Well 2
Monthly
Measured
Total Oil
Production
Production
Separator
Well 3
Metered or monthly values at tank battery: qo= 450 B/D, qw= 20 B/D
There are some problems to avoid in history matching that might not be obvious to an
inexperienced engineer. One of the most common problems is the wrong rate problem.
In the example reservoir in Figure 6, a number of wells are producing from the oil leg of the
reservoir at low solution gas/oil ratios. One well is near the gas/oil contact and produces at high
gas/oil ratios. The simulator run is made with a fixed oil production rate and the gas/oil ratio is
calculated from the cell saturations. In this case, the calculated gas/oil ratio is much too high
because the model oil saturation is approaching residual oil saturation. The model gas/oil ratio
approaches infinity and so does the gas production rate since the oil production rate is fixed. We
find that the pressure behavior of the model is meaningless because the gas rate in this one well is
extremely high.
In order to solve this problem, the total production rate must not be controlled by the lower
rate phase (oil in this case) if the total reservoir production rate is to be accurately matched. This
total production rate must be matched before the reservoir pressure can be matched.
A simple check to make at the end of a history match run is to make sure that all three phases
produce about the same cumulative production as observed in the field. If one of the phases is
significantly off, then that rate must be adjusted before pressure can be matched.
Some simulators make an automatic adjustment of the rates to honor the observed total
reservoir rates, rather than the oil rates. This requires data input of oil, gas, and water production
rates. At each timestep, the total reservoir production rate is calculated and matched with the
observed field rate. When this calculation is performed, the total reservoir production rate will be
correct and the pressure match can then be performed.
Solution. The problem is that qg approaches infinity since the model kro approaches zero. We can
see the effect on qt by remembering that:
7
History Matching
www.petroman.ir
When using the total production rate, input a value of (qg)max, which is the highest value of qg
seen in the field. A second solution method is to input qg instead of qo when GOR > 4,000 scf/bbl
and subsequently calculate qo. A third method is to input qt (or qo, qw, and qg) observed and then
match (qt)model as shown in the following table:
qo
qw
WOR
qt
FIELD
50
100
2.0
150
MODEL
50
150
3.0
200
37.5
112.5
3.0
150
MODEL (Adjusted)
Gas
qo
= 3 BOPD
GOR = 20,000 scf/bbl
Simulation
Field Data
Oil
Water
Time
Fig. 6 - Wrong rate problem for the well near gas-oil contact (GOC).
8
History Matching
www.petroman.ir
Pressure Analysis
We can analyze different aspects of pressure performance and make some generalities about
how this may be adjusted.
Gradients (Cross-Section). The pressure gradients can be observed from pressure profiles
constructed at particular times where field data is available. The gradients may lead to information
about the continuity of the formation or the transmissibility of the reservoir.
k
a) u t =
b) faults
c) h contour
In figure 7, the model shows different trend than the field data. One of the probable cause
for the difference is that high absolute permeability used in the simulation model. Decreasing
absolute permeability in the simulation may improve the match.
Pseudosteady-State (Closed reservoir). Pseudosteady-state behavior can be analyzed when it
occurs in the model and in the field. The matching of this data requires that the total reservoir
production rate be correct. History matching can lead to better estimates of pore volume or total
compressibility.
qt
dp
= dt
V p ct
(1)
Liquid/rock compressibilty
dominates
Average
p
Gassy compressibility
dominates
Transient
state
Pseudo-steady state
Well
Pressure
Bubble point
Time
Time
History Matching
www.petroman.ir
(2)
After the reservoir pressure goes below bubble point, total compressibility increases
drastically because of increase in oil compressibility and generation of gas saturation and gas
compressibility. Below bubble point, oil compressibility dominates over other terms.
Transient Pressure Behavior. Transient behavior can be analyzed using the dimensionless
pressure and dimensionless time and the results can be used as another tool in the parameter
adjustment process during history matching. The results of this analysis apply only to transient flow
periods.
It should be noted that mobility is a factor in the magnitude of the pressure drop (as indicated
by the dimensionless pressure equation) and the timing at which a particular feature occurs (from
dimensionless time). The thickness only affects the magnitude of pressure drop. The porosity or
total compressibility affect only the timing.
pD =
kh( pi - pwf )
tD =
141.2 q B
h*2
k*2
0.000264 k t hrs
ct*2 or *2
10
History Matching
www.petroman.ir
(3)
ct rw 2
V p So
(4)
Bo
qt
dp
= dt
V p ct
(5)
OOIP can be accurately determined when the following conditions are met:
If
(1) good depletion history
(a) good data
(b) significant pressure decline
(c) volumetric (closed) reservoir
and
(2) ct is dominated by coSo
and
(3) good PVT data for oil
Assume:
ct = c f + co S o + c w S w + c g S g
(6)
ct co S o
(7)
Then,
- qt
dp
=
V p co S o
dt
1
qt
co Bo OOIP
(8)
Or:
- q t dp
OOIP =
/
co Bo dt
(9)
Notes:
(1) co is more likely to dominate below the bubble point. Above the bubble point cf and cw
become important and data may not be accurate.
(2) Original gas in place (OGIP) is measured by extrapolating a plot of p/z versus Gp. Hence,
if a strong water drive is present in the reservoir, wrong OGIP will be obtained.
(3) If a good history match is obtained, it does not necessarily mean a good prediction of
OOIP can be obtained.
11
History Matching
www.petroman.ir
(4) Generally, values of permeability and porosity used in the simulation are obtained from
the core analysis. During core analysis, permeability is calculated by establishing a
steady state flow through the core. The porosity is obtained by expansion process
following Boyles law. Good evaluation during core analysis will lead to good history
matching and OOIP calculation.
(5) If injection has been started at the beginning of the life of the reservoir, then it is highly
unlikely that a good prediction of OOIP can be obtained.
Continuity Problem
Sometimes, it is difficult to predict the continuity of the reservoir. For example, suppose,
two identical wells are producing at the same rate as shown in figure 11. In this case, we cannot
conclude about the continuity at the no flow boundary which is midway between the two wells.
Presence of continuity at the no flow boundary may not be important during primary depletion.
However, continuity is important during secondary recovery when one of the wells may be changed
to an injection well. Interference well testing can determine continuity. This type of factor should
be taken into account for successful prediction using simulator.
Well 2
Well 1
Reservoir
No flow boundary
12
History Matching
www.petroman.ir
13
History Matching
www.petroman.ir
14
History Matching
www.petroman.ir
Make Run
All q = 0
Check
Run
Validity
Sight-Check
for Data
Errors
Are
Fluids
Equilibrated ?
No
YES
GO TO STEP 2
15
History Matching
www.petroman.ir
Identify
Problem and
Correct
Step 2 is the pressure match. During this step it is required that the total reservoir production
rate be specified for each well so that the wrong rate problem can be avoided. The producing
ratios may be in error, but the total reservoir production rate must be correct. When we have
accomplished this, then we can think of modeling Martin's equation using total mobility, total
compressibility, and total reservoir production rate.
The first step is to match the average reservoir pressure. This is equivalent to checking the
material balance for the reservoir. Adjustment in the pore volume or compressibility must be made
at this point. If an aquifer exists, then the level of aquifer support will be matched at this point.
The next step is to check pressure gradients at particular times to see if the reservoir
continuity and transmissibility are correct.
Finally, when the gross reservoir adjustments are completed, then individual well areas can
be adjusted.
After step 2 is completed, then we can go to step 3.
16
History Matching
www.petroman.ir
Are
qt
Close ?
Check Total
Rates
Adjust
Rate
qmax
Controls
NO
YES
p
Level
OK ?
Check Material
Balance
Adjust
, ct, h
Area
NO
YES
Are
p gradients
OK ?
Check
Gradients
NO
Adjust
kh/
YES
Fine Tune
Near Well
NO
Individual
Wells
OK ?
YES
GO TO STEP 3
17
History Matching
www.petroman.ir
Adjust
kh/
18
History Matching
www.petroman.ir
We enter step 3 with the correct total reservoir producing rates, but may have erroneous
producing ratios. The adjustments in step 3 are ones that involve saturations, relative permeabilities,
water/oil contacts, and gas/oil contacts. The relative permeability adjustments that are required to
match the producing ratios may be done with pseudo relative permeabilities if a 2-D model or a
coarse 3-D model is used.
19
History Matching
www.petroman.ir
Check Run
Validity
FROM STEP 2
Are
WOR & GOR
OK ?
Check Total
Rates
Adjust
Field-Wide
Kr Curves
Leaving t
Same
NO
YES
Check Material
Balance
Are
Well
WOR & GOR
OK ?
Adjust
Contacts
Special Kr
Curves &
Completions
NO
YES
Are
Well q
OK ?
Check
Gradients
NO
Put in
Rate (Oil)
YES
Fine Tune
Near Wells
NO
Is
p match
OK ?
YES
GO TO STEP 4
20
History Matching
www.petroman.ir
Go To
Step 2
21
History Matching
www.petroman.ir
Step 4 is the final step in history matching. Once the pressures and producing ratios have
been matched then the productivity indices can be matched. This is done by adjusting the model's
productivity index (PI) to give the actual field bottom hole flowing pressures at the end of the history
period. The bottom hole flowing pressure must be accurate at this point because it will usually
control the producing rates during the prediction period.
For a 2-D model, the productivity index adjustment can be made by hand without even
making a simulator run. The bottom hole pressure calculation is external to the cell pressure
calculation in the simulator and is not involved in the matching of cell pressures.
If the model is a 3-D model with commingled wells, then the adjustment of productivity
indices has to be done on an individual layer basis. This affects the flow splitting from the
different producing intervals in the model wells, which may in turn affect the pressure behavior.
This type of adjustment can be tedious but can be aided by downhole producing information that
indicates rates from different producing intervals (i.e., tracer surveys or flowmeter surveys).
22
History Matching
www.petroman.ir
Adjust P.I. To
Match Pwf
Make
Final
Run
GO TO PREDICTION
23
History Matching
www.petroman.ir
24
History Matching
www.petroman.ir
SYNTHETIC HISTORY
MATCHING
RUN #1
RUN
#1
ACTUAL
HISTORY
Change
Data
RUN #2
RUN
#2
ACTUAL
HISTORY
Change
Data
RUN #3
Actual
History
RUN
#3
ACTUAL
HISTORY
25
History Matching
www.petroman.ir
EXERCISES
RESERVOIR SIMULATION
CLASS PROBLEM NO. 62
The reservoir shown below is homogeneous, isotropic, and 2-D. Wells A and B are producing at
a constant rate for 5 years. At that time, well A is converted to an injection well. The injection
rate in Well A and the production rate in well B are held constant over the remaining 5 years.
The observed pressure data and the results of history match run #1 are attached. The input and
output data files for the simulation runs are also provided.
Analyze the pressure performance and recommend the changes required in reservoir parameters
to match the pressure behavior.
I=1
J=1
2
3
4
5
6
7
26
History Matching
www.petroman.ir
PROB_62
TIME, YRS
Well A
0.12
0.2
0.29
0.36
0.46
0.55
0.63
0.71
0.78
0.87
0.95
1.03
1.12
1.2
1.28
1.36
1.46
1.53
1.61
1.7
1.78
1.87
1.95
2.01
2.11
2.18
2.28
2.36
2.44
2.52
2.6
2.68
2.76
2.83
2.92
3
3.09
3.17
3.25
3.34
3.42
3.51
3.6
3.67
3.75
3.84
3.91
3.99
4.07
4.16
4.25
4.32
4.41
4.49
4.57
4.66
4.74
4.82
4.9
4.97
5.06
5.14
5.23
5.3
Obs. Press.
Well A
3821.9
3784.6
3744.7
3710.1
3667.5
3632.8
3595.6
3561
3523.7
3486.4
3449.2
3411.9
3379.9
3340
3305.4
3270.8
3233.5
3190.9
3156.3
3119
3084.4
3041.8
3009.9
2975.3
2946
2916.7
2900.6
2889.9
2876.6
2865.9
2852.5
2841.8
2828.4
2820.4
2812.3
2801.6
2790.9
2782.8
2772.1
2761.4
2753.4
2742.6
2734.6
2721.2
2710.5
2702.5
2691.8
2681
2673
2664.9
2654.2
2648.8
2635.5
2627.4
2622
2616.6
2605.9
2597.8
2589.8
2579.1
2730.4
2802
2860.4
2918.7
TIME, YRS
Well B
0.14
0.23
0.31
0.39
0.49
0.57
0.64
0.74
0.8
0.9
0.98
1.06
1.14
1.23
1.31
1.38
1.48
1.56
1.64
1.73
1.81
1.89
1.97
2.05
2.13
2.22
2.3
2.38
2.47
2.54
2.62
2.7
2.79
2.85
2.96
3.03
3.11
3.2
3.29
3.36
3.44
3.53
3.61
3.68
3.78
3.86
3.94
4.02
4.09
4.18
4.27
4.35
4.43
4.51
4.6
4.68
4.77
4.85
4.93
5.01
5.08
5.17
5.26
5.33
Obs. Press.
Well B
TIME, YRS
3854.8
3817.8
3778.1
3738.5
3698.9
3661.9
3624.9
3587.9
3548.3
3511.2
3476.9
3439.9
3397.6
3360.6
3321
3284
3244.4
3207.4
3173
3130.7
3096.4
3059.4
3022.4
2982.7
2958.9
2927.2
2916.6
2900.7
2887.5
2876.9
2866.3
2855.6
2842.4
2834.4
2823.8
2810.6
2800
2792
2778.7
2770.8
2760.2
2746.9
2738.9
2731
2720.3
2712.4
2701.8
2693.8
2685.8
2677.8
2664.6
2656.6
2648.6
2640.7
2627.4
2622.1
2616.8
2608.8
2598.2
2590.2
2582.2
2576.9
2571.6
2568.9
27
History Matching
www.petroman.ir
0.003
0.08
0.16
0.25
0.33
0.41
0.49
0.58
0.66
0.74
0.82
0.90
0.99
1.07
1.15
1.23
1.32
1.40
1.48
1.56
1.64
1.73
1.81
1.89
1.97
2.05
2.14
2.22
2.30
2.38
2.47
2.55
2.63
2.71
2.79
2.88
2.96
3.04
3.12
3.21
3.29
3.37
3.45
3.53
3.62
3.70
3.78
3.86
0.00
4.03
4.11
4.19
4.27
4.36
4.44
4.52
4.60
4.68
4.77
4.85
4.93
5.00
5.08
5.16
H.M. #1
Well A
3991.3
3957.1
3937.1
3917.1
3877.2
3857.3
3817.4
3797.5
3757.6
3737.6
3717.7
3677.8
3640.6
3620.7
3600.7
3580.8
3540.9
3520.9
3480
3469
3458.9
3449.1
3439.4
3429.9
3420.7
3411.9
3403.3
3394.9
3386.6
3378.7
3370.8
3363
3355.3
3347.7
3340.1
3332.7
3325.3
3318
3310.8
3303.8
3296.8
3290.1
3283.4
3276.8
3270.2
3263.7
3257.2
3250.8
3244.4
3238.1
3231.9
3225.7
3219.6
3213.5
3207.5
3201.7
3195.9
3190.1
3184.3
3178.6
3173
3168.3
3188.3
3288.3
H.M. #1
Well B
3991.3
3957.1
3937.1
3917.1
3877.2
3857.3
3817.4
3797.5
3757.6
3737.6
3717.7
3677.8
3640.6
3620.7
3600.7
3580.8
3540.9
3520.9
3480
3469
3458.9
3449.1
3439.4
3429.9
3420.7
3411.9
3403.3
3394.9
3386.6
3378.7
3370.8
3363
3355.3
3347.7
3340.1
3332.7
3325.3
3318
3310.8
3303.8
3296.8
3290.1
3283.4
3276.8
3270.2
3263.7
3257.2
3250.8
3244.4
3238.1
3231.9
3225.7
3219.6
3213.5
3207.5
3201.7
3195.9
3190.1
3184.3
3178.6
3173
3168.3
3166.6
3162.9
PROB_62
TIME, YRS
Well A
5.37
5.46
5.54
5.61
5.7
5.78
5.86
5.94
6.03
6.12
6.21
6.29
6.38
6.45
6.53
6.61
6.69
6.77
6.86
6.93
7.01
7.09
7.2
7.27
7.34
7.42
7.5
7.59
7.67
7.75
7.84
7.93
8.02
8.1
8.18
8.27
8.33
8.42
8.51
8.59
8.66
8.75
8.83
8.91
8.99
9.06
9.14
9.23
9.31
9.4
9.48
9.56
9.65
9.73
9.82
9.9
9.93
Obs. Press.
Well A
2985
3067.3
3298.3
3425.7
3478.7
3505.2
3526.3
3539.5
3550.1
3565.9
3576.4
3587
3600.2
3610.7
3621.3
3629.1
3631.7
3639.6
3642.1
3647.4
3652.6
3657.8
3663
3670.9
3678.8
3684.1
3694.6
3705.1
3718.3
3728.9
3747.4
3763.2
3787
3802.9
3821.4
3834.6
3845.1
3850.3
3855.5
3863.4
3868.7
3873.9
3876.5
3881.7
3886.9
3894.8
3897.4
3902.6
3905.2
3910.4
3918.3
3923.5
3928.7
3936.6
3939.2
3944.4
3949.7
TIME, YRS
Well B
5.42
5.51
5.58
5.66
5.74
5.82
5.9
5.98
6.06
6.15
6.25
6.32
6.47
6.58
6.66
6.75
6.82
6.9
6.98
7.06
7.13
7.21
7.26
7.29
7.38
7.46
7.55
7.65
7.73
7.81
7.89
7.97
8.05
8.11
8.2
8.3
8.37
8.45
8.56
8.64
8.72
8.8
8.88
8.97
9.05
9.13
9.2
9.27
9.36
9.46
9.54
9.62
9.7
9.78
9.86
9.95
Obs. Press.
Well B
TIME, YRS
2563.5
2563.5
2560.8
2558.1
2558
2558
2557.9
2557.9
2563.1
2563
2563
2565.6
2568.1
2573.3
2573.2
2575.8
2578.4
2581
2580.9
2586.1
2586.1
2588.7
2591.3
2593.9
2596.5
2599.1
2599
2598.9
2604.2
2604.1
2609.3
2611.9
2614.5
2614.4
2619.7
2622.2
2624.8
2627.4
2635.2
2635.2
2640.4
2643
2648.2
2650.8
2650.7
2656
2661.2
2663.8
2666.4
2668.9
2674.1
2676.7
2679.3
2681.9
2687.1
2689.7
28
History Matching
www.petroman.ir
H.M. #1
Well A
H.M. #1
Well B
5.25
5.33
5.41
5.49
5.58
5.66
5.74
5.82
5.90
5.99
6.07
6.15
6.23
6.32
6.40
6.48
6.56
6.64
6.73
6.81
6.89
6.97
7.05
7.14
7.22
7.30
7.38
7.47
7.55
7.63
7.71
7.79
7.88
7.96
8.04
8.12
8.21
8.29
8.37
8.45
8.53
8.62
8.70
8.78
8.86
8.95
9.03
9.11
9.19
9.27
9.36
9.44
9.52
9.60
9.68
9.77
9.85
3308.3
3328.3
3348.3
3356.5
3360.5
3365
3370
3375.6
3381.8
3388.6
3394.2
3400.1
3404.2
3409.5
3415
3420.3
3423.9
3426.9
3429.7
3432.2
3434.6
3436.9
3439.4
3441.9
3444.5
3447.3
3450.1
3452.9
3455.8
3458.9
3461.7
3464.6
3467.6
3470.6
3473.7
3476.9
3480.1
3483.4
3486.7
3489.9
3493.1
3496.2
3499.3
3502.4
3505.7
3508.6
3511.5
3514.3
3517.2
3520
3522.8
3525.6
3528.4
3531.2
3534
3536.8
3539.6
3162.4
3162.4
3162.9
3164
3165.2
3166.6
3167.9
3169.3
3170.6
3171.8
3173.1
3174.4
3175.7
3177
3178.2
3179.4
3180.7
3182.1
3183.5
3184.9
3186.4
3187.8
3189.3
3190.7
3192.1
3193.5
3195
3196.4
3197.8
3199.2
3200.6
3202
3203.3
3204.7
3206.1
3207.4
3208.7
3210
3211.3
3212.7
3214
3215.4
3216.8
3218.2
3219.5
3220.9
3222.3
3223.7
3225.1
3226.5
3227.9
3229.3
3230.7
3232
3233.4
3234.9
3236.3
9.93
3542.4
3237.8
10.00
3544.6
3239
PRESSURE, PSI
3,500
3,000
2,500
2,000
0
10
TIME, DAYS
OBSERVED
H.M. #1
PRESSURE, PSIA
3500
3000
2500
2000
0
TIME, YEARS
OBSERVED
29
History Matching
www.petroman.ir
H.M. #1
10
/
SIMULATOR INPUT DATA
/
Problem 62
/
Run 1
/
/
SINGLE VALUE DATA
/
IMAX 7
/Number of grid blocks in X-direction
JMAX 7
/Number of grid blocks in Y-direction
PREF
CROC
CWAT
COIL
CRVO
VISW
PRES
/
P
Bo
Bg
Vo
Vg
Rs
/
psi
rb/stb rcf/scf cp
cp
scf/stb
500 1.19 .04
0.75 .0118 350
1000 1.25 .015 0.56 .0132 475
2000 1.41 .007 0.38 .0168 850
/PVT table
3000 1.61 .005 0.27 .0215 1250
3500 1.77 .004 0.25 .024 1575
END
/
/ GRID DATA
/
DX 7*600
/Reference depth for top of grid blocks
DY 7*600
/
H
Por
Perm Elev
GRID
000
30
0.25 50
-5000
/Thickness, porosity, perm, & elev.
PO
000
4000
/Initial reservoir pressure
PSAT 0 0 0 3500
/Bubble point pressure
SW
0 0 00.3
/Initial water saturation
SG
000
0.0
/Initial gas saturation
END
/
/ SCHEDULE DATA
/
NAME 1
I = 2, J = 4, K =1 PI = 3.375
/Well location and productivity index
QO 1
150.0
/Initial production rate
PLIM 1
100.0
/BHP constraint
NAME 2
I = 6, J = 4, K =1 PI = 3.375
QO 2
150.0
PLIM 2
100.0
WLRT 1
PSRT 1
DELT
30
/Timestep increment
DTMAX
30
TIME
1825
QW 2
-400.0
/Well status changed to water injection
PLIM 2
20000
TIME
3650
END
31
History Matching
www.petroman.ir
Problem No. 62
History Match Run 1 - Output
NUMBER OF GRIDS
REGION ACTIVE INACTIVE
------ ------ -------TOTAL
49
0
STEP
---TS
1
TS
2
TS
3
TS
4
TS
5
TS
6
TS
7
TS
8
TS
9
TS 10
TS 11
TS 12
TS 13
TS 14
TS 15
TS 16
TS 17
TS 18
TS 19
TS 20
TS 21
TS 22
TS 23
TS 24
TS 25
TS 26
TS 27
TS 28
TS 29
TS 30
TS 31
TS 32
TS 33
TS 34
TS 35
TS 36
TS 37
TS 38
TS 39
TS 40
TS 41
TS 42
TS 43
TS 44
TS 45
TS 46
TS 47
TS 48
TS 49
TS 50
TS 51
TS 52
TS 53
TS 54
TS 55
TS 56
TS 57
TS 58
TS 59
TS 60
TIME
DAYS
---30
60
90
120
150
180
210
240
270
300
330
360
390
420
450
480
510
540
570
600
630
660
690
720
750
780
810
840
870
900
930
960
990
1020
1050
1080
1110
1140
1170
1200
1230
1260
1290
1320
1350
1380
1410
1440
1470
1500
1530
1560
1590
1620
1650
1680
1710
1740
1770
1798
DELT
DAYS
-----30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
27.50
*****
OIL
(MMSTB)
---------9.44359
PAVG
PSI
---3970
3941
3911
3881
3852
3822
3792
3762
3733
3703
3673
3643
3613
3584
3554
3524
3494
3487
3476
3466
3455
3446
3436
3426
3417
3408
3399
3390
3382
3373
3365
3357
3349
3341
3334
3326
3319
3311
3304
3297
3290
3283
3276
3269
3263
3256
3250
3243
3237
3231
3225
3219
3213
3207
3201
3195
3189
3184
3178
3173
FIELD PRODUCTION
OIL
GAS
WATER
B/D
MCF/D
B/D
------- ------- ------300
473
5
300
473
5
300
473
5
300
473
5
300
473
5
300
473
5
300
472
5
300
473
5
300
473
5
300
473
5
300
472
5
300
473
5
300
472
5
300
472
5
300
472
5
300
473
5
300
468
5
300
467
5
300
465
5
300
463
5
300
461
5
300
459
5
300
457
5
300
455
5
300
453
5
300
452
5
300
450
5
300
448
5
300
446
5
300
445
5
300
443
5
300
442
5
300
440
5
300
439
5
300
437
5
300
436
5
300
434
5
300
433
5
300
431
5
300
430
5
300
429
5
300
427
5
300
426
5
300
425
5
300
423
5
300
422
5
300
421
5
300
419
5
300
418
5
300
417
5
300
416
5
300
415
5
300
414
5
300
412
5
300
411
5
300
410
5
300
409
5
300
408
5
300
407
5
300
406
5
*****
SOLVENT
(MMMSCF)
--------.000000
FIELD INJECTION
GAS
WATER
MCF/D
B/D
------- ------0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
PORE
VOLUME
(MMRCF)
----------132.300
AVERAGE SATURATIONS
OIL
GAS
WATER SOLVENT
(FRC) (FRC) (FRC)
(FRC)
----- ----- --------.700 .000
.300
.000
32
History Matching
www.petroman.ir
MAXIMUM
SATURATION
ERROR I J K
----- -- -- -.000 2 4 1
.000 6 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 6 4 1
.000 2 4 1
.000 6 4 1
.000 6 4 1
.000 2 4 1
.000 6 4 1
-.001 2 4 1
.000 6 4 1
.000 6 4 1
.000 6 4 1
.000 6 4 1
.000 2 4 1
.000 2 4 1
.000 6 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 6 4 1
.000 2 4 1
.000 6 4 1
.000 2 4 1
.000 6 4 1
.000 6 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 6 4 1
.000 6 4 1
.000 6 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 6 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 6 4 1
.000 2 4 1
.000 6 4 1
.000 2 4 1
.000 6 4 1
.000 6 4 1
.000 2 4 1
.000 6 4 1
.000 2 4 1
TIMESTEP
61 AT
1825.0 DAYS
NAME I J K STAT
PI
---- -- -- -- ---- ----1
2 4 1 QO
3.4
2
6 4 1 QO
3.4
PWF
(PSI)
-----3130.0
3130.0
WELLS
----2
I=
1
2
3
J 4
5
6
7
1
3171.37
3169.47
3165.90
3162.15
3165.90
3169.47
3171.37
STEP
---TS 61
TS 62
TS 63
TS 64
TS 65
TS 66
TS 67
TS 68
TS 69
TS 70
TS 71
TS 72
TS 73
TS 74
TS 75
TS 76
TS 77
TS 78
TS 79
TS 80
TS 81
TS 82
TS 83
TS 84
TS 85
TS 86
TS 87
TS 88
TS 89
TS 90
TS 91
TS 92
TS 93
TS 94
TS 95
TS 96
TS 97
TS 98
TS 99
TS 100
TS 101
TS 102
TS 103
TS 104
TS 105
TS 106
TS 107
TS 108
TS 109
TS 110
TS 111
TIME
DAYS
---1825
1855
1885
1915
1945
1975
2005
2035
2065
2095
2125
2155
2185
2215
2245
2275
2305
2335
2365
2395
2425
2455
2485
2515
2545
2575
2605
2635
2665
2695
2725
2755
2785
2815
2845
2875
2905
2935
2965
2995
3025
3055
3085
3115
3145
3175
3205
3235
3265
3295
3325
* * * WELL RATES * * *
OIL
GAS
WATER
(B/D)
(MCF/D)
(B/D)
-------- -------- -------150.0
202.4
2.4
150.0
202.4
2.4
300.0
.0
404.8
.0
AVERAGE SATURATIONS
OIL GAS
WATER
(FRC) (FRC) (FRC)
----- ----- ----.630 .069 .301
GOR
WCT
(SCF/B) (PCT)
------- ----1349.3
1.6
1349.3
1.6
4.8
.0
1349.3
PAVG
PSI
---3168
3170
3171
3173
3175
3176
3178
3180
3182
3183
3185
3187
3189
3191
3192
3194
3196
3198
3199
3201
3203
3205
3206
3208
3210
3212
3213
3215
3217
3219
3221
3222
3224
3226
3228
3230
3231
3233
3235
3237
3239
3241
3243
3244
3246
3248
3250
3252
3254
3256
3258
CUMULATIVE PRODUCTION
OIL
GAS
WATER
MB
MMCF
MB
------- ------- ------274
405
4
274
405
4
547
0
810
0
SATURATIONS
SO SG SW
(FRC PORVOL)
--- --- --.62 .08 .30
.62 .08 .30
9
0
2
2
PRESSURE (PSI)
TIME= 1825.00
2
3
4
5
6
3171.44
3171.82
3172.03
3171.82
3171.44
3169.29
3170.17
3170.63
3170.17
3169.29
3164.24
3167.10
3168.30
3167.10
3164.24
3152.77
3163.85
3166.54
3163.85
3152.77
3164.24
3167.10
3168.30
3167.10
3164.24
3169.29
3170.17
3170.63
3170.17
3169.29
3171.44
3171.82
3172.03
3171.82
3171.44
DELT
DAYS
-----27.50
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
1.6
PORE VOLUME
AVERAGED
PRESSURE
---------3167.9
FIELD PRODUCTION
OIL
GAS
WATER
B/D
MCF/D
B/D
------- ------- ------300
405
5
150
202
2
150
202
2
150
202
2
150
202
2
150
202
2
150
202
2
150
202
2
150
202
2
150
202
2
150
203
2
150
203
2
150
203
2
150
203
2
150
203
2
150
203
2
150
203
2
150
204
2
150
204
2
150
204
2
150
204
2
150
204
2
150
204
2
150
205
2
150
205
2
150
205
2
150
205
2
150
205
2
150
205
2
150
206
2
150
206
2
150
206
2
150
206
2
150
206
2
150
206
2
150
207
2
150
207
2
150
207
2
150
207
2
150
207
2
150
207
2
150
208
2
150
208
2
150
208
2
150
208
2
150
208
2
150
208
2
150
209
2
150
209
2
150
209
2
150
209
2
7
3171.37
3169.47
3165.90
3162.15
3165.90
3169.47
3171.37
FIELD INJECTION
GAS
WATER
MCF/D
B/D
------- ------0
0
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
0
-400
33
History Matching
www.petroman.ir
MAXIMUM
SATURATION
ERROR I J K
----- -- -- -.000 2 4 1
.000 6 4 1
.000 6 4 1
.000 6 4 1
.000 6 4 1
.000 6 4 1
.000 6 4 1
.000 6 4 1
.000 6 4 1
.000 6 4 1
.000 6 4 1
.000 6 4 1
.000 6 4 1
.000 6 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
.000 2 4 1
TS
TS
TS
TS
TS
TS
TS
TS
TS
TS
112
113
114
115
116
117
118
119
120
121
3355
3385
3415
3445
3475
3505
3535
3565
3595
3623
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
27.50
3260
3262
3264
3266
3268
3270
3272
3274
3276
3278
150
150
150
150
150
150
150
150
150
150
209
209
209
210
210
210
210
210
211
211
2
2
2
2
2
2
2
2
2
2
0
0
0
0
0
0
0
0
0
0
TIMESTEP
122 AT
NAME I J K STAT
PI
---- -- -- -- ---- ----1
2 4 1 QO
3.4
2
6 4 1 QWI
3.4
FIELD TOTALS - PRODUCED:
INJECTED:
CONTROL
------QO
QWI
WELLS
----1
1
I=
1
2
3
J 4
5
6
7
1
3255.99
3254.16
3250.65
3246.88
3250.65
3254.16
3255.99
STEP
---TS 122
TIME
DAYS
---3650
* * * WELL RATES * * *
OIL
GAS
WATER
(B/D)
(MCF/D)
(B/D)
-------- -------- -------150.0
210.9
2.4
.0
.0
-400.0
150.0
.0
210.9
.0
AVERAGE SATURATIONS
OIL GAS
WATER
(FRC) (FRC) (FRC)
----- ----- ----.623 .046 .331
GOR
WCT
(SCF/B) (PCT)
------- ----1405.9
1.6
.0
.0
2.4
-400.0
1405.9
PAVG
PSI
---3280
.2E-06
.2E-06
.2E-06
.3E-06
.1E-06
.9E-07
.6E-06
.3E-06
.4E-06
.5E-06
.8E-06
.8E-06
.4E-06
.5E-06
.2E-06
.6E-06
.3E-06
.2E-06
.4E-06
.2E-06
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
2
2
2
2
2
2
2
2
2
2
4
4
4
4
4
4
4
4
4
4
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
0
0
0
0
0
0
0
0
0
0
1.6
PORE VOLUME
AVERAGED
PRESSURE
---------3279.5
CUMULATIVE PRODUCTION
OIL
GAS
WATER
MB
MMCF
MB
------- ------- ------548
781
9
274
405
-726
821
0
1185
0
SATURATIONS
SO SG SW
(FRC PORVOL)
--- --- --.64 .06 .30
.31 .03 .66
13
-730
2
2
PRESSURE (PSI)
TIME= 3650.00
2
3
4
5
6
3258.37
3263.29
3269.98
3277.48
3284.36
3256.37
3262.05
3269.69
3278.65
3287.94
3251.44
3259.37
3268.63
3281.93
3338.40
3239.89
3255.92
3267.37
3327.44
3480.76
3251.44
3259.37
3268.63
3281.93
3338.40
3256.37
3262.05
3269.69
3278.65
3287.94
3258.37
3263.29
3269.98
3277.49
3284.36
DELT
DAYS
-----27.50
-.1E-06
.3E-06
.2E-06
.2E-06
.2E-06
.4E-07
.3E-06
.3E-06
-.3E-07
.7E-07
3650.0 DAYS
REGION
PWF
(PSI)
-----3217.1
3520.3
-400
-400
-400
-400
-400
-400
-400
-400
-400
-400
FIELD PRODUCTION
OIL
GAS
WATER
B/D
MCF/D
B/D
------- ------- ------150
211
2
7
3288.60
3293.39
3306.47
3362.86
3306.47
3293.39
3288.60
FIELD INJECTION
GAS
WATER
MCF/D
B/D
------- ------0
-400
34
History Matching
www.petroman.ir
MAXIMUM
SATURATION
ERROR I J K
----- -- -- -.000 2 4 1
RESERVOIR SIMULATION
CLASS PROBLEM NO. 63 (SYNTHETIC HISTORY MATCHING)
I=1
J=1
2
3
4
5
6
7
Attached is the pressure versus time history of well C, which produced oil at constant rate
from the above reservoir. Well A displayed the same behavior. The reservoir is 2-D,
homogeneous, and isotropic.
The pressure versus time plot for history match # 1 is also given.
Analyze the pressure histories and recommend data changes.
35
History Matching
www.petroman.ir
36
History Matching
www.petroman.ir
63 Data
37
History Matching
www.petroman.ir
38
History Matching
www.petroman.ir
RESERVOIR SIMULATION
CLASS PROBLEM NO. 65 (SYNTHETIC HISTORY MATCHING)
I=1
10
11
12
13
14
15
J=1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Well GE-209, located in the center of the reservoir's oil zone (shown above), has been producing
at a constant rate for 1.5 years. The reservoir is homogeneous, isotropic, and effectively 2-D.
Observed pressure data, as well as the results of history match run #1, are attached.
39
History Matching
www.petroman.ir
PROBLEM NO. 65
H.M. #1
2500
2391
2335
2280
2252
2234
2221
2212
2204
2199
2194
2191
2189
2187
2185
2184
2183
Observed
2500
2298
2202
2114
2069
2041
2021
2007
1997
1990
1984
1981
1978
1976
1975
1975
1974
CELL PRESSURE,
psia
0.2
0.4
0.6
0.8
TIME, years
Observed
40
History Matching
www.petroman.ir
H.M. #1
1.2
1.4
1.6
/
SIMULATOR INPUT DATA
/
Problem 65
/
Run 1
/
SINGLE VALUE DATA
/
IMAX 15
/number of grids in X-direction
JMAX 15
/number of grids in Y-direction
PREF
CROC
CWAT
COIL
CRVO
VISW
2500
3.0E-6
3.3E-6
16.95E-6
62.162E-6
0.4
/PVT table
/
DX 12800 6400 3200 1600 800 400 400 400 400 400 800 1600 3200 6400 9000
DY 12800 6400 3200 1600 800 400 400 400 400 400 800 1600 3200 6400 9000
/
H
Por
Perm Elev
GRID
0 0 0 110 0.22 25
-3100
PO
0 0 1 2500
PSAT
0 0 0 1000
SW
0 0 0 1.0
SG
0 0 0 0.0
/
/...................................Input Sw in the oil region
/
WINDOW I = 7 TO 9, J = 5 TO 5, K = 1 TO 1
SW 0 0 0 0.24
WINDOW I = 6 TO 10, J = 6 TO 6, K = 1 TO 1
SW 0 0 0 0.24
WINDOW I = 5 TO 11, J = 7 TO 9, K = 1 TO 1
SW 0 0 0 0.24
WINDOW I = 6 TO 10, J = 10 TO 10, K = 1 TO 1
SW 0 0 0 0.24
WINDOW I = 7 TO 9, J = 11 TO 11, K = 1 TO 1
SW 0 0 0 0.24
END
/
/
SCHEDULE DATA
/
NAME 1
I = 8 J = 8 K =1 PI = 3.375
QO 1
100.0
PLIM 1
1000.0
WLRT 1 PSRT 1
DSTAR
0.05
DSLIM
0.05
OMEGA
1.1
DELT
10
DTMAX
10
TIME 18.2625 / .05 years
TIME 36.525 / .10 years
TIME 73.05 / .20 years
TIME 109.575 / .30 years
TIME 146.1
/ .40 years
TIME 182.625 / .50 years
TIME 219.15 / .60 years
TIME 255.675 / .70 years
42
History Matching
www.petroman.ir
TIME 292.2
TIME 328.725
TIME 365.25
TIME 401.775
TIME 438.3
TIME 474.825
TIME 511.35
TIME 547.875
/ .80 years
/ .90 years
/ 1.00 years
/ 1.10 years
/ 1.20 years
/ 1.30 years
/ 1.40 years
/ 1.50 years
43
History Matching
www.petroman.ir
Problem No. 65
History Match Run 1 - Output
BO
(RB/STB)
-------1.130
1.210
1.150
BG
(RCF/SCF)
--------.1700
.0170
.0040
OIL VISC
(CP)
-------17.000
17.000
17.000
GAS VISC
(CP)
-------.020
.060
.100
RS
(SCF/STB)
--------150.0
350.0
350.0
KRG
-------.0000
1.0000
KROG
-------1.0000
.0000
PCOG
--------.000
.000
KRW
-------.0000
.0000
.0000
.0700
.1300
.1900
.2540
.2700
1.0000
NUMBER OF GRIDS
REGION ACTIVE INACTIVE
------ ------ -------TOTAL
225
0
FIELD PRODUCTION
TIME
DELT
STEP DAYS
DAYS
---- ---- -----TS
1
9
9.13
KROW
PCOW
---------------1.0000
.000
1.0000
.000
1.0000
.000
.1900
.000
.1000
.000
.0200
.000
.0000
.000
.0000
.000
.0000
.000
*****
INITIAL FLUID IN PLACE
OIL
GAS
WATER
(MMSTB)
(MMMSCF)
(MMSTB)
---------------------------21.7627
7.61750
9902.27
FIELD INJECTION
MATERIAL BALANCE ERROR
PAVG
OIL
GAS
WATER
GAS
PSI
B/D
MCF/D
B/D
MCF/D
---- ------- ------- ------- ------2500
100
35
0
0
................................................ TIMESTEP
................................................
**** CURRENT FLUID IN PLACE ****
REGION
OIL
GAS
(MMSTB)
(BSCF)
------ ------------ -----------TOTAL
21.7609
7.61687
NAME I J K STAT
PI
---- -- -- -- ---- ----1
8 8 1 QO
3.4
PWF
(PSI)
-----1742.8
STEP
---TS
2
TS
3
WELLS
----1
TIME
DAYS
---18
27
DELT
DAYS
-----9.13
9.13
*****
SOLVENT
(MMMSCF)
--------.000000
PORE
VOLUME
(MMRCF)
----------55293.1
AVERAGE SATURATIONS
OIL
GAS
WATER SOLVENT
(FRC) (FRC) (FRC)
(FRC)
----- ----- --------.003 .000
.997
.000
MAXIMUM
WATER
OIL
GAS
WATER
B/D
FRC
FRC
FRC
------- ------- ------- ------0 -.3E-06 -.4E-06 -.2E-09
2 AT
SATURATION
ERROR I J K
----- -- -- -.000 8 8 1
18.3 DAYS
PORE
AVERAGE SATURATIONS
PORE VOLUME
WATER
VOLUME
OIL GAS
WATER
AVERAGED
(MMSTB)
(MMRCF)
(FRC) (FRC) (FRC)
PRESSURE
------------ --------------- ----- -------------9902.27
55293.1
.003 .000 .997
2500.0
* * * WELL RATES * * *
OIL
GAS
WATER
(B/D)
(MCF/D)
(B/D)
-------- -------- -------100.0
35.0
.4
100.0
.0
35.0
.0
.4
.0
GOR
WCT
(SCF/B) (PCT)
------- ----350.0
.4
350.0
.4
CUMULATIVE PRODUCTION
OIL
GAS
WATER
MB
MMCF
MB
------- ------- ------2
1
0
2
0
1
0
SATURATIONS
SO SG SW
(FRC PORVOL)
--- --- --.76 .00 .24
0
0
1
1
FIELD PRODUCTION
FIELD INJECTION
MATERIAL BALANCE ERROR
MAXIMUM
PAVG
OIL
GAS
WATER
GAS
WATER
OIL
GAS
WATER
SATURATION
PSI
B/D
MCF/D
B/D
MCF/D
B/D
FRC
FRC
FRC
ERROR I J K
---- ------- ------- ------- ------- ------- ------- ------- ------- ----- -- -- -2500
100
35
0
0
0 -.6E-07 .5E-07 -.4E-09
.000 8 8 1
2500
100
35
0
0
0 -.3E-06 -.2E-06 -.4E-09
.000 8 8 1
44
History Matching
www.petroman.ir
................................................ TIMESTEP
................................................
4 AT
36.5 DAYS
NAME I J K STAT
PI
---- -- -- -- ---- ----1
8 8 1 QO
3.4
PWF
(PSI)
-----1688.5
STEP
---TS
4
TS
5
TS
6
TS
7
100.0
.0
WELLS
----1
TIME
DAYS
---37
47
57
65
DELT
DAYS
-----9.13
10.00
10.00
8.26
* * * WELL RATES * * *
OIL
GAS
WATER
(B/D)
(MCF/D)
(B/D)
-------- -------- -------100.0
35.0
.6
35.0
.0
PAVG
PSI
---2500
2500
2500
2500
GOR
WCT
(SCF/B) (PCT)
------- ----350.0
.5
.6
.0
350.0
CUMULATIVE PRODUCTION
OIL
GAS
WATER
MB
MMCF
MB
------- ------- ------4
1
0
.5
4
0
FIELD PRODUCTION
OIL
GAS
WATER
B/D
MCF/D
B/D
------- ------- ------100
35
1
100
35
1
100
35
1
100
35
1
FIELD INJECTION
GAS
WATER
MCF/D
B/D
------- ------0
0
0
0
0
0
0
0
8 AT
AVERAGE SATURATIONS
OIL GAS
WATER
(FRC) (FRC) (FRC)
----- ----- ----.003 .000 .997
REGION
PWF
(PSI)
-----1635.5
STEP
---TS
8
TS
9
TS 10
TS 11
WELLS
----1
TIME
DAYS
---73
83
93
101
DELT
DAYS
-----8.26
10.00
10.00
8.26
* * * WELL RATES * * *
OIL
GAS
WATER
(B/D)
(MCF/D)
(B/D)
-------- -------- -------100.0
35.0
.7
100.0
.0
35.0
.0
PAVG
PSI
---2500
2500
2500
2500
12 AT
CONTROL
------QO
WELLS
----1
TIME
DELT
100.0
.0
PORE VOLUME
AVERAGED
PRESSURE
---------2499.9
CUMULATIVE PRODUCTION
OIL
GAS
WATER
MB
MMCF
MB
------- ------- ------7
3
0
.7
7
0
3
0
35.0
.0
SATURATIONS
SO SG SW
(FRC PORVOL)
--- --- --.76 .00 .24
0
0
.7
.0
MAXIMUM
SATURATION
ERROR I J K
----- -- -- -.000 8 8 1
.000 8 8 1
.000 8 8 1
.000 8 8 1
GOR
WCT
(SCF/B) (PCT)
------- ----350.0
.7
350.0
.7
PORE VOLUME
AVERAGED
PRESSURE
---------2499.9
CUMULATIVE PRODUCTION
OIL
GAS
WATER
MB
MMCF
MB
------- ------- ------11
4
0
11
0
4
0
SATURATIONS
SO SG SW
(FRC PORVOL)
--- --- --.76 .00 .24
0
0
1
1
FIELD PRODUCTION
OIL
GAS
WATER
109.6 DAYS
AVERAGE SATURATIONS
OIL GAS
WATER
(FRC) (FRC) (FRC)
----- ----- ----.003 .000 .997
* * * WELL RATES * * *
OIL
GAS
WATER
(B/D)
(MCF/D)
(B/D)
-------- -------- -------100.0
35.0
.7
PAVG
350.0
FIELD INJECTION
GAS
WATER
MCF/D
B/D
------- ------0
0
0
0
0
0
0
0
REGION
1
1
................................................ TIMESTEP
................................................
NAME I J K STAT
PI
---- -- -- -- ---- ----1
8 8 1 QO
3.4
MAXIMUM
SATURATION
ERROR I J K
----- -- -- -.000 8 8 1
.000 8 8 1
.000 8 8 1
.000 8 8 1
73.1 DAYS
GOR
WCT
(SCF/B) (PCT)
------- ----350.0
.7
.7
.0
FIELD PRODUCTION
OIL
GAS
WATER
B/D
MCF/D
B/D
------- ------- ------100
35
1
100
35
1
100
35
1
100
35
1
PWF
(PSI)
-----1608.4
0
0
1
1
................................................ TIMESTEP
................................................
NAME I J K STAT
PI
---- -- -- -- ---- ----1
8 8 1 QO
3.4
1
0
SATURATIONS
SO SG SW
(FRC PORVOL)
--- --- --.76 .00 .24
FIELD INJECTION
GAS
WATER
45
History Matching
www.petroman.ir
MAXIMUM
SATURATION
STEP
---TS 12
TS 13
TS 14
TS 15
DAYS
---110
120
130
138
DAYS
-----8.26
10.00
10.00
8.26
PSI
---2500
2500
2500
2500
B/D
------100
100
100
100
MCF/D
------35
35
35
35
B/D
------1
1
1
1
MCF/D
------0
0
0
0
................................................ TIMESTEP
................................................
16 AT
* * * WELL RATES * * *
OIL
GAS
WATER
(B/D)
(MCF/D)
(B/D)
-------- -------- -------100.0
35.0
.7
STEP
---TS 16
TS 17
TS 18
TS 19
100.0
.0
WELLS
----1
TIME
DAYS
---146
156
166
174
FIELD
DELT
DAYS
-----8.26
10.00
10.00
8.26
35.0
.0
CONTROL
------QO
20 AT
STEP
---TS 20
TS 21
TS 22
TS 23
TIME
DAYS
---183
193
203
211
DELT
DAYS
-----8.26
10.00
10.00
8.26
35.0
.0
PAVG
PSI
---2500
2500
2500
2500
5
0
350.0
24 AT
SATURATION
ERROR I J K
----- -- -- -.000 4 3 1
.000 5 2 1
.000 8 8 1
.000 7 7 1
PORE VOLUME
AVERAGED
PRESSURE
---------2499.8
CUMULATIVE PRODUCTION
OIL
GAS
WATER
MB
MMCF
MB
------- ------- ------18
6
0
.7
18
0
6
0
35.0
SATURATIONS
SO SG SW
(FRC PORVOL)
--- --- --.76 .00 .24
0
0
.7
MAXIMUM
SATURATION
ERROR I J K
----- -- -- -.000 2 2 1
.000 4 1 1
.000 13 8 1
.000 4 1 1
219.2 DAYS
AVERAGE SATURATIONS
OIL GAS
WATER
(FRC) (FRC) (FRC)
----- ----- ----.003 .000 .997
* * * WELL RATES * * *
OIL
GAS
WATER
(B/D)
(MCF/D)
(B/D)
-------- -------- -------100.0
35.0
.7
100.0
0
0
182.6 DAYS
FIELD INJECTION
GAS
WATER
MCF/D
B/D
------- ------0
0
0
0
0
0
0
0
REGION
15
0
SATURATIONS
SO SG SW
(FRC PORVOL)
--- --- --.76 .00 .24
1
1
................................................ TIMESTEP
................................................
NAME I J K STAT
PI
---- -- -- -- ---- ----1
8 8 1 QO
3.4
.7
GOR
WCT
(SCF/B) (PCT)
------- ----350.0
.7
.7
.0
FIELD PRODUCTION
OIL
GAS
WATER
B/D
MCF/D
B/D
------- ------- ------100
35
1
100
35
1
100
35
1
100
35
1
PWF
(PSI)
-----1568.9
CUMULATIVE PRODUCTION
OIL
GAS
WATER
MB
MMCF
MB
------- ------- ------15
5
0
AVERAGE SATURATIONS
OIL GAS
WATER
(FRC) (FRC) (FRC)
----- ----- ----.003 .000 .997
* * * WELL RATES * * *
OIL
GAS
WATER
(B/D)
(MCF/D)
(B/D)
-------- -------- -------100.0
35.0
.7
100.0
.0
WELLS
----1
350.0
PORE VOLUME
AVERAGED
PRESSURE
---------2499.9
FIELD INJECTION
MATERIAL BALANCE ERROR
MAXIMUM
GAS
WATER
GAS
WATER
OIL
GAS
WATER
MCF/D
B/D
MCF/D
B/D
FRC
FRC
FRC
------- ------- ------- ------- ------- ------- ------35
1
0
0
.8E-08 .1E-07 -.1E-06
35
1
0
0
.8E-07 -.3E-06 .2E-06
35
1
0
0 -.5E-07 .2E-06 -.7E-09
35
1
0
0 -.6E-07 .1E-06 -.6E-09
1
1
REGION
NAME I J K STAT
PI
---- -- -- -- ---- ----1
8 8 1 QO
3.4
ERROR I J K
----- -- -- -.000 8 8 1
.000 8 8 1
.000 8 8 1
.000 8 8 1
146.1 DAYS
GOR
WCT
(SCF/B) (PCT)
------- ----350.0
.7
.7
.0
................................................ TIMESTEP
................................................
PWF
(PSI)
-----1578.3
FRC
FRC
FRC
------- ------- -------.5E-06 -.1E-06 -.6E-09
.1E-07 .2E-06 -.7E-09
.1E-06 -.3E-06 .7E-07
.7E-07 .3E-06 .7E-07
AVERAGE SATURATIONS
OIL GAS
WATER
(FRC) (FRC) (FRC)
----- ----- ----.003 .000 .997
REGION
NAME I J K STAT
PI
---- -- -- -- ---- ----1
8 8 1 QO
3.4
B/D
------0
0
0
0
GOR
WCT
(SCF/B) (PCT)
------- ----350.0
.7
350.0
46
History Matching
www.petroman.ir
.7
PORE VOLUME
AVERAGED
PRESSURE
---------2499.8
CUMULATIVE PRODUCTION
OIL
GAS
WATER
MB
MMCF
MB
------- ------- ------22
8
0
22
SATURATIONS
SO SG SW
(FRC PORVOL)
--- --- --.76 .00 .24
INJECTED:
CONTROL
------QO
STEP
---TS 24
TS 25
TS 26
TS 27
.0
WELLS
----1
TIME
DAYS
---219
229
239
247
DELT
DAYS
-----8.26
10.00
10.00
8.26
.0
PAVG
PSI
---2500
2500
2500
2500
.0
FIELD PRODUCTION
OIL
GAS
WATER
B/D
MCF/D
B/D
------- ------- ------100
35
1
100
35
1
100
35
1
100
35
1
FIELD INJECTION
GAS
WATER
MCF/D
B/D
------- ------0
0
0
0
0
0
0
0
................................................ TIMESTEP
................................................
28 AT
NAME I J K STAT
PI
---- -- -- -- ---- ----1
8 8 1 QO
3.4
FIELD TOTALS - PRODUCED:
INJECTED:
CONTROL
------QO
STEP
---TS 28
TS 29
TS 30
TS 31
WELLS
----1
TIME
DAYS
---256
266
276
284
DELT
DAYS
-----8.26
10.00
10.00
8.26
* * * WELL RATES * * *
OIL
GAS
WATER
(B/D)
(MCF/D)
(B/D)
-------- -------- -------100.0
35.0
.7
100.0
.0
35.0
.0
PAVG
PSI
---2500
2500
2500
2500
32 AT
CONTROL
------QO
STEP
---TS 32
TS 33
TS 34
TS 35
WELLS
----1
TIME
DAYS
---292
302
312
320
DELT
DAYS
-----8.26
10.00
10.00
8.26
100.0
.0
35.0
.0
26
0
9
0
SATURATIONS
SO SG SW
(FRC PORVOL)
--- --- --.76 .00 .24
0
0
MAXIMUM
SATURATION
ERROR I J K
----- -- -- -.000 15 13 1
.000 3 1 1
.000 2 1 1
.000 15 2 1
292.2 DAYS
350.0
PORE VOLUME
AVERAGED
PRESSURE
---------2499.7
CUMULATIVE PRODUCTION
OIL
GAS
WATER
MB
MMCF
MB
------- ------- ------29
10
0
.7
29
0
10
0
SATURATIONS
SO SG SW
(FRC PORVOL)
--- --- --.76 .00 .24
0
0
1
1
FIELD PRODUCTION
OIL
GAS
WATER
B/D
MCF/D
B/D
------- ------- ------100
35
1
100
35
1
100
35
1
100
35
1
.7
GOR
WCT
(SCF/B) (PCT)
------- ----350.0
.7
.7
.0
FIELD INJECTION
GAS
WATER
MCF/D
B/D
------- ------0
0
0
0
0
0
0
0
................................................ TIMESTEP
................................................
REGION
PORE VOLUME
AVERAGED
PRESSURE
---------2499.7
CUMULATIVE PRODUCTION
OIL
GAS
WATER
MB
MMCF
MB
------- ------- ------26
9
0
AVERAGE SATURATIONS
OIL GAS
WATER
(FRC) (FRC) (FRC)
----- ----- ----.003 .000 .997
* * * WELL RATES * * *
OIL
GAS
WATER
(B/D)
(MCF/D)
(B/D)
-------- -------- -------100.0
35.0
.7
PAVG
PSI
---2500
2500
2500
2500
350.0
FIELD INJECTION
GAS
WATER
MCF/D
B/D
------- ------0
0
0
0
0
0
0
0
REGION
MAXIMUM
SATURATION
ERROR I J K
----- -- -- -.000 3 1 1
.000 13 1 1
.000 11 1 1
.000 3 1 1
1
1
FIELD PRODUCTION
OIL
GAS
WATER
B/D
MCF/D
B/D
------- ------- ------100
35
1
100
35
1
100
35
1
100
35
1
PWF
(PSI)
-----1556.3
255.7 DAYS
GOR
WCT
(SCF/B) (PCT)
------- ----350.0
.7
.7
.0
................................................ TIMESTEP
................................................
NAME I J K STAT
PI
---- -- -- -- ---- ----1
8 8 1 QO
3.4
AVERAGE SATURATIONS
OIL GAS
WATER
(FRC) (FRC) (FRC)
----- ----- ----.003 .000 .997
REGION
PWF
(PSI)
-----1561.8
1
1
PORE
VOLUME
(MMRCF)
36 AT
328.7 DAYS
AVERAGE SATURATIONS
OIL GAS
WATER
(FRC) (FRC) (FRC)
47
History Matching
www.petroman.ir
PORE VOLUME
AVERAGED
PRESSURE
MAXIMUM
SATURATION
ERROR I J K
----- -- -- -.000 2 1 1
.000 3 1 1
.000 15 7 1
.000 7 1 1
NAME I J K STAT
PI
---- -- -- -- ---- ----1
8 8 1 QO
3.4
PWF
(PSI)
-----1552.1
STEP
---TS 36
TS 37
TS 38
TS 39
DELT
DAYS
-----8.26
10.00
10.00
8.26
35.0
.0
PAVG
PSI
---2500
2500
2500
2500
* * * WELL RATES * * *
OIL
GAS
WATER
(B/D)
(MCF/D)
(B/D)
-------- -------- -------100.0
35.0
.7
100.0
.0
WELLS
----1
TIME
DAYS
---329
339
349
357
------------ ----------9902.27
55293.0
.7
.0
FIELD PRODUCTION
OIL
GAS
WATER
B/D
MCF/D
B/D
------- ------- ------100
35
1
100
35
1
100
35
1
100
35
1
FIELD INJECTION
GAS
WATER
MCF/D
B/D
------- ------0
0
0
0
0
0
0
0
40 AT
STEP
---TS 40
TS 41
TS 42
TS 43
WELLS
----1
TIME
DAYS
---365
375
385
394
DELT
DAYS
-----8.26
10.00
10.00
8.26
* * * WELL RATES * * *
OIL
GAS
WATER
(B/D)
(MCF/D)
(B/D)
-------- -------- -------100.0
35.0
.7
100.0
.0
35.0
.0
PAVG
PSI
---2500
2500
2499
2499
CONTROL
------QO
STEP
---TS 44
TS 45
WELLS
----1
TIME
DAYS
---402
412
DELT
DAYS
-----8.26
10.00
MAXIMUM
SATURATION
ERROR I J K
----- -- -- -.000 13 12 1
.000 15 9 1
.000 4 2 1
.000 9 1 1
PORE VOLUME
AVERAGED
PRESSURE
---------2499.5
CUMULATIVE PRODUCTION
OIL
GAS
WATER
MB
MMCF
MB
------- ------- ------37
13
0
.6
37
0
13
0
35.0
.0
SATURATIONS
SO SG SW
(FRC PORVOL)
--- --- --.76 .00 .24
0
0
.6
.0
MAXIMUM
SATURATION
ERROR I J K
----- -- -- -.000 12 2 1
.000 8 3 1
.000 7 2 1
.000 5 2 1
401.8 DAYS
AVERAGE SATURATIONS
OIL GAS
WATER
(FRC) (FRC) (FRC)
----- ----- ----.003 .000 .997
* * * WELL RATES * * *
OIL
GAS
WATER
(B/D)
(MCF/D)
(B/D)
-------- -------- -------100.0
35.0
.6
PAVG
PSI
---2499
2499
350.0
44 AT
100.0
.0
0
0
365.3 DAYS
FIELD INJECTION
GAS
WATER
MCF/D
B/D
------- ------0
0
0
0
0
0
0
0
REGION
12
0
SATURATIONS
SO SG SW
(FRC PORVOL)
--- --- --.76 .00 .24
1
1
................................................ TIMESTEP
................................................
NAME I J K STAT
PI
---- -- -- -- ---- ----1
8 8 1 QO
3.4
33
0
GOR
WCT
(SCF/B) (PCT)
------- ----350.0
.6
.7
.0
FIELD PRODUCTION
OIL
GAS
WATER
B/D
MCF/D
B/D
------- ------- ------100
35
1
100
35
1
100
35
1
100
35
1
PWF
(PSI)
-----1546.5
.7
AVERAGE SATURATIONS
OIL GAS
WATER
(FRC) (FRC) (FRC)
----- ----- ----.003 .000 .997
REGION
PWF
(PSI)
-----1549.0
350.0
CUMULATIVE PRODUCTION
OIL
GAS
WATER
MB
MMCF
MB
------- ------- ------33
12
0
1
1
................................................ TIMESTEP
................................................
NAME I J K STAT
PI
---- -- -- -- ---- ----1
8 8 1 QO
3.4
GOR
WCT
(SCF/B) (PCT)
------- ----350.0
.7
---------2499.6
GOR
WCT
(SCF/B) (PCT)
------- ----350.0
.6
350.0
.6
PORE VOLUME
AVERAGED
PRESSURE
---------2499.5
CUMULATIVE PRODUCTION
OIL
GAS
WATER
MB
MMCF
MB
------- ------- ------40
14
0
40
0
14
0
SATURATIONS
SO SG SW
(FRC PORVOL)
--- --- --.76 .00 .24
0
0
1
1
FIELD PRODUCTION
OIL
GAS
WATER
B/D
MCF/D
B/D
------- ------- ------100
35
1
100
35
1
FIELD INJECTION
GAS
WATER
MCF/D
B/D
------- ------0
0
0
0
48
History Matching
www.petroman.ir
MAXIMUM
SATURATION
ERROR I J K
----- -- -- -.000 4 2 1
.000 15 2 1
TS
TS
46
47
422
430
10.00
8.26
2499
2499
100
100
35
35
1
1
0
0
................................................ TIMESTEP
................................................
48 AT
NAME I J K STAT
PI
---- -- -- -- ---- ----1
8 8 1 QO
3.4
FIELD TOTALS - PRODUCED:
INJECTED:
CONTROL
------QO
STEP
---TS 48
TS 49
TS 50
TS 51
WELLS
----1
TIME
DAYS
---438
448
458
467
DELT
DAYS
-----8.26
10.00
10.00
8.26
* * * WELL RATES * * *
OIL
GAS
WATER
(B/D)
(MCF/D)
(B/D)
-------- -------- -------100.0
35.0
.6
100.0
.0
35.0
.0
PAVG
PSI
---2499
2499
2499
2499
52 AT
CONTROL
------QO
STEP
---TS 52
TS 53
TS 54
TS 55
TIME
DAYS
---475
485
495
503
DELT
DAYS
-----8.26
10.00
10.00
8.26
35.0
.0
PAVG
PSI
---2499
2499
2499
2499
CONTROL
WELLS
350.0
56 AT
35.0
.0
44
0
15
0
0
0
SATURATIONS
SO SG SW
(FRC PORVOL)
--- --- --.76 .00 .24
0
0
MAXIMUM
SATURATION
ERROR I J K
----- -- -- -.000 1 1 1
.000 1 1 1
.000 6 1 1
.000 15 1 1
PORE VOLUME
AVERAGED
PRESSURE
---------2499.3
CUMULATIVE PRODUCTION
OIL
GAS
WATER
MB
MMCF
MB
------- ------- ------47
17
0
.6
47
0
17
0
SATURATIONS
SO SG SW
(FRC PORVOL)
--- --- --.76 .00 .24
0
0
.6
.0
MAXIMUM
SATURATION
ERROR I J K
----- -- -- -.000 5 1 1
.000 6 1 1
.000 11 1 1
.000 15 5 1
511.3 DAYS
AVERAGE SATURATIONS
OIL GAS
WATER
(FRC) (FRC) (FRC)
----- ----- ----.003 .000 .997
* * * WELL RATES * * *
OIL
GAS
WATER
(B/D)
(MCF/D)
(B/D)
-------- -------- -------100.0
35.0
.6
100.0
.0
1
1
474.8 DAYS
FIELD INJECTION
GAS
WATER
MCF/D
B/D
------- ------0
0
0
0
0
0
0
0
REGION
1
1
1
1
FIELD PRODUCTION
OIL
GAS
WATER
B/D
MCF/D
B/D
------- ------- ------100
35
1
100
35
1
100
35
1
100
35
1
PWF
(PSI)
-----1542.2
.6
GOR
WCT
(SCF/B) (PCT)
------- ----350.0
.6
.6
.0
................................................ TIMESTEP
................................................
NAME I J K STAT
PI
---- -- -- -- ---- ----1
8 8 1 QO
3.4
1
1
PORE VOLUME
AVERAGED
PRESSURE
---------2499.4
CUMULATIVE PRODUCTION
OIL
GAS
WATER
MB
MMCF
MB
------- ------- ------44
15
0
AVERAGE SATURATIONS
OIL GAS
WATER
(FRC) (FRC) (FRC)
----- ----- ----.003 .000 .997
* * * WELL RATES * * *
OIL
GAS
WATER
(B/D)
(MCF/D)
(B/D)
-------- -------- -------100.0
35.0
.6
100.0
.0
WELLS
----1
350.0
FIELD INJECTION
GAS
WATER
MCF/D
B/D
------- ------0
0
0
0
0
0
0
0
REGION
2
1
1
1
................................................ TIMESTEP
................................................
NAME I J K STAT
PI
---- -- -- -- ---- ----1
8 8 1 QO
3.4
.000 15
.000 1
438.3 DAYS
GOR
WCT
(SCF/B) (PCT)
------- ----350.0
.6
.6
.0
FIELD PRODUCTION
OIL
GAS
WATER
B/D
MCF/D
B/D
------- ------- ------100
35
1
100
35
1
100
35
1
100
35
1
PWF
(PSI)
-----1543.3
AVERAGE SATURATIONS
OIL GAS
WATER
(FRC) (FRC) (FRC)
----- ----- ----.003 .000 .997
REGION
PWF
(PSI)
-----1544.7
0
0
GOR
WCT
(SCF/B) (PCT)
------- ----350.0
.6
350.0
49
History Matching
www.petroman.ir
.6
PORE VOLUME
AVERAGED
PRESSURE
---------2499.3
CUMULATIVE PRODUCTION
OIL
GAS
WATER
MB
MMCF
MB
------- ------- ------51
18
0
51
0
18
0
0
0
SATURATIONS
SO SG SW
(FRC PORVOL)
--- --- --.76 .00 .24
------QO
STEP
---TS 56
TS 57
TS 58
TS 59
----1
TIME
DAYS
---511
521
531
540
DELT
DAYS
-----8.26
10.00
10.00
8.26
PAVG
PSI
---2499
2499
2499
2499
FIELD PRODUCTION
OIL
GAS
WATER
B/D
MCF/D
B/D
------- ------- ------100
35
1
100
35
1
100
35
1
100
35
1
FIELD INJECTION
GAS
WATER
MCF/D
B/D
------- ------0
0
0
0
0
0
0
0
................................................ TIMESTEP
................................................
60 AT
STEP
---TS 60
WELLS
----1
TIME
DAYS
---548
DELT
DAYS
-----8.26
* * * WELL RATES * * *
OIL
GAS
WATER
(B/D)
(MCF/D)
(B/D)
-------- -------- -------100.0
35.0
.6
100.0
.0
35.0
.0
PAVG
PSI
---2499
.6
.0
MAXIMUM
SATURATION
ERROR I J K
----- -- -- -.000 5 1 1
.000 15 7 1
.000 4 1 1
.000 5 1 1
547.9 DAYS
AVERAGE SATURATIONS
OIL GAS
WATER
(FRC) (FRC) (FRC)
----- ----- ----.003 .000 .997
REGION
NAME I J K STAT
PI
---- -- -- -- ---- ----1
8 8 1 QO
3.4
GOR
WCT
(SCF/B) (PCT)
------- ----350.0
.6
350.0
.6
PORE VOLUME
AVERAGED
PRESSURE
---------2499.2
CUMULATIVE PRODUCTION
OIL
GAS
WATER
MB
MMCF
MB
------- ------- ------55
19
0
55
0
19
0
SATURATIONS
SO SG SW
(FRC PORVOL)
--- --- --.76 .00 .24
0
0
1
1
FIELD PRODUCTION
OIL
GAS
WATER
B/D
MCF/D
B/D
------- ------- ------100
35
1
FIELD INJECTION
GAS
WATER
MCF/D
B/D
------- ------0
0
50
History Matching
www.petroman.ir
MAXIMUM
SATURATION
ERROR I J K
----- -- -- -.000 3 1 1
RESERVOIR SIMULATION
CLASS PROBLEM NO. 66 (SYNTHETIC HISTORY MATCHING)
I = 1
5
DX = DY = H = 400
The above homogeneous, isotropic, 1-D, oil zone is attached to an aquifer of uncertain
properties and dimensions. Geological work has shown that there are five possible aquifer
structures. They are represented by Reservoirs A through E in Figure 1.
The reservoir is produced at a variable rate as shown by figure 2. Figure 3 shows the
production cell pressure response versus time for six years of production for the base case (no
aquifer). Also shown in Figure 3 is a pressure plot for a six year simulation run of the oil zone
with an aquifer in communication.
Reservoirs A through E (Figure 1) are the same oil zone, but with aquifers of
different sizes and configurations attached. The data set for the base case and grid modifications
for each of the aquifers are given in Tables A and B respectively.
Analyze Figure 2. Decide which aquifer case (A-E) is plotted, then make a computer run
to confirm your choice.
51
History Matching
www.petroman.ir
BASE RESERVOIR
RESERVOIR A
IMAX = 6
RESERVOIR B
IMAX = 13
RESERVOIR C
IMAX = 8
RESERVOIR D
IMAX = 11
RESERVOIR E
IMAX = 11
52
History Matching
www.petroman.ir
qo (STB/D)
600
500
400
300
200
100
0
0
3
Time (yrs)
53
History Matching
www.petroman.ir
TABLE A
/
/
/
/
/
IMAX
CROC
CWAT
COIL
CRVO
VISW
DENO
DENG
DENW
METH
2
/ D-4 solution method
STONE = T
/Method for 3-phase Kr calculation
NOMAPS
NOTABS
END
/
/
.TABLE SECTION
/
KRW
/
Sw
Krw
Krow Pcow
0.0
0.0
1.00 0.0
0.20 0.06 0.7
0.0
0.25 0.016 0.465 0.0
0.30 0.026 0.283 0.0
0.35 0.039 0.187 0.0
/Water-Oil Kr table
0.40 0.055 0.127 0.0
0.45 0.073 0.085 0.0
0.50 0.095 0.060 0.0
0.55 0.121 0.040 0.0
0.60 0.153 0.027 0.0
0.65 0.198 0.018 0.0
0.70 0.254 0.010 0.0
0.75 0.325 0.007 0.0
0.80 0.420 0.0
0.0
1.00 1.000 0.0
0.0
KRG
/
Sg
Krg
Krog Pcog
0.0
0.0
1.0
0.0
1.00 0.0
1.0
0.0
PRES
/
/
P
psi
14.7
264.7
514.7
1014.7
Bo
rb/stb
1.0620
1.1500
1.2070
1.2950
Bg
rcf/scf
0.935829
0.067902
0.035229
0.018007
Vo
cp
1.040
0.9750
0.9100
0.8300
Vg
Rs
cp
scf/stb
0.0080
1.1
0.00960
90.4
0.01120 180.2
0.01400 371.1
2014.7
2514.7
3014.7
4014.7
9014.7
5014.7
9014.7
1.4350
1.5000
1.5650
1.6950
1.5790
1.8270
2.3570
0.009063
0.007266
0.006064
0.004554
0.002167
0.003644
0.002167
0.6950
0.6410
0.5940
0.5100
0.7400
0.4490
0.2030
0.01890
0.02080
0.02280
0.02680
0.47000
0.03090
0.04700
54
History Matching
www.petroman.ir
636.1
774.8
929.8
1270.0
1270.0
1618.0
2984.0
END
/
/
GRID DATA
/
WINDOW I=1 TO 5 J = 1 TO 1
ELEV
0
0
0
H
0
0
0
POR
0
0
0
WSAT
0
0
0
GSAT
0
0
0
PSAT
0
0
0
PO
0
0
0
PERM
0
0
0
DX
5*400
DY
1*400
END
/
SCHEDULE DATA
OMEGA=1.0
WLRT
2
PSRT
2
DELT=10
DTCUT=0.5
DTFAC=2.0
DTMIN=1
SGMAX=0.04
SWMAX=0.04
DOUT=365
DTMAX=60
WELL=1
I=1
J=1
M=1
QO=700
PI=10.59
TIME 365
M=1
QO=250
TIME 730
M=1
QO=50
TIME 1095
M=1
QO=700
TIME 1460
M=1
QO=400
TIME 1642
M=1
QO=200
TIME 1825
M=1
QO=50
TIME 2190
END
K = 1 TO 1
.7200
400
.25
.20
.0
1000
5000
10
K=1
55
History Matching
www.petroman.ir
TABLE B
MODIFICATIONS FOR DIFFERENT CASES
PROBLEM NO. 66
CELL
DI
1500
POR
SOI
1500
0.1
CASE A:
CASE B:
6
7
8
9
10
11
12
13
1
1
1
1
1
1
1
1
1500
1500
1500
1500
1500
1500
1500
1500
0
0
0
0
0
0
0
0
0.1
0.1
0.1
600
700
800
0
0
0
6
7
8
9
10
11
1000
0.5
1000
1000
1000
1000
600
700
800
900
1000
1100
0
0
0
0
0
0
6
7
8
9
10
11
1
1000
1000
1000
1000
1000
600
700
800
900
1000
1100
0
0
0
0
0
0
CASE C:
6
7
8
600
700
800
CASE D:
CASE E:
NOTE:
56
History Matching
www.petroman.ir
REFERENCES
1. Mattax, C.C. and Dalton, R.L.: Reservoir Simulation, SPE Monograph 13, Richardson, TX (1990)
87-98.
2. Peaceman, D.W.: "Interpretation of Well-Block Pressures in Numerical Reservoir Simulation,"
SPEJ (June 1978) 183-194; Trans., AIME, 265.
3. van Poollen, H.K., Breitenbach, E.A., and Thurnau, D.H.: "Treatment of Individual
Wellsand Grids in Reservoir Modeling," SPEJ (Dec. 1968) 341-46.
4. Slater, G.E. and Durrer, E..: "Adjustment of Reservoir Simulator Models to Match
Field
Performance," SPEJ (Sept. 1971) 295-305; Trans., AIME, 251.
5. Carter, R.D. et al.: "Performance Matching With Constraints," SPEJ (April 1974) 187-91; Trans.,
AIME, 257.
6. Veatch, R.W. Jr. and Thomas, G.W.: "A Direct Approach for History Matching," paper SPE
3515 presented at the 1971 Annual Meeting, New Orleans, Oct. 3-6.
7. Coats, K.H. and Dempsey, J.R.: "A New Approach for Determining Reservoir Description from
Field Performance SPEJ (March 1970) 66-74; Trans., AIME, 249.
8. Bradford, R.N. and Hrkel, E.J.: "Evaluation of a Reservoir Simulation Study: West Civit
Dykeman Sand Unit," JPT (Dec. 1979) 1599-1604.
9. Brummett, W.M. et al.: "Reservoir Description by Simulation at SACROC- A Case
History," JPT (Oct. 1976) 1241-1255.
57
History Matching
www.petroman.ir
CHAPTER 4
Coning
Introduction
Analysis of Forces
Coning Simulation
Addington's Method
References
Exercises
Appendix
www.petroman.ir
Coning
Introduction
Water coning. Many wells produce from oil zones underlain by "bottom water". When the well is
produced, water moves up toward the wellbore in a cone shape. At certain conditions, water breaks
through into the well and concurrent oil and water production begins. The phenomena is referred to
as 'water coning'.
The phenomenon of coning was studied analytically by Muskat and several key factors were
determined. However this analysis was for steady state and also assumed with no multi-phase flow
or saturation gradients occur. The main result of the analysis was that a critical rate exists above
which the cone reaches the perforations and the water or gas is produced. Below this critical rate,
the cone reaches a certain height depending on the rate and other parameters.
Some of the factors affecting coning are listed. Coning increases with:
(a) increasing w/o (=M)
(b) increasing kv/kh (or continuity)
(c) increasing q
(d) decreasing hw
(e) decreasing ()/144
(f) decreasing (kh)
(g) increasing Pc
Gas
Oil
Oil
W ater
Coning
www.petroman.ir
water
coning
h cone
Oil
h
Water
qcritical
2
Coning
www.petroman.ir
Analysis of forces
hap
144
(psi) - 1
(1)
(psi)
Coning condition:
(p)gravity = (/144)hap
(2)
Estimation of (p)viscous :
0.00708 k h
qB =
h
( p )viscous
o
(3)
r
ln e + Spartial penetration
rw
p / 144
g
Gas
h ap
Gas
Z
h
OIL
Bubble
/ 144
p
h
viscous
bp
Fig. 5 - Schematic
Fig. 6 - Z vs p plot
3
Coning
www.petroman.ir
Gravity Number
or,
N grav
q o Bo o ln e + Spp
rw
144
=
q w
0.00633 A k avg
(4)
(5)
from these, the higher the flow rate the lower the gravity number.
Coning conditions
Ngrav < 1
(q o)c =
+ Spp
Bo o ln
rw
4
Coning
www.petroman.ir
(6)
qc =
qc =
qc =
(d) Schols 7
re
( / 144) k o ( h2 - h2p )
qc =
0.432 +
re
887.89 o Bo
ln
rw
qc =
5(Analytical)
6(Analytical&Experimental)
o Bo
0.14
(Experimental)
7.0915 x 10 -3 h 2 ( / 144)k o
( r D e , , )
Bo o
0.0142 h2 k h ( / 144)
qc =
q cD
qc =
2
k o ( / 144) h p
1-
75.15 Bo o h
( h t )2.238[ln r e]1.99
h2 ( / 144) k h
qc =
q cD
282.41 Bo o
5
Coning
www.petroman.ir
Non-Steady-State:
(a) Addington13
Coning simulation
qc =
2
kh h p hap
1
(137.9)2
kv
Bo
kh
01
.
(Simulation)
F1 F2
A special type of simulation is required to model the coning of water or gas into a producing
well. The phenomenon is only local to the vicinity of the wellbore and the simulation must be
concentrated around the wellbore area. For this region a radial geometry must be used.
The reservoir simulation of coning behavior will predict the changes with time and give a
complete performance prediction for a well. The purpose of a coning study is usually to predict the
performance of a well after matching history. In some reservoir studies, the purpose of a coning
simulation is to study the completion of the well and try to determine the effects of various
workover, stimulation, and recompletion plans. Coning simulation has been performed to set field
rules for conservation in some cases.
Another common use of coning simulation is to develop pseudo-relative permeability curves
for use in field scale 2-D or 3-D models. Since the field scale model cannot predict accurately
coning behavior, this behavior must be studied separately in radial coordinates and the results of the
coning study must then be put into the field scale model in the form of pseudo-relative permeabilities
or other calculation procedure.
The purposes of coning study are to match history & describe the reservoir, predict, study
workovers, set field rules, and develop kr curves for 2-D/3-D models.
A special type of reservoir simulator is required for coning simulation. It is necessary to use
"implicit saturation" calculations in order to make practical simulation runs. We will describe the
reason for this.
A typical simulation grid has small grids near the wellbore. These small grids are required to
model the rapidly changing saturations and flow rates near the wellbore. An evenly spaced
logarithmic spacing is usually used which represents the pressure gradient as a straight line with the
logarithm of radius, at least for an ideal problem. This improves the accuracy of the calculations.
Care should be taken not to take the first cell unduly small because this is the cell which has the
highest velocities and is the most likely to be unstable.
We note that the velocity of the fluid flow is the greatest near the wellbore. But the pore
volumes of the cells have the smallest values near the wellbore. this results in an extremely high
through-put of fluids through a cell near the wellbore during a timestep. It is not unusual to have
several hundreds of pore volumes of throughput in a cell during a timestep. the IMPES method tends
to be very unstable with these high throughput rates. this instability results in oscillating solution
values such as saturations. This oscillation usually leads to the solution "blowing up" and the
solution becomes meaningless.
6
Coning
www.petroman.ir
10
20
rw
log r
Fig. 7 - r vs Z plot
IMPES k
r
Sw
Fig. 9 - t vs Sw plot
Implicit
simulators.
The solution to the stability problem is to treat relative permeability and capillary pressure
implicitly. That is, to solve the timesteps with these values evaluated at the end of the timestep rather
than at the beginning of the timestep. there are a number of published methods of performing the
implicit calculations. The fundamentals of most of the methods involves extrapolating the old
relative permeability to a new value by using a "chord slope" of the relative permeability curves
multiplied times the saturation change.
The typical results from a coning simulation will be plots such as water/oil ratio versus time.
If two or more rates are used for the simulation, it is important to re-plot the results on a cumulative
production basis. If the water/oil ratio is the same for two rate cases when plotted on a cumulative oil
production basis, then we would say that the cases are not rate sensitive. the higher rate does not
damage the water/oil ratio performance of the well.
7
Coning
www.petroman.ir
Implicit methods.
(8)
WOR
WOR
q
Not Rate
Sensitive
Np
An example of the application of a coning simulator can be found in an actual coning study
that was performed. The reservoir was producing large amounts of water and it was important to
determine whether a workover or recompletion was desirable.
In this case the water/oil ratio was not accurately known nor was the stratification of the
reservoir known, at least away from the wellbore. Whether cross-flow was predominant or whether
the layers were isolated was also not known. The stimulation runs were made with these uncertain
variables varied for different cases. This is called a "sensitivity analysis". A number of simulation
runs were made varying the rate, and completion interval along with different values of the uncertain
reservoir parameters. The result of this sensitivity analysis was that none of the variables had much
effect on the water/oil ratio performance within the limits that were investigated. Therefore, it was
decided that a reduction in rate or a workover was not called for in this case.
Several rules can be applied to coning simulation. In the radial direction the spacing should
be evenly spaced logarithmically near the wellbore. This the same as having each cell radius equal to
the previous radius times a factor. For example, values of 2, 4, 8 would be an example of geometric
spacing with a factor of 2.0. Near the outer boundary of the radial direction, the spacing should be
more evenly spaced on a coordinate basis, because the pressure gradients tend to flatten rather than
8
Coning
www.petroman.ir
being a semi-log straight with radius. The total number of grids in the radial direction is normally
around eight to ten.
The vertical direction is very important in a coning simulation. Normally a well is stratified
with layers that are roughly normal to the wellbore, so the vertical flow is similar to serial flow
going through various layers of transmissibility. The vertical grid spacing should be chosen in a
manner which describes the stratification of the well accurately. If the permeability profile of the
wellbore can be represented by a "stair step" such as in the diagram in the sketch, then the vertical
grids should coincide with the stair steps. If an impermeable break is identified, it should be located
between cells, rather than in the middle of the cell.
2
Serial flow
p
in
p
out
Because of the layered nature of a typical reservoir, horizontal flow tends to be parallel along
the layers and the average horizontal permeability should include all of the layers that are contained
within a cell.
The vertical flow in the layered reservoir tends to be serial flow through the layers. The
correct vertical permeability for flow between vertical cells in a 3 -D model or radial model should
be a harmonic average of the reservoir layers between the mid-points of adjacent simulator cells.
9
Coning
www.petroman.ir
3-D or RADIAL
Horizontal:
kh h
kh =
h
(9)
(arithmetic)
Vertical:
kv =
h
h
( )
kw
(harmonic)
(10)
10
Coning
www.petroman.ir
Addington's Method
This method shows how to obtain a correlation to predict gas coning behavior based on a
radial numerical model which can be used in a 3-D coarse simulator. The correlation allows to
predict not only the critical coning rate but also the GOR of a well after coning has been achieved.
Purpose - To develop a method to use in a 3-D coarse simulator.
log GOR
h ap BT
h ap
(b) Do regression with hBT, m for model parameters (qt, hp, kh, kv/kh, etc.).
(c) Correlate hBT, m with these (model parameters).
11
Coning
www.petroman.ir
BT
BT
2
1
P1= qt(kv/kh)0.1of1f2/(kh*hp1/2)
(11)
(12)
F2= (Acres/160)0.1
(13)
P2= qt(kv/kh)0.5oF1F3/(kh*hp1/2)
(14)
F3= (Acres/160)0.5
(15)
Sg Vp = gas volume
(15)
k=1
14
Sg = 1 - Swc
(16)
Sg = 1 - Swc - Sor
(17)
Sg = Sgc
(18)
12
Coning
www.petroman.ir
(19)
t
GOR
Consider an oil field (reservoir) overlain by a large gas cap. Each well is capable of gas
coning if produced above its critical rate. Abopve each wells critical rate, its GOR is rate dependent.
Now, suppose a fixed oil rate, (qo)max, is produced from the field. An optimal allocation of oil rates to
individual wells would minimize the total field gas rate. The problem can be stated as:
Objective function:
minimize
Constrain :
(q g ) i
(21)
(q o ) i = (q o ) max
(22)
13
Coning
www.petroman.ir
Now suppose a fixed gas rate, (qg)max, is produced from the same field. In this case, an
optimal allocatation of gas rates to individual wells would maximize total field oil rates. This
problem can be stated as:
Objective function:
maximize
Constrain :
(q o ) i
(23)
(q g ) i = (q g ) max
(24)
The maximum field rate for oil might be a pipeline capacity, and the maximum field rate for
gas might be a plant processing capacity.
Optimization creterion
i,j = 1,....,n
(25)
It should be pointed out that (dqg/dqo)i is a derivative and not a GOR. This derivative shows
the change in the gas rate with an incremental change in the oil rate and changes with rate under gas
coning conditions.
This optimization algorithm can optimize actual field rates using well test data and can be
used in a general 3-dimensional reservoir simulator to optimize well rates during simulation.
qo max
GOR
qg max
14
Coning
www.petroman.ir
(26)
Horizontal wells:
qc =
k v k h k ' ro Lh / 144
q cD
where,
qc = critical coning rate, stb/D
kh = horizontal permeability, md
k ro = oil relative permeability at Swc
o = oil viscosity, cp
= water-oil gravity difference, psi/ft
h = initial oil formation thickness, ft
qcD = dimensionless critical coning rate
L = length of horizontal well, ft
15
Coning
www.petroman.ir
(27)
For a tank reservoir, the hwb correlation can be used to calculate water breakthrough time for
vertical and horizontal wells.
t bt =
( N p ) bt
(28)
qt
For constant rate cases, WOR after breakthrough can be predicted from
Vertical wells:
hbp h wb
(29)
Horizontal wells:
16
Coning
www.petroman.ir
hbp h wb
(30)
REFERENCES
1. Addington, D.V., "An approach to Gas-Coning Correlations for a Large Grid Cell Reservoir
Simulator" JPT November 1981.
2. Yang W., Wattenbarger R.A.,"Water Coning Calculations for Vertical and Horizontal Wells"
SPE 22931
3. Urbanczyk, Christopher H., Wattenbarger, R.A., "Optimization of Well Rates under Gas Coning
Conditions" SPE 21677
17
Coning
www.petroman.ir
EXERCISES
RESERVOIR SIMULATION
CLASS PROBLEM NO. 41 (CONING FUNCTIONS)
Use the Addington method and correlation to do the following:
a. Calculate qo critical
b. Draw GOR curve for this rate
c. Calculate GOR at 2 x qo critical
d. Draw the GOR vs hap curves for both rates
Data:
kv / kh = 0.1
o
= 5.0 cp
kh
= 1200 md-feet
ht
= 500 feet
hbp = 0
hp
= 60 feet
= 850 scf/stb
= 1.42 rb/stb
18
Coning
www.petroman.ir
RESERVOIR SIMULATION
CLASS PROBLEM NO. 95
Find kh and kv for each layer form the following core analysis.
layer
h
1
1
1
1
kh
50
10
20
100
kv
25
5
10
50
1
1
1
1
1
200
300
50
20
10
100
150
25
10
5
1
1
1
1
5
6
8
10
2.5
3
4
5
1
1
1
1
20
100
500
10
10
50
250
5
1
1
1
1
50
60
80
90
25
30
40
45
19
Coning
www.petroman.ir
APPENDIX
SPE 22931 "Water Coning Calculations for Vertical and Horizontal Wells"
SPE 21677 "Optimization of Well Rates under Gas Coning Conditions"
20
Coning
www.petroman.ir
CHAPTER 5
Compositional Simulation
Introduction
Applications
PVT Behavior
Equations for Compositional Simulation
Regression of EOS Parameters
Modeling of Surface Facilities
Lab Tests
Gas Cycling in Volatile Oil and Gas Condensate
Reservoirs
Vaporizing Gas Drive (VGD)
Field Examples
Fundamentals
Heavy Fraction Characterization
Pseudoization
Regrouping Scheme of Heavy Components
Compositional Simulation
Heavy Ends Characterization
Default Fluid Predictions
Pre - Regression
Component Pseudoization
Regression
Initialization Data for VIP
Run Data for VIP
Data Files
PVT Matching
www.petroman.ir
Fundamentals
Heavy Fraction Characterization
Pseudoization
Regrouping Scheme of Heavy Components
Compositional Simulation
Heavy Ends Characterization
Default Fluid Predictions
Pre - regression
Component Pseudoization
Regression
Initialization Data For VIP
Run Data For VI
Data Files
Heavy.dat
Heavy.eos
Default.dat
Regression.dat
Pseudo.dat
Pseudo.eos
Final.dat
Final.eos
Spe3i.dat
Spe3r.dat
1
Compositional Simulation
www.petroman.ir
Compositional Simulation
Introduction
In compositional simulation of oil and gas reservoirs equation-of-state (EOS) methods are
seeing increasing usage over more traditional K-value methods for phase equilibrium calculations. It
has been found by several authors that equations-of-state are, in general, not able to accurately
predict reservoir fluid behavior using theoretical EOS parameters. It has been found that "tuning"
the EOS (by modifying the EOS parameters) is required to adequately match laboratory derived
PVT data.
In this chapter, we describe how an equation-of-state is "tuned" to match laboratory data
using nonlinear regression. We present partial justification for modification of EOS parameters and
then discuss some practical guidelines for efficient and meaningful fluid characterization. Although
we will mention component pseudoization (lumping) since it is usually part of the fluid
characterization process, the emphasis of this section will be on regression and not on pseudoization.
Most of the material presented in this text was obtained from Coats and Smart.1
1
Compositional Simulation
www.petroman.ir
Applications
A compositional simulator is used in a variety of situations in which a black oil simulator
does not adequately describe the fluid behavior. Each of the applications listed requires a special
phase behavior computation and may require special laboratory analysis to determine the behavior
of the fluids. Some applications are as follows:
A)
B)
C)
D)
E)
2
Compositional Simulation
www.petroman.ir
PVT Behavior
The PVT behavior of the reservoir fluids must be well understood in order to do a competent
job of compositional simulation. We will briefly review some of the fundamentals. At any
particular time, a mixture of fluids may be in a vapor-liquid equilibrium at a particular temperature
and pressure. For most reservoirs the temperature can be considered to be constant. The variation of
the equilibrium with pressure and also with composition is also important. When the vapor and
liquid phases are in equilibrium, the molecules of each component are condensing and vaporizing at
the same rate as shown in Fig. 1. The equilibrium can be described by x, y, and z values for all the
components. A K-value can be calculated for the equilibrium for each component.
equilibrium
Vapor
Liquid
yi
xi
Fig. 1
Prediction of PVT Behavior. The K-values have different patterns that can be observed. It is
common to see a log-log plot of K vs. p which shows a family of curves tending to converge to a Kvalue of 1.0 at a "convergence pressure" as in Fig. 2. The behavior of this family of curves depends
on the composition as well as on temperature and pressure. For example, the amount of intermediate
hydrocarbons has a significant effect on convergence pressure.
3
Compositional Simulation
www.petroman.ir
c2
c1
c3
1.0
Log k
c5
Leaner Mixture
cn
Lower c2 - c5
Log pressure
Fig. 2
Several methods are commonly used in compositional simulators to model the phase
behavior. K-value methods can be used with a systematic method of modifying convergence
pressures with compositions. Equation of state methods are also in common usage and have the
advantage of predicting densities as well as the molecular equilibrium.
Much research has been directed toward predicting phase behavior given temperature,
pressure and composition. However, for reservoir simulation studies laboratory composition data is
almost always available for the mixtures expected to occur in the reservoir. For that reason, it is not
as important to predict fluid behavior because the computational procedure must be modified to
match the laboratory data. All of the phase behavior methods use various parameters that can be
adjusted to accomplish the match of the laboratory data. It should be noted that a simulator using an
equation of state method can run as much as ten times slower than a simulator using the K-value
methods. This loss of computer efficiency tends to offset the advantages of the equation of state
methods. Some simulation engineers use equation of state methods to match laboratory data and
then convert to K-value methods with independent density calculation methods for the reservoir
simulation runs.
PVT Prediction Methods. The methods used for PVT prediction are as follows:
1) K-value, pk (convergence pressure) method.(This method is faster and can be matched with
lab data and EOS , but it needs correction for the density of oil and the density of gas)
2) Equation of State method (EOS).( It matches more accurate with the lab data and we can get
the density of oil and gas)
a) Peng-Robinson
b) Redlich-Kwong
4
Compositional Simulation
www.petroman.ir
5
Compositional Simulation
www.petroman.ir
r
r
( y$ i g u g )+ ( x$ i o u o )= [ ( S o o x$ i + S g g y$ i )]
t
vapor flow + liquid flow
(1)
storage
S o ,g
(2)
1 .0
o , g , o , g = f(p, T, x i , y i , z i )
(3)
where a and b are characteristics of the particular gas, R is the universal gas constant, and VM is the
molar volume.
Peng - Robinson Equation. Peng and Robinson later proposed the following:
[p +
V M (V M
aT
]( V M - b) = RT
+ b) + b( V M - b)
(5)
aT = ac
(6)
R2 T 2C
AC = 0.45724
(7)
PC
1/ 2
1/ 2 = 1 + m(1 - T r )
b = 0.07780
(8)
RT c
pc
(9)
(10)
[p +
aT
]( V M - b) = RT
V M ( V M + b)
a c = 0.42747
R Tc
pc
(11)
(12)
(13)
RT c
b = 0.08664
pc
7
Compositional Simulation
www.petroman.ir
(14)
= -( log pvr + 1)
(15)
8
Compositional Simulation
www.petroman.ir
F =
W | ( d - d ) /
j
jC
dj
j=1
(16)
where djC and dj are calculated and observed values of observation j, respectively, and W
is a weighting factor.
Using the regression algorithm above and the guidelines contained in this text, it has been
found that agreement between laboratory data and regressed EOS results is generally good to
excellent.
Justification for Altering Theoretical EOS Parameters. Thus far, no justification has been given
for alteration of the theoretical values of EOS parameters. In the following sections we present
partial justification for the modification of EOS parameters for methane and the plus fraction.
Methane Fraction. The theoretical a and b values in both the PR and SRK EOS are based on the
van der Waals conditions of p/V=2p/V2=0 at the critical point. The component temperature
functions in the SRK and PR EOS essentially reflect satisfaction of pure-component density and
vapor-pressure data below critical temperature. At reservoir conditions, methane is usually well
above its critical point, where there is no theory or clear-cut guide to selection or alteration of 's.
Since the theoretical values are based on p and T conditions far removed from the p and T
conditions of interest, one could argue that the 's for methane should be determined from matching
of laboratory-derived PVT data in the p and T ranges of interest.
Plus Fraction. The primary argument for altering EOS parameters of the plus fraction is that the
plus fraction, unlike all other components, is a mixture of many components.
General Procedure for Fluid Characterization. In this section, we describe the procedure
commonly used to characterize a particular fluid for use in compositional simulation. Computer
programs (such as Integrated Technologies' EOS-PAK) are available commercially which can
9
Compositional Simulation
www.petroman.ir
facilitate greatly the computations required in the following steps. Some or all of the following steps
may be employed in a typical fluid characterization.
1. Decide if and how to split the plus fraction into extended fractions. The actual splitting is usually
done using the method presented by Whitson5. There is some debate as to whether the mole
fractions of the extended fractions should be roughly equal or decreasing (e.g., exponentially) with
increasing molecular weight.
2. Pseudoize (lump) the components into a fewer number of pseudo-components. This is done
primarily for economic reasons (fewer components results in faster run times and lower costs). The
focus of this section is not on pseudoization; however, there are references available which offer
some guidelines for optimal pseudoization.6
3. Match the laboratory expansion data using a nonlinear regression algorithm. The regression data
set to be matched may include results from multiple fluid samples and results from tests run at
different temperatures. In general, the regression data set should include all measured laboratory
data. Any calculated data that are reported as part of a fluid analysis should not be included. The
regression variable set should be chosen such that it possesses the characteristics of an optimal
regression variable set.
4. Match the laboratory viscosity data using a nonlinear regression algorithm. The regression data
set should include all measured viscosity data. The regression variable set should include only the
critical z-factors (which affect viscosity only and do not affect the results of expansion tests).
5. At this point the EOS fluid characterization is complete, as all laboratory measured data have
been matched. However, additional adjustments to the fluid data may be required before
compositional simulation can be started. For example, it may be necessary to alter the initial fluid
composition to match the initial saturation pressure or gas-oil ratio observed in the field. In addition,
it may be necessary to determine the optimum separator conditions (only for predictive studies; for
history matching, the actual separator conditions should be used).
Practical Considerations for Minimizing Splitting of C7+ Fraction. Coats and Smart found that
extensive splitting of the C7+ fraction is not necessary to match laboratory data, in contrast to
previous authors. In general, none to four splits of the C7+ fraction are all that are required. A
general, a priori guide to the extent of C7+ splitting required can be obtained from the experimentally
observed range of C7+ molecular weights during an expansion or multiple-contact test. The greater
the range of C7+ molecular weights, the more C7+ fractions are required.
In addition, the degree of splitting is also affected by the anticipated recovery process. For
example, gas cycling processes will require more splitting than depletion/waterflooding processes.
Minimizing Number of Regression Variables. Usually only five parameters - a and b for
methane and the heaviest fraction and the methane/heaviest binary interaction coefficient - are
necessary and sufficient for good matches of laboratory data. For the purposes of this discussion, the
"heaviest" fraction is the heaviest single fraction used in the EOS calculations. For example, if the
10
Compositional Simulation
www.petroman.ir
original plus fraction, C7+, is split into three components, say F7, F8, and F9, then F9 is the new
heaviest fraction as defined above.
A general rule for selecting regression variables is to exclude any EOS parameter that, by
inspection, cannot affect significantly the calculated value of any of the regression data. For
example, if the fluid composition in a regression data set has a very small amount of a particular
component, then it would not make sense to include that particular component's 's or binaries as
regression variables. Conversely, it makes sense to include the EOS parameters of a component that
is predominate (such as methane) in the fluid compositions.
Regression of Surface Separation Data. Coats and Smart found that regressed EOS surface
separation results were the same regardless of whether surface separation data were included in the
regression data set. In other words, an EOS tuned with only data at reservoir conditions will, in
general, adequately predict surface separation data.
They did not address the opposite situation, i.e. whether including the surface separation data
in the regression data set could harm the match of data at reservoir conditions. They implied that
there was nothing wrong with including the separation data, but that it was not necessary. Of course,
this applies only if the surface separation data are measured in the laboratory. Surface separation
data presented in standard fluid analysis reports are commonly calculated (using K-values) and not
measured. In general, calculated data (such as surface separation data) should not be included in a
regression data set; only measured data should be included.
Characteristics of an Optimal Regression Variable Set. The characteristics of an optimal
regression variable set are:
1. The regression converges,
2. The values that the variables converge upon are realistic (reasonably close to theoretical values),
3. Deletion of any member of the variable set results in either or both of (a) significantly worse data
match and (b) unrealistic variable values, and
4. Addition of any other EOS parameter results in either or both of (a) nonconvergence and (b)
insignificantly better data matches.
Nonconvergence is dependent on both the variable set and the regression data set. It may be
possible to solve a nonconvergence problem by adding more data to the regression data set. For a
fixed regression data set, the general remedy for nonconvergence is to simply remove one of the
regression variables.
The general remedy for convergence to unrealistic variable values is to add another variable
to the regression variable set. For example, although negative values of binary interaction
coefficients may result in a good match of lab data, negative values are not reasonable. It may be
possible, by adding another regression variable, to obtain an equally good match of the lab data with
more reasonable positive interaction coefficients.
11
Compositional Simulation
www.petroman.ir
From personal experience, it has also been determined that (1) convergence and (2) realistic
variable values are also affected by the weighting factors assigned to the various regression data
items. It may be possible to fix problems with either or both of these characteristics by changing the
weighting factors assigned to certain data items (e.g. assigning lower weighting factors to data that
are suspect).
If a poor data match results from a regression that satisfies the above criteria for an optimal
regression variable set, then this indicates either erroneous data or inadequacy of the EOS to match
the particular fluid data (it may be difficult to distinguish between the two). It is sometimes possible
to detect erroneous data by simple mass balances on CVD data and/or surface separation data.
Example Applications. Examples of EOS regression to match laboratory data are presented by
Coats and Smart1 and the third SPE comparative solution project, SPE 12278.7
12
Compositional Simulation
www.petroman.ir
Residue Gas
Plant
etc.
GAS
Compressor
Make-up Gas
F3 = 0.95
F 6 = 1.00
Prod.
1
S.T
Oil Sales
Stage Separation
INJ.
Fig. 3
The producing fluid is separated with stage separation to improve the oil recovery. The
separator calculation must be performed at the separator temperature and pressure given. The
separator liquid goes to the stock tank and is sold as oil.
The separator gas then continues to a gas liquids plant which strips the gas liquids from the
input stream. This is modeled by specifying a fraction of each component which is recovered as gas
liquids.
The remaining residue gas may be re-injected into the reservoir after compression to a
suitable injection pressure. It may be necessary to add make-up gas to the gas injection in order to
compensate for shrinkage after liquid removal.
The simulation of the surface facilities as well as the reservoir performance makes this a very
interesting project. The optimization of the operations might include runs in which the plant
configuration is modified as well as runs with the usual injection and production parameters
modified.
13
Compositional Simulation
www.petroman.ir
Lab Tests
Certain types of laboratory tests are usually performed for compositional problems. Before
performing compositional simulation, it is necessary to simulate laboratory tests with whatever
phase behavior and fluid property calculations which are going to be used in the simulation. We will
review some of the basic laboratory tests that are performed for gas condensate reservoirs.
Constant Composition Expansion Test. The first test is constant composition expansion,
sometimes called flash liberation. Refer to Fig. 4.
GAS CONDENSATE
Constant Composition Expansion
4500
4300
3500
dew pt.
3000
2500
1000
pressure
2-phase
Vol.
% Liq.
Vliq
pressure
pressure
Fig . 4
The PVT cell is expanded from above the dew point to pressures usually near the expected
abandonment pressure. Below the dew point pressure, the total volume of the two phase mixture is
measured. Sometimes the liquid volume is measured and sometimes not.
The two phase expansion curve must be matched by the simulator. Two phase
compressibility can be calculated directly from this volumetric data. This compressibility value
would be at any measured pressure for the combined liquid and gas compressibilities, at least for the
composition used.
The liquid percent curve is presented if the liquid volumes are measured in the laboratory. It
is common for the liquid to form below the dew point and then vanish again at lower pressures. This
14
Compositional Simulation
www.petroman.ir
would result in no residual oil saturation if this could be accomplished in the reservoir, but this test is
only applicable for constant composition.
Fig. 5 shows a phase diagram for a constant composition fluid. For this fluid, the nature of
constant composition expansion depends on the temperature of the fluid. This figure shows a bubble
point fluid at low temperatures and a fairly dry condensate which vaporizes completely during
pressure decreases at higher temperatures.
3
2
crit.
pt.
1. black oil
2. volatile oil
3. gas condensate
(high intermediate
4. gas condendsate
5. gas
100%
80%
press.
60%
40%
20%
0%
temperature
Fig. 5
Fig. 6 presents a family of curves which represent the pressure traverses of the various
temperatures shown in Fig. 5. For the first two temperatures, bubble points are observed. The
second temperature has greater shrinkage since it is closer to the critical point. At temperatures
above the critical temperature, the dew point elevates with increases in temperature and then
decreases with further increases in temperature. At the highest temperature, the liquid completely
vaporizes with pressure reduction.
15
Compositional Simulation
www.petroman.ir
b.p.
b.p.
100
T1
T2
% Liq.
T3
T4
T
0
d.p.
d.p.
d.p.
Pressure
Fig. 6
Behavior of the constant composition expansion is useful in understanding reservoir fluid
behavior. It is important to match this behavior with the simulator. However, the composition of the
reservoir fluid during depletion usually changes dramatically when two phase flow occurs.
Constant Volume Depletion Test. Fig. 7 depicts a constant volume depletion test (or simply,
depletion test). This test is normally done for a dew point fluid. This is a test which approximates
the constant reservoir volume fluid expansion.
GAS CONDENSATE:
constant volume depletion
Gp
Gp
4500
4300
dew point
3800
yi
yi
3000
3000
3800
Fig. 7
For this case, gas is withdrawn at pressures below the dew point, representing gas production
from the reservoir. Liquid is not taken out of the PVT cell because liquid usually does not flow in
the reservoir. Rather than "bleeding off" the reservoir gas, the gas production is taken out of the
PVT cell at discrete intervals. By performing the test in this manner the exact composition of the
equilibrium gas is known at the pressure at which the gas is withdrawn. This composition is
measured from the produced gas at that particular pressure. The liquid composition cannot be
16
Compositional Simulation
www.petroman.ir
measured without destroying the sample, but it can be calculated from a material balance.
Experimental errors affect the accuracy of the material balance, however.
The standard results of this test are direct measurements which are represented by three
sketches shown in Fig. 8.
cumm. gas
below D.P.
c1
c2
% Liq.
log y
c3
Gp
c4
c5
Pressure
Pressure
Pressure
Fig. 8
The liquid percent curve can be modified to show oil saturation by accounting for the
connate water saturation. The composition of the producing gas stream is presented and can be used
to calculate the gas-oil ratio or gas-liquid ratio in the producing stream during depletion. And
finally, the cumulative gas production shows how pressure declines with gas production. These
three curves must be matched with the simulator to validate the phase behavior and fluid property
calculations. The liquid density is not measured directly, nor is the gas density. The validation of
the simulator includes the density calculations as well as the phase equilibrium calculations.
The surface separation and calculation of the gas-oil ratio must also be calculated by the
simulator during the simulation runs. This information is usually reported in the depletion
laboratory test results. These calculations are relatively simple because NGAA K-values can usually
be used at surface conditions.
17
Compositional Simulation
www.petroman.ir
Displacement Process. The injected lean gas miscibility displaces the in-situ rich gas toward the
producing well. The liquid generally does not move due to unfavorable mobility ratios. Also the oil
saturation can drop below the residual oil saturation. However, the oil vaporizes behind the
displacement front. The intermediates in the oil vaporize due to the contact with the injected lean
gas. This vaporization process "strips" the reservoir liquid of its intermediates which are then
transported in the gas phase to the producer. This miscible process is referred to as a multiple contact
miscible process (also called developed miscibility) as opposed to a first contact miscible process. If
the reservoir fluid is considered as being composed of three pseudo components (C ), intermediates
(C2-6) and heavies (C7+), a ternary diagram can be used to represent the reservoir fluid and the
displacement process at a given temperature and pressure. Two fluids are considered first contact
miscible if all possible mixtures of the two yield a single phase fluid at a given pressure and
temperature. If the path connecting the injected gas composition and the reservoir oil composition
does not pass through the two phase region, the process is termed first contact miscible. This path is
known as the dilution path and represents the composition changes as the injected gas displaces
reservoir oil. In doing so, we are neglecting the pressure change that is, of course, an essential
ingredient in making the fluids flow in the reservoir. Ternary diagrams represent constant pressure
and temperature conditions, but illustrate the mechanics of a miscible displacement process.
1
injected. The dry gas miscibly displaces the rich gas toward the producing wells. The liquid
normally does not move because it is below the residual oil saturation. However, the oil does
vaporize behind the displacement front. The lighter ends of the oil vaporize because of the contact
with the dry injection gas instead of the rich reservoir gas that it previously contacted. As this
vaporization continues, the oil saturation decreases and a significant amount of the oil components
are transported to the producing wells in the gas phase.
GAS CYCLING
dry gas
production
new equilibrium
vaporization of intermediates
displacement front
dry
rich
G
O
Fig. 9
This gas cycling process is modeled with a compositional simulator. The displacement
mechanism is modeled simultaneously with the phase behavior.
19
Compositional Simulation
www.petroman.ir
Field Examples
Three field simulation studies are summarized in this section. All three reservoirs are retrograde
reservoirs. In each case, gas cycling was used for a variety of reasons. However the underlying
mechanism of displacement is the vaporizing gas drive as described earlier.
Carson Creek Field, Alberta. The Carson Creek field is a retrograde gas condensate field of
Central Alberta, discovered in 1957. The field encompasses about 18,000 acres and has two gas
pools, "A" and "B". The original gas-in-place in the two pools was about 400 BSCF. Methane
constituted 70.2 mole percent of the original reservoir fluid. Gas cycling operations began in 1962;
original gas cycling operations consisted of processing between 70 and 90 MMSCF/D raw gas to
extract about 4,000 STB/D condensate and re-injecting all residue gases into the reservoir. The
cycling also processed 5 to 10 MMSCF/D solution gas from an adjacent field and was re-injected.
Gas cycling was employed in this field for both pressure maintenance and to carry out a miscible
flood.
A compositional simulator was used to simulate gas cycling in this field. The model uses Kvalues as a function of temperature, pressure and composition to model phase equilibrium. Density,
viscosity and compressibility of the fluids were treated as functions of composition and pressure.
History matches of production, injection and pressure in each well were obtained. Also, a match of
the pentanes-plus composition of each produced well stream was obtained.
Bonnie Glen D 3-A Pool Model, Alberta. The Bonnie Glen D 3-A pool, is one of the most prolific
fields in Canada. The original oil in place is estimated at 6.57 MMSTB, with an initial gas cap of
44.5 MMSCF. Since simultaneous production of gas cap during the life of the oil column could be
detrimental to oil recovery, gas cap production would normally be deferred until depletion of the oil
column. Gas cycling can however, can be carried out concurrently with continued depletion of oil
column with a beneficial effect on the overall recovery of hydrocarbon liquids. The increase in
recovery results from vaporization of retrograde liquids in the gas cap plus vaporization of lighter
components from the original oil column.
A three dimensional three phase black oil simulation study was first conducted to history
match the reservoir performance. This was followed by conversion to a compositional formulation.
The compositional formulation used K-value correlations to model phase equilibria.
Kaybob South Field , Alberta. The Kaybob South Field is one of the largest oil fields in Canada.
Initial gas in place in the field was estimated at 4,000 BSCF. The reservoir encompasses an area of
57,000 acres and is totally underlain by water (bottom water drive). Because of the possibility of
significant retrograde liquid losses upon depletion, it was decided to implement gas cycling in the
field. Reservoir simulation was used to design the gas cycling project and study its potential impact.
The design and planning of the cycling project began in 1967, without full delineation of the
reservoir limits. Several cycling patterns were simulated (involving different injection well patterns).
The effect of different cycling rates was also studied by simulating various cases. Studies were
undertaken to determine the degree of revaporization expected from cycling with residue gas.
21
Compositional Simulation
www.petroman.ir
Table 1
Reservoir Fluid Characteristics for
Volatile Oils and Retrograde Condensates
Fluid Classification
Volatile Oils
Greenish to orange
42-45
Gas Condensates
Yellowish
45-60
1200-3500
8000-30000
1.7-3.0
6.0-20.0
150-300
150-300
3000-7500
(Bubble Point)
22
Compositional Simulation
www.petroman.ir
1500-9000
(Dew Point)
PVT MATCHING
Fundamentals:
Heavy Fraction Characterization. During the development of the application of EOS's to naturally
occuring hydrocarbon mixtures, it has become clear that insufficient description of heavier
hydrocarbons (e.g., heptanes and heavier) reduces the accuracy of PVT predictions. Volatile oil and
gas-condensate volumetric phase behavior is particularly sensitive to composition and properties of
the heaviest components. Therefore we have to develop a comprehensive method to characterize
compositional variation, which we call "molar distribution". A three-parameter gamma function was
chosen for describing molar distribution.
(x - h )a-1 exp [(x - h) / b]
p(x) =
a
b (a)
(1)
p(x) is called the probability density function (or three parameter gamma function). Whereas x
corresponds to measured C7+ molecular weight. For a = 1, the distribution is exponential. Values less
than one give accelerated exponential distribution, while values greater than one yield left-skewed
distributions. Note that as a approaches infinity, the distribution becomes normal, though "folded"
at h, the minimum molecular weight included in the C7+ fraction.
a, b, and h are parameters defining the distribution. h can be estimated accurately since it
represents the minimum molecular weight to be included in the fraction. If a is given, b is found
directly from h, a, and the measured C7+ molecular weight, M7+. a can be fitted to measured molarand weight-distribution data, or estimated using an empirical relation.
The parameter h is defined as the minimum molecular weight expected to occur in the Cn+
fraction. That is, there is zero probability [p(x) = 0] for occurrence of compounds with molecular
weight less than h. If the C7+ is considered, then h = 92 (the molecular weight of toluene) would be a
good estimate for h. Experience has shown that a good approximation of h is given by
h = 14n - 6
(2)
for a Cn+ fraction. Eq. 2 is a useful empirical relation but should not be considered a restraint on the
model.
Considering the remaining two parameters in Eq. 1, a and b, a useful property of the
three-parameter gamma function is that the product ab equals the arithmetic average molecular
weight, Mn+ , minus h.
Mn+ - h = ab
(3)
There are several empirical correlations available for estimating a from randomly sampled
data such as fully extended molar distribution. b is easily calculated from the other variables.
b = (Mn+ - h)/a
(4)
(5)
Y = ln[(Mn+ - h)/mG]
(6)
where,
and
(7)
mG = [ ( M i - h ) ]
Zi
1/ Z n+
i=n
Eq. 5 is valid for 0 < Y < 0.5772 (i.e., a >= 1.0). The variable mG merely represents a geometric
average molecular-weight variable. For Eq. 5 to be useful for the molar distribution problem, it is
necessary to have measured SCN (single-carbon-number) mole fractions and molecular weights
accurately. They should also constitute a full compositional analysis, preferably having a
diminishing or negligible quantity of the last fraction, N. Since such analyses are nearly nonexistent,
a set of correction tables has been developed when only partial analyses are available (the limitation
of a >= 1.0 mentioned previously is also lifted in the present use of Eq. 5).
Due to the limitation of the Eq. 5, we may determine a by minimization of a error function.
The error function, E(a) , used to optimize a is defined as the sum of the squares of differences in
measured and calculated compositions and is given by
(8)
E(a) =
(Z i
Z)
i=n
E(a) =
( f
wi
- f )
i=n
if a weight distribution is to fit (the more common case). Calculated values are marked with a tilde.
24
Compositional Simulation
www.petroman.ir
25
Compositional Simulation
www.petroman.ir
The first of the two pseudoization conditions is satisfied by defining pseudo component properties
as:
(10)
n
n
(l)
(l)
2
.5
(l)
l
(l)
A(l) = a p( T c ) / T 2 pc = x i x j (1 - d ij )( Ai A j )
i=1 j=1
(l)
= b p T c / T p
l
(11)
(l)
c
l
j
Bj
j=1
Tc
(l)
(12)
Pc(l) =
(l)
x j pcj
j=1
(13)
(l)
T cj
j=1
vc
(l)
(14)
(l)
j
v cj
j=1
(15)
(l)
(l)
j
Mj
(l)
*
j
j=1
*(l)
Mj
j=1
0.5
(16)
( x
(l)
j
0.5
Mj )
j=1
a a
i
i=1
ij
(i)
(j) 0.5
(1- d )( A A )
j=1
i=1
j=1
(17)
x x (1 - d
i
26
Compositional Simulation
www.petroman.ir
0.5
ij
)( Ai A j )
The pair of pseudo components are arbitrarily labeled components 1 and 2 on the left-hand
side of Eq. 17. {xi} is the n-component composition of an arbitrary mixture of the two pseudo
components, i.e.
xi = a1xi1 + a2xi
(18)
(19)
(20)
where MN is the molecular weight of the last SCN group (which may actually be a plus fraction),
and I = 1, 2, .... Ng. Molecular weights of SCN groups falling within the boundaries of these values
are included in the MCN group, I.
27
Compositional Simulation
www.petroman.ir
Compositional Simulation:
There are two basic steps to work on compositional simulation. One is the PVT match, the
other is the simulation of the compositional model. The following are the step-wise procedures and
the corresponding data files. DESKTOP_PVT package was used for PVT match, compositional
simulator is VIP. Data are from Third SPE Comparative Solution Project: Gas Cycling of
Retrograde Condensate Reservoirs1 (we name this paper as SPE3).
1- Heavy Ends Characterization. This is the first step for PVT matching. The following
procedure is to be followed:
a- Invoke DESKTOP_PVT. The Peng-Robinson EOS is selected by picking CONFIG
from the main menu, then EOS submenu, then typing PR. (noted as CONFIG/EOS,
PR).
b- Turn on heavy end characterization by choosing CONFIG/Heavy, Y.
c- Enter heavy end parameters by selecting (from main menu) HEAVY/Parameter. At
least three numbers must be input to complete this step, they are: molecular weight,
specific gravity, and mole fraction of C7+ . These numbers can be found in SPE3.
d- Specify the names and number of heavy component groups by choosing PSEUDOCOMPONENT NO. AND NAME.
e- Select cut-off molecular weights and specify 130 for bracket 1 and 180 for bracket 2.
f- After inputting these data, you click CALCULATE option in the HEAVY menu.
g- Inspect the data by selecting HEAVY/Review.
h- Save the calculated EOS parameters by selecting HEAVY/Save EOS. Name the file
heavy.eos.
i- Save the data by choosing FILE/Save and name the file heavy.dat.
Molecular Weight =
140
Specific gravity =
0.774
Mole Fraction
=
0.0659
Cut-off Molecular Weight
130
180
Names of Heavy Components: HVY1 HVY2 HVY3
Output:
HEAVY.DAT, HEAVY.EOS
2- Default Fluid Predictions. At this step, all components from SPE3 should be input except
heavy components. The following procedure is to be followed:
a- Choose CONFIG/Test Type.
b- Select, Cnst Composition, Cnst Volume, and/or Swelling.
c- Choose CONFIG/Heavy, N, to turn off the heavy components option.
d- Choose COMPONENT/System, and highlight the components required (do not include
the C7+).
e- Load The EOS for the heavy component by choosing COMPONENT/Append EOS,
heavy.eos.
f- Enter pure component mole fraction by choosing COMPONENT/Composition.
g- Enter the data for the tests by choosing TESTS/Cnst Composition.
h- Repeat for the Cnst Volume and Swelling tests.
28
Compositional Simulation
www.petroman.ir
29
Compositional Simulation
www.petroman.ir
4- Component Pseudoization. This step is called the pseudoization. The pseudo components are
listed below. Once you give the names of pseudo components and components included in
them, the program is capable of calculating regrouped lab data. Test data must be reentered for
the new pseudo components. New files will be generated: pseudo.dat and pseudo.eos. The
following procedure is to be followed:
abcdefghijkl-
Choose COMPONENT/Load EOS, regress.eos. This will load the regressed EOS.
Turn off regression by choosing CONFIG/Regression, N.
Turn Pseudoization on by selecting CONFIG/Pseudoization, Y.
Choose PSEUDO/Pseudo Name. Give names to the new pseudo components.
Choose PSEUDO/Parameter. Specify the components in Pseudo-COMP P1 Lump.
Repeat for all pseudo components.
Choose PSEUDO/Calculate.
Save the calculated EOS parameters by selecting PSEUDO/Save EOS. Name the file
pseudo.eos.
Choose PSEUDO/Replace EOS. This replaces the active EOS.
Turn off pseudoization by choosing CONFIG/Pseudoization, N.
Reenter the test data by choosing TESTS/Const Composition. Press F2 and load the
global reference composition. Repeat for other tests.
Run the program by selecting RUN/Go.
Save the data by choosing FILE/Save and name the file pseudo.dat.
Pseudo Components:
P1
C1 + N2
P2
C2 + CO2
P3
C3 + NC4 + IC4
P4
C6 + NC5 + IC5
P5
HVY1
P6
HVY2
P7
HVY3
Test Types:
Constant Composition Expansion, Constant Volume Depletion and
Swelling
Output:
PSEUDO.DAT, PSEUDO.EOS
5- Regression. This is the final step for PVT match. Procedures are the same as we did in the step
3. Regression is based on the data obtained in the step 4. The main problem is the selection of
variables to be regressed in order to obtain a good match. There is no definite way to proceed.
In this case trial and error method was used to get a good match. Selecting proper variables is a
time consuming effort. The following procedure is to be followed:
a- Choose CONFIG/Regression, Y, to turn on regression.
b- Choose (from main menu), REGRES/Variable. Choose EOS Property and enter the
number for the regression variable in the appropriate place (delete old numbers if
present). Then Choose Binary Coeff and enter the number for the regression variable in
the appropriate place. Then Exit.
30
Compositional Simulation
www.petroman.ir
cdefghijk-
Choose REGRES/Limits and change the minimum to 0.7 and maximum to 1.3.
Choose REGRES/Control and change the no. of iterations to 20.
Run the program by Choosing RUN/Go.
View the data by choosing REPORT/Graphics.
Save the calculated EOS parameters by selecting REPORT/Save EOS. Name the file
final.eos.
Choose COMPONENT/Load EOS, final.eos. This will overwrite the old EOS.
Choose CONFIG/Regression, N, to turn off regression.
Run the program by Choosing RUN/Go.
Save the data by choosing FILE/Save and name the file final.dat.
Pseudo Components:
P1
C1 + N2
P2
C2 + CO2
P3
C3 + NC4 + IC4
P4
C6 + NC5 + IC5
P5
HVY1
P6
HVY2
P7
HVY3
Regress Variables:
P1-ZC
1
P2-ZC
2
P5-ZC
3
P6-ZC
4
P7-ZC
5
P1-OMEGGA
6
P1-OMEGAB
7
P5,P6,P7-OMEGAA
8
P5,P6,P7-OMEGAB
9
DJK P1-P7
10
Output:
FINAL.DAT, FINAL.EOS
6- Initialization Data for Compositional Simulation. After step 5, the adjusted EOS properties
are obtained. These data are considered as the "correct values" and input in the initialization file
along with other data from SPE3. The format of initialization file is attached in the following
pages. VIP will create a spe3i.rst file when you type: "corerun" in the terminal. It should be
noted that you should check the "corefil.dat" in the current directory before you type command
"corerun."
Data Sources:
File Name:
31
Compositional Simulation
www.petroman.ir
7- Run Data for Compositional Simulation. This is the last step for compositional simulation.
After creating the simulation run data file-spe3r.dat and having a proper restart file-spe3i.rst
from step 6. you can type the command "execrun", VIP will automatically produce the
information specified by the user in the spe3r.dat. The format of the run data file spe3r.dat is
attached in the following pages.
Data Sources:
File Name:
SPE12278, Final.dat
spe3r.dat
32
Compositional Simulation
www.petroman.ir
Data Files
The following pages contain the data files and the output files from the VIP simulator.
Below is a list of the enclosed material.
Heavy.dat
Heavy.eos
Default.dat
Regression.dat
Regression.eos
Pseudo.dat
Pseudo.eos
Final.dat
Final.eos
Spe3i.dat
Spe3r.dat
33
Compositional Simulation
www.petroman.ir
HEAVY.DAT
SPLIT
MWPLUS GPLUS ZPLUS NG MWGRP
140.00 0.7740 0.06590 3 130.00 180.00
PROPERTY CORRELATION SIMULATION
TC
RIAZI-DAUBERT
PC
RIAZI-DAUBERT
ACENTRIC EDMISTER
ZC
RIAZI-DAUBERT
CONMWI
MWC6C7
92.00
MWINC
12.000
END
34
Compositional Simulation
www.petroman.ir
HEAVY.EOS
EOS PR
COMPONENTS
HVY1 HVY2 HVY3
C
C PLUS FRACTION PSEUDO COMPONENT PROPERTIES
C
PROPERTIES
COMP MW
TC
PC
ZC
Accentric OMEGA A
HVY1 107.77 561.98 410.55 26239 .31504
X
HVY2 148.05 677.63 325.18 24598 .41381
X
HVY3 224.01 843.58 244.42 23072 .57895
X
OMEGA B
X
X
X
DJK C1
HVY1
.036113
HVY2
.041436
HVY3
.049070
ENDEOS
C
C PLUS FRACTION PSEUDO COMPONENT COMPOSITIONS
C
COMPOSITION
.034766 .019684 .011451
DEFAULT.DAT
EOS PR
COMPONENTS
CO2 N2 C1 C2 C3 NC4 IC4 NC5 IC5 NC6 HVY1 HVY2 HVY3
NCV 13
PROPERTIES F
PSIA
COMP MW
TC
PC
ZC ACENTRIC OMEGAA OMEGAB PCHOR
CO2 44.010 87.90 1070.9 .27420 .22250
.457236 .077796
49.60
N2
28.013 -232.40 493.0 .29100 .03720
.457236 .077796
35.00
C1
16.043 -116.60 667.8 .28900 .01260
.457236 .077796
71.00
C2
30.070 90.10 707.8 .28500 .09780
.457236 .077796
111.00
C3
44.097 206.00 616.3 .28100 .15410
.457236 .077796
151.00
NC4 58.124 305.70 550.7 .27400 .20150
.457236 .077796
191.00
IC4
58.124 275.00 529.1 .28300 .18400
.457236 .077796
191.00
NC5 72.151 385.70 488.6 .26200 .25240
.457236 .077796
231.00
IC5
72.151 369.10 490.4 .27300 .22860
.457236 .077796
231.00
NC6 86.178 453.70 436.9 .26400 .29980
.457236 .077796
271.00
HVY1 107.770 561.98 410.5 .26239 .31504
.457236 .077796
353.20
HVY2 148.050 677.63 325.2 .24598 .41381
.457236 .077796
474.02
HVY3 224.010 843.58 244.4 .23072 .57895
.457236 .077796
677.90
DJK CO2
N2
.000000
C1
.150000
C2
.150000
C3
.150000
NC4
.150000
IC4
.150000
NC5
.150000
IC5
.150000
NC6
.150000
HVY1 .150000
HVY2 .150000
HVY3 .150000
DJK N2
C1
.120000
C2
.120000
C3
.120000
NC4
.120000
IC4
.120000
NC5
.120000
IC5
.120000
NC6
.120000
HVY1 .120000
HVY2 .120000
HVY3 .120000
DJK C1
C2
.000000
C3
.000000
NC4
.020000
IC4
.020000
NC5
.020000
IC5
.020000
NC6
.025000
HVY1 .036113
HVY2 .041436
HVY3 .049070
DJK C2
C3
.000000
NC4
.010000
IC4
.010000
NC5
.010000
IC5
.010000
NC6
.010000
HVY1 .010000
36
Compositional Simulation
www.petroman.ir
HVY2 .010000
HVY3 .010000
DJK C3
NC4
.010000
IC4
.010000
NC5
.010000
IC5
.010000
NC6
.010000
HVY1 .010000
HVY2 .010000
HVY3 .010000
DJK NC4
IC4
.000000
NC5
.000000
IC5
.000000
NC6
.000000
HVY1 .000000
HVY2 .000000
HVY3 .000000
DJK IC4
NC5
.000000
IC5
.000000
NC6
.000000
HVY1 .000000
HVY2 .000000
HVY3 .000000
DJK NC5
IC5
.000000
NC6
.000000
HVY1 .000000
HVY2 .000000
HVY3 .000000
DJK IC5
NC6
.000000
HVY1 .000000
HVY2 .000000
HVY3 .000000
DJK NC6
HVY1 .000000
HVY2 .000000
HVY3 .000000
DJK HVY1
HVY2 .000000
HVY3 .000000
DJK HVY2
HVY3 .000000
HIDEAL
CO2 2.1028E+02 5.0361E+00 4.4507E-03 -1.1658E-06 1.5274E-10 -5.7829E-15
N2
-2.6164E+01 7.1489E+00 -4.9835E-04 4.4513E-07 -9.0210E-11 4.4521E-15
C1
-1.1193E+02 9.1718E+00 -4.7216E-03 6.7887E-06 -2.4493E-09 3.1208E-13
C2
-6.3147E-01 7.9649E+00 -7.5217E-04 8.7905E-06 -3.8672E-09 5.4789E-13
C3
-3.2544E+01 7.6111E+00 4.1469E-03 9.5048E-06 -4.7227E-09 7.0237E-13
NC4 4.3186E +02 5.7293E+00 1.5646E-02 3.0120E-06 -2.4420E-09 3.8132E-13
IC4
6.6831E+02 2.7132E+00 1.9460E-02 8.3832E-07 -1.8392E-09 3.1555E-13
NC5 1.9605E+03 -2.0202E-01 3.1752E-02 -6.2257E-06 5.8994E-10 -1.4225E-14
IC5
1.9930E+03 -2.2728E+00 3.3903E-02 -7.0911E-06 7.4305E-10 -2.1274E-14
NC6 -6.3694E+02 1.9744E+01 -7.0295E-03 3.9020E-05 -2.1744E-08 4.0917E-12
HVY1 0.0000E+00 4.4727E+00 4.4674E-02 -6.3579E-06 0.0000E+00 0.0000E+00
HVY2 0.0000E+00 2.7007E+00 5.8820E-02 -7.8301E-06 0.0000E+00 0.0000E+00
HVY3 0.0000E+00 4.9001E+00 8.9064E-02 -1.1580E-05 0.0000E+00 0.0000E+00
ENDEOS
C
PVTFILE
KVFILE
C
CCEXP RUN1
COMPOSITION
37
Compositional Simulation
www.petroman.ir
38
Compositional Simulation
www.petroman.ir
REGRESS.DAT
REGRESS
VARIABLE MIN INIT MAX
1
.8000 1.0000 1.2000
2
.8000 1.0000 1.2000
3
.8000 1.0000 1.2000
4
.8000 1.0000 1.2000
5
.8000 1.0000 1.2000
6
.8000 1.0000 1.2000
7
.8000 1.0000 1.2000
8
.8000 1.0000 1.2000
9
.8000 1.0000 1.2000
10
.8000 1.0000 1.2000
11
.8000 1.0000 1.2000
IMAX
IPRINT
H
TOL1
TOL2
TOL3
20
1
.200000 .001000 .001000 .010000
COMP MW TC PC ZC ACENTRIC OMEGAA OMEGAB PCHOR
C1
X X X
X
X
1
2
X
HVY1 X X X X
X
3
6
X
HVY2 X X X X
X
4
7
X
HVY3 X X X X
X
5
8
X
DJK C1
HVY1 9
HVY2 10
HVY3 11
ENDREG
C
C
EOS PR
COMPONENTS
CO2 N2 C1 C2 C3 NC4 IC4 NC5 IC5 NC6 HVY1 HVY2 HVY3
NCV 13
PROPERTIES F
PSIA
COMP MW
TC
PC
ZC
ACENTRIC OMEGAA OMEGAB PCHOR
CO2
44.010 87.90 1070.9 .27420 .22250 .457236 .077796
49.60
N2
28.013 -232.40 493.0 .29100 .03720 .457236 .077796
35.00
C1
16.043 -116.60 667.8 .28900 .01260 .457236 .077796
71.00
C2
30.070 90.10 707.8 .28500 .09780 .457236 .077796 111.00
C3
44.097 206.00 616.3 .28100 .15410 .457236 .077796 151.00
NC4
58.124 305.70 550.7 .27400 .20150 .457236 .077796 191.00
IC4
58.124 275.00 529.1 .28300 .18400 .457236 .077796 191.00
NC5
72.151 385.70 488.6 .26200 .25240 .457236 .077796 231.00
IC5
72.151 369.10 490.4 .27300 .22860 .457236 .077796 231.00
NC6
86.178 453.70 436.9 .26400 .29980 .457236 .077796 271.00
HVY1 107.770 561.98 410.5 .26239 .31504 .457236 .077796 353.20
HVY2 148.050 677.63 325.2 .24598 .41381 .457236 .077796 474.02
HVY3 224.010 843.58 244.4 .23072 .57895 .457236 .077796 677.90
DJK CO2
N2
.000000
C1
.150000
C2
.150000
C3
.150000
NC4
.150000
IC4
.150000
NC5
.150000
IC5
.150000
NC6
.150000
HVY1 .150000
HVY2 .150000
HVY3 .150000
DJK N2
C1
.120000
39
Compositional Simulation
www.petroman.ir
C2
.120000
C3
.120000
NC4
.120000
IC4
.120000
NC5
.120000
IC5
.120000
NC6
.120000
HVY1 .120000
HVY2 .120000
HVY3 .120000
DJK C1
C2
.000000
C3
.000000
NC4
.020000
IC4
.020000
NC5
.020000
IC5
.020000
NC6
.025000
HVY1 .036113
HVY2 .041436
HVY3 .049070
DJK C2
C3
.000000
NC4
.010000
IC4
.010000
NC5
.010000
IC5
.010000
NC6
.010000
HVY1 .010000
HVY2 .010000
HVY3 .010000
DJK C3
NC4
.010000
IC4
.010000
NC5
.010000
IC5
.010000
NC6
.010000
HVY1 .010000
HVY2 .010000
HVY3 .010000
DJK NC4
IC4
.000000
NC5
.000000
IC5
.000000
NC6
.000000
HVY1 .000000
HVY2 .000000
HVY3 .000000
DJK IC4
NC5
.000000
IC5
.000000
NC6
.000000
HVY1 .000000
HVY2 .000000
HVY3 .000000
DJK NC5
IC5
.000000
NC6
.000000
HVY1 .000000
HVY2 .000000
HVY3 .000000
DJK IC5
NC6
.000000
HVY1 .000000
HVY2 .000000
HVY3 .000000
DJK NC6
40
Compositional Simulation
www.petroman.ir
HVY1 .000000
HVY2 .000000
HVY3 .000000
DJK HVY1
HVY2 .000000
HVY3 .000000
DJK HVY2
HVY3 .000000
HIDEAL
CO2 2.1028E+02 5.0361E+00 4.4507E-03 -1.1658E-06 1.5274E-10 -5.7829E-15
N2
-2.6164E+01 7.1489E+00 -4.9835E-04 4.4513E-07 -9.0210E-11 4.4521E-15
C1
-1.1193E+02 9.1718E+00 -4.7216E-03 6.7887E-06 -2.4493E-09 3.1208E-13
C2
-6.3147E-01 7.9649E+00 -7.5217E-04 8.7905E-06 -3.8672E-09 5.4789E-13
C3
-3.2544E+01 7.6111E+00 4.1469E-03 9.5048E-06 -4.7227E-09 7.0237E-13
NC4 4.3186E+02 5.7293E+00 1.5646E-02 3.0120E-06 -2.4420E-09 3.8132E-13
IC4
6.6831E+02 2.7132E+00 1.9460E-02 8.3832E-07 -1.8392E-09 3.1555E-13
NC5 1.9605E+03 -2.0202E-01 3.1752E-02 -6.2257E-06 5.8994E-10 -1.4225E-14
IC5
1.9930E+03 -2.2728E+00 3.3903E-02 -7.0911E-06 7.4305E-10 -2.1274E-14
NC6 -6.3694E+02 1.9744E+01 -7.0295E-03 3.9020E-05 -2.1744E-08 4.0917E-12
HVY1 0.0000E+00 4.4727E+00 4.4674E-02 -6.3579E-06 0.0000E+00 0.0000E+00
HVY2 0.0000E+00 2.7007E+00 5.8820E-02 -7.8301E-06 0.0000E+00 0.0000E+00
HVY3 0.0000E+00 4.9001E+00 8.9064E-02 -1.1580E-05 0.0000E+00 0.0000E+00
ENDEOS
C
PVTFILE
KVFILE
C
CCEXP RUN1
COMPOSITION
.012100 .019400 .659900 .086900 .059100 .027800 .023900
.011200 .015700 .018100 .034766 .019684 .011451
TEMP 200.0
F
DEWPT 3428.0 PSIG
PRES VREL XLIQ
ZG
6000.0 0.8045 0.0 1.129
5500.0 0.8268 0.0 1.063
5000.0 0.8530 0.0 0.9980
4500.0 0.8856 0.0 0.9930
4000.0 0.9284 0.0 0.8690
3600.0 0.9745 0.0 0.8220
3428.0 1.000 0.0 0.0
3400.0 1.004 0.009000 0.0
3350.0 1.014 0.02700 0.0
3200.0 1.047 0.08100 0.0
3000.0 1.100 0.1500 0.0
2800.0 1.164 0.0 0.0
2400.0 1.341 0.0 0.0
2000.0 1.611 0.0 0.0
1600.0 2.041 0.0 0.0
1300.0 2.554 0.0 0.0
1030.0 3.293 0.0 0.0
836.00 4.139 0.0 0.0
C
CVDEP RUN1
COMPOSITION
.012100 .019400 .659900 .086900 .059100 .027800 .023900
.011200 .015700 .018100 .034766 .019684 .011451
TEMP 200.0 F
DEWPT 3428.0 PSIG
PRES
3428.0 3000.0 2400.0 1800.0 1200.0 700.00
MW HVY1 HVY3 140.00 127.00 118.00 111.00 106.00 105.00
ZGAS
0.8030 0.7980 0.8020 0.8300 0.8770 0.9240
VPROD
0.0 0.09095 0.2470 0.42026 0.59687 0.7402
SLIQ
0.0 0.1500 0.1990 0.1920 0.1710 0.1520
C
SWELL RUN1
COMPOSITION
41
Compositional Simulation
www.petroman.ir
42
Compositional Simulation
www.petroman.ir
REGRESS.EOS
EOS PR
COMPONENTS
CO2 N2 C1 C2 C3 NC4 IC4 NC5 IC5 NC6 HVY1 HVY2 HVY3
NCV 13
PROPERTIES F PSIA
COMP MW TC
PC ZC ACENTRIC OMEGAA OMEGAB PCHOR
CO2 44.01 87.90 1070.90 .2742 .2225
.4572360 .0777960 49.6
N2
28.01 -232.40 493.00 .2910 .0372
.4572360 .0777960 35.0
C1
16.04 -116.60 667.80 .2890 .0126
.3776518 .0727877 71.0
C2
30.07 90.10 707.80 .2850 .0978
.4572360 .0777960 111.0
C3
44.10 206.00 616.30 .2810 .1541
.4572360 .0777960 151.0
NC4 58.12 305.70 550.70 .2740 .2015
.4572360 .0777960 191.0
IC4
58.12 275.00 529.10 .2830 .1840
.4572360 .0777960 191.0
NC5 72.15 385.70 488.60 .2620 .2524
.4572360 .0777960 231.0
IC5
72.15 369.10 490.40 .2730 .2286
.4572360 .0777960 231.0
NC6 86.18 453.70 436.90 .2640 .2998
.4572360 .0777960 271.0
HVY1 107.77 561.98 410.50 .2624 .3150
.4169631 .0933548 353.2
HVY2 148.05 677.63 325.20 .2460 .4138
.5486750 .0883512 474.0
HVY3 224.01 843.58 244.40 .2307 .5789
.5344757 .0933546 677.9
DJK N2
CO
.0000000
DJK C1
CO2
.1500000
N2
.1200000
DJK C2
CO2
.1500000
N2
.1200000
C1
.0000000
DJK C3
CO2
.1500000
N2
.1200000
C1
.0000000
C2
.0000000
DJK NC4
CO2
.1500000
N2
.1200000
C1
.0200000
C2
.0100000
C3
.0100000
DJK IC4
CO2
.1500000
N2
.1200000
C1
.0200000
C2
.0100000
C3
.0100000
NC4
.0000000
DJK NC5
CO2
.1500000
N2
.1200000
C1
.0200000
C2
.0100000
C3
.0100000
NC4
.0000000
IC4
.0000000
DJK IC5
CO2
.1500000
N2
.1200000
C1
.0200000
C2
.0100000
C3
.0100000
43
Compositional Simulation
www.petroman.ir
NC4
.0000000
IC4
.0000000
NC5
.0000000
DJK NC6
CO2
.1500000
N2
.1200000
C1
.0250000
C2
.0100000
C3
.0100000
NC4
.0000000
IC4
.0000000
NC5
.0000000
IC5
.0000000
DJK HVY1
CO2
.1500000
N2
.1200000
C1
-.1638391
C2
.0100000
C3
.0100000
NC4
.0000000
IC4
.0000000
NC5
.0000000
IC5
.0000000
NC6
.0000000
DJK HVY2
CO2
.1500000
N2
.1200000
C1
-.1585253
C2
.0100000
C3
.0100000
NC4
.0000000
IC4
.0000000
NC5
.0000000
IC5
.0000000
NC6
.0000000
HVY1 .0000000
DJK HVY3
CO2
.1500000
N2
.1200000
C1
.2490480
C2
.0100000
C3
.0100000
NC4
.0000000
IC4
.0000000
NC5
.0000000
IC5
.0000000
NC6
.0000000
HVY1 .0000000
HVY2 .0000000
C
HSTAR
CO2
2.10280E+02 5.03610E+00 4.45070E-03 -1.16580E-06 1.52740E-10
-5.78290E-15
N2
-2.61640E+01 7.14890E+00 -4.98350E-04 4.45130E-07 -9.02100E-11
4.45210E-15
C1
-1.11930E+02 9.17180E+00 -4.72160E-03 6.78870E-06 -2.44930E-09
3.12080E-13
C2
-6.31470E-01 7.96490E+00 -7.52170E-04 8.79050E-06 -3.86720E-09
5.47890E-13
C3
-3.25440E+01 7.61110E+00 4.14690E-03 9.50480E-06 -4.72270E-09
7.02370E-13
NC4
4.31860E+02 5.72930E+00 1.56460E-02 3.01200E-06 -2.44200E-09
3.81320E-13
IC4
6.68310E+02 2.71320E+00 1.94600E-02 8.38320E-07 -1.83920E-09
3.15550E-13
NC5
1.96050E+03 -2.02020E-01 3.17520E-02 -6.22570E-06 5.89940E-10
-1.42250E-14
44
Compositional Simulation
www.petroman.ir
IC5
45
Compositional Simulation
www.petroman.ir
PSEUDO.DAT
EOS PR
COMPONENTS
P1 P2 P3 P4 P5 P6 P7
NCV 7
PROPERTIES F
PSIA
COMP MW
TC
PC
ZC ACENTRIC OMEGAA OMEGAB PCHOR
P1
16.382 -119.91 662.8 .28883 .01330
.376528 .072862 69.97
P2
31.774 89.83 752.2 .28944 .11304
.451923 .079260 103.50
P3
50.642 245.90 581.0 .28180 .17244
.459850 .078437 169.66
P4
77.793 407.26 468.4 .26789 .26316
.460949 .078234 247.09
P5
107.770 561.98 410.5 .26240 .31500 .416963 .093355 353.20
P6
148.050 677.63 325.2 .24600 .41380 .548675 .088351 474.00
P7
224.010 843.58 244.4 .23070 .57890 .534476 .093355 677.90
DJK P1
P2
.000947
P3
.007835
P4
.021885
P5
-.160885
P6
-.155665
P7
.244672
DJK P2
P3
.003695
P4
.010541
P5
.010541
P6
.010541
P7
.010541
DJK P3
P4
.002281
P5
.002281
P6
.002281
P7
.002281
DJK P4
P5
.000000
P6
.000000
P7
.000000
DJK P5
P6
.000000
P7
.000000
DJK P6
P7
.000000
HIDEAL
P3
0.0000E+00 3.0481E+00 2.2327E-02 -3.1330E-06 0.0000E+00 0.0000E+00
P4
0.0000E+00 1.3339E+00 3.0999E-02 -4.1340E-06 0.0000E+00 0.0000E+00
P6
0.0000E+00 2.5153E+01 7.7904E-02 -1.2416E-05 0.0000E+00 0.0000E+00
ENDEOS
C
PVTFILE
KVFILE
C
CCEXP RUN1
COMPOSITION
.679300 .099000 .110800 .045000 .034766 .019684 .011451
TEMP 200.0 F
DEWPT 3428.0 PSIG
PRES VREL XLIQ
ZG
6000.0 0.8045 0.0 1.129
5500.0 0.8268 0.0 1.063
5000.0 0.8530 0.0 0.9980
4500.0 0.8856 0.0 0.9930
4000.0 0.9284 0.0 0.8690
46
Compositional Simulation
www.petroman.ir
3600.0
3428.0
3400.0
3350.0
3200.0
3000.0
2800.0
2400.0
2000.0
1600.0
1300.0
1030.0
836.00
C
CVDEP RUN1
COMPOSITION
.679300 .099000 .110800 .045000 .034766 .019684 .011451
TEMP 200.0 F
DEWPT 3428.0 PSIG
PRES
3428.0 3000.0 2400.0 1800.0 1200.0 700.00
MW P5 P7 140.00 127.00 118.00 111.00 106.00 105.00
ZGAS
0.8030 0.7980 0.8020 0.8300 0.8770 0.9240
VPROD
0.0 0.09095 0.2470 0.42026 0.59687 0.7402
SLIQ
0.0 0.1500 0.1990 0.1920 0.1710 0.1520
C
SWELL RUN1
COMPOSITION
.679300 .099000 .110800 .045000 .034766 .019684 .011451
YINJ
.946800 .052700 .000500 .000000 .000000 .000000 .000000
TEMP
PSIG
ZINJ
0.0
0.1271
0.3046
0.5384
0.6538
C
END
200.0 F
VREL PSAT TYPE
1.000 3428.0 DEWPT
1.122 3635.0 DEWPT
1.354 4015.0 DEWPT
1.925 4610.0 DEWPT
2.504 4880.0 DEWPT
47
Compositional Simulation
www.petroman.ir
PSEUDO.EOS
C EQUATION OF STATE DATA PRODUCED BY DESKTOP-PVT FOR PSEUDO COMPONENTS
C
EOS PR
COMPONENTS
P1 P2 P3 P4 P5 P6 P7
PROPERTIES F PSIA
COMP MW TC PC ZC ACENTRIC OMEGAA OMEGAB PCHOR
P1 16.38 -119.91 662.81 .2888 .0133
.3765281
.0728615 70.0
P2 31.77 89.83 752.18 .2894 .1130
.4519229
.0792598 103.5
P3 50.64 245.90 581.03 .2818 .1724
.4598499
.0784368 169.7
P4 77.79 407.26 468.43 .2679 .2632
.4609493
.0782338 247.1
P5 107.77 561.98 410.50 .2624 .3150
.4169631
.0933548 353.2
P6 148.05 677.63 325.20 .2460 .4138
.5486750
.0883512 474.0
P7 224.01 843.58 244.40 .2307 .5789
.5344757
.0933546 677.9
DJK P2
P1 .0009470
DJK P3
P1 .0078346
P2 .0036951
DJK P4
P1 .0218854
P2 .0105413
P3 .0022813
DJK P5
P1 -.1608847
P2 .0105413
P3 .0022813
P4 .0000000
DJK P6
P1 -.1556652
P2 .0105413
P3 .0022813
P4 .0000000
P5 .0000000
DJK P7
P1 .2446722
P2 .0105413
P3 .0022813
P4 .0000000
P5 .0000000
P6 .0000000
C
HSTAR
P1 0.00000E+00 0.00000E+00 0.00000E+00 0.00000E+00 0.00000E+00
0.00000E+00
P2 0.00000E+00 0.00000E+00 0.00000E+00 0.00000E+00 0.00000E+00
0.00000E+00
P3 0.00000E+00 3.04810E+00 2.23275E-02 -3.13303E-06 0.00000E+00
0.00000E+00
P4 0.00000E+00 1.33391E+00 3.09992E-02 -4.13396E-06 0.00000E+00
0.00000E+00
P5 0.00000E+00 0.00000E+00 0.00000E+00 0.00000E+00 0.00000E+00
0.00000E+00
P6 0.00000E+00 2.51533E+01 7.79036E-02 -1.24159E-05 0.00000E+00
0.00000E+00
P7 0.00000E+00 0.00000E+00 0.00000E+00 0.00000E+00 0.00000E+00
0.00000E+00
C
ENDEOS
C
C Pseudoized Component Composition
C
48
Compositional Simulation
www.petroman.ir
COMPOSITION
.679300 .099000 .110800 .045000 .034766 .019684 .011451
49
Compositional Simulation
www.petroman.ir
FINAL.DAT
REGRESS
VARIABLE MIN INIT MAX
1
.8000 1.0000 1.2000
2
.8000 1.0000 1.2000
3
.8000 1.0000 1.2000
4
.8000 1.0000 1.2000
5
.8000 1.0000 1.2000
6
.8000 1.0000 1.2000
7
.8000 1.0000 1.2000
8
.8000 1.0000 1.2000
9
.8000 1.0000 1.2000
10
.8000 1.0000 1.2000
11
.8000 1.0000 1.2000
12
.8000 1.0000 1.2000
13
.8000 1.0000 1.2000
IMAX
IPRINT
H
TOL1
TOL2
TOL3
20
1
.200000 .001000 .001000 .010000
COMP MW TC PC ZC ACENTRIC OMEGAA OMEGAB PCHOR
P1
X X
X
1
X
6
7
X
P2
X X
X
2
X
X
X
X
P5
11 X
X
3
X
8
9
X
P6
12 X
X
4
X
8
9
X
P7
13 X
X
5
X
8
9
X
DJK P1
P7
10
ENDREG
C
C
EOS PR
COMPONENTS
P1 P2 P3 P4 P5 P6 P7
NCV 7
PROPERTIES F
PSIA
COMP MW TC
PC
ZC ACENTRIC OMEGAA OMEGAB PCHOR
P1
16.382 -119.91 662.8 .28883
.01330
.376528 .072862 69.97
P2
31.774 89.83 752.2 .28944
.11304
.451923 .079260 103.50
P3
50.642 245.90 581.0 .28180
.17244
.459850 .078437 169.66
P4
77.793 407.26 468.4 .26789
.26316
.460949 .078234 247.09
P5
107.770 561.98 410.5 .26240
.31500
.416963 .093355 353.20
P6
148.050 677.63 325.2 .24600
.41380
.548675 .088351 474.00
P7
224.010 843.58 244.4 .23070
.57890
.534476 .093355 677.90
DJK P1
P2
.000947
P3
.007835
P4
.021885
P5
-.160885
P6
-.155665
P7
.244672
DJK P2
P3
.003695
P4
.010541
P5
.010541
P6
.010541
P7
.010541
DJK P3
P4
.002281
P5
.002281
P6
.002281
P7
.002281
DJK P4
P5
.000000
50
Compositional Simulation
www.petroman.ir
P6
.000000
P7
.000000
DJK P5
P6
.000000
P7
.000000
DJK P6
P7
.000000
HIDEAL
P3
0.0000E+00 3.0481E+00 2.2327E-02 -3.1330E-06 0.0000E+00 0.0000E+00
P4
0.0000E+00 1.3339E+00 3.0999E-02 -4.1340E-06 0.0000E+00 0.0000E+00
P6
0.0000E+00 2.5153E+01 7.7904E-02 -1.2416E-05 0.0000E+00 0.0000E+00
ENDEOS
C
PVTFILE
KVFILE
C
CCEXP RUN1
COMPOSITION
.679300 .099000 .110800 .045000 .034766 .019684 .011451
TEMP 200.0 F
DEWPT 3428.0 PSIG
PRES VREL XLIQ
ZG
6000.0 0.8045 0.0 1.129
5500.0 0.8268 0.0 1.063
5000.0 0.8530 0.0 0.9980
4500.0 0.8856 0.0 0.9930
4000.0 0.9284 0.0 0.8690
3600.0 0.9745 0.0 0.8220
3428.0 1.000 0.0 0.0
3400.0 1.004 0.009000 0.0
3350.0 1.014 0.02700 0.0
3200.0 1.047 0.08100 0.0
3000.0 1.100 0.1500 0.0
2800.0 1.164 0.0 0.0
2400.0 1.341 0.0 0.0
2000.0 1.611 0.0 0.0
1600.0 2.041 0.0 0.0
1300.0 2.554 0.0 0.0
1030.0 3.293 0.0 0.0
836.00 4.139 0.0 0.0
C
CVDEP RUN1
COMPOSITION
.679300 .099000 .110800 .045000 .034766 .019684 .011451
TEMP 200.0 F
DEWPT 3428.0 PSIG
PRES
3428.0 3000.0 2400.0 1800.0 1200.0 700.00
MW P5 P7 140.00 127.00 118.00 111.00 106.00 105.00
ZGAS
0.8030 0.7980 0.8020 0.8300 0.8770 0.9240
VPROD
0.0 0.09095 0.2470 0.42026 0.59687 0.7402
SLIQ
0.0 0.1500 0.1990 0.1920 0.1710 0.1520
C
SWELL RUN1
COMPOSITION
.679300 .099000 .110800 .045000 .034766 .019684 .011451
YINJ
.946800 .052700 .000500 .000000 .000000 .000000 .000000
TEMP
PSIG
ZINJ
0.0
0.1271
0.3046
0.5384
200.0 F
VREL PSAT TYPE
1.000 3428.0 DEWPT
1.122 3635.0 DEWPT
1.354 4015.0 DEWPT
1.925 4610.0 DEWPT
51
Compositional Simulation
www.petroman.ir
FINAL.EOS
EOS PR
COMPONENTS
P1 P2 P3 P4 P5 P6 P7
NCV 7
PROPERTIES F PSIA
COMP MW TC
PC ZC ACENTRIC OMEGAA OMEGAB PCHOR
P1
16.38 -119.91 662.80 .2888 .0133
.3528063 .0710390 70.0
P2
31.77 89.83 752.20 .2894 .1130
.4519230 .0792600 103.5
P3
50.64 245.90 581.00 .2818 .1724
.4598500 .0784370 169.7
P4
77.79 407.26 468.40 .2679 .2632
.4609490 .0782340 247.1
P5 102.03 561.98 410.50 .2624 .3150
.4192163 .0950905 353.2
P6 150.30 677.63 325.20 .2460 .4138
.5516401 .0899935 474.0
P7 233.65 843.58 244.40 .2307 .5789
.5373643 .0950905 677.9
DJK P2
P1
.0009470
DJK P3
P1
.0078350
P2
.0036950
DJK P4
P1
.0218850
P2
.0105410
P3
.0022810
DJK P5
P1
-.1608850
P2
.0105410
P3
.0022810
P4
.0000000
DJK P6
P1
-.1556650
P2
.0105410
P3
.0022810
P4
.0000000
P5
.0000000
DJK P7
P1
.2481725
P2
.0105410
P3
.0022810
P4
.0000000
P5
.0000000
P6
.0000000
C
HSTAR
P1
0.00000E+00 0.00000E+00 0.00000E+00 0.00000E+00 0.00000E+00
0.00000E+00
P2
0.00000E+00 0.00000E+00 0.00000E+00 0.00000E+00 0.00000E+00
0.00000E+00
P3
0.00000E+00 3.04810E+00 2.23270E-02 -3.13300E-06 0.00000E+00
0.00000E+00
P4
0.00000E+00 1.33390E+00 3.09990E-02 -4.13400E-06 0.00000E+00
0.00000E+00
P5
0.00000E+00 0.00000E+00 0.00000E+00 0.00000E+00 0.00000E+00
0.00000E+00
P6
0.00000E+00 2.51530E+01 7.79040E-02 -1.24160E-05 0.00000E+00
0.00000E+00
P7
0.00000E+00 0.00000E+00 0.00000E+00 0.00000E+00 0.00000E+00
0.00000E+00
C
52
Compositional Simulation
www.petroman.ir
ENDEOS
C
C Global Reference Component Composition
C
COMPOSITION
.679300 .099000 .110800 .045000 .034766 .019684 .011451
53
Compositional Simulation
www.petroman.ir
Spe3i.dat
C VIP-EXEC COMPOSITIONAL EXAMPLE #1
C
C DATA SET NAME: SPE3I.DAT
C
C FEATURES DEMONSTRATED:
C (1) RECTANGULAR GRID 9X9X4
C (2) GAS CONDENSATE, GAS CYCLING
C
C ----------------------------------------C
INIT
TITLE1
SPE COMPARATIVE PROJECT 3
TITLE2
DATA SET NAME: SPE3I.DAT
TITLE3
SINGLE WELL RECTANGULAR GRID GAS CONDENSATE
C
NCOL 80
NLINES 60
C
PRINT ALL
C
C INITIAL SIMULATION DATE:
DATE 01 01 1992
C
C
NX NY NZ NCOMP
9
9 4
7
C
C CONSTANT RESERVOIR PROPERTIES
DWB BWI VW CW CR TRES TS PS
1. .9813 .3 3.0E-6 4.0E-6 200. 60. 14.7
C
C
C
NOLIST
TABLES
IEQUIL PINIT
DEPTH PCWOC WOC PCGOC GOC
1
3543
7500
0
7500
0
C
PSAT
7500
3443
C
SWT 1
SW KRW
KROW PCWO
0.00 0.0 0.801 75.0
0.04 0.0 0.8005 71.0
0.08 0.0 0.8003 70.0
0.12 0.0 0.8001 60.0
0.16 0.00 0.80 50.0
0.2 0.002 0.65 30.0
0.24 0.01 0.513 21.0
0.28 0.02 0.4 15.5
0.32 0.033 0.315 12.0
0.36 0.049 0.25 9.20
0.40 0.066 0.196 7.00
0.44 0.090 0.15 5.30
0.48 0.119 0.112 4.20
0.52 0.15 0.082 3.40
54
Compositional Simulation
www.petroman.ir
0.56
0.60
0.64
0.68
0.72
0.76
0.80
0.84
0.88
0.92
0.96
1.00
0.186
0.227
0.277
0.330
0.39
0.462
0.54
0.62
0.71
0.80
0.90
1.00
0.060 2.70
0.04 2.10
0.024 1.70
0.012 1.30
0.005 1.00
0.0 0.70
0.0 0.50
0.0 0.40
0.0 0.30
0.0 0.20
0.0 0.10
0.0 0.00
C
SGT 1
SG KRG KROG PCGO
0.0 0.0 0.801 0.0
0.04 0.005 0.650 0.0
0.08 0.013 0.513 0.0
0.12 0.026 0.400 0.0
0.16 0.040 0.315 0.0
0.20 0.058 0.250 0.0
0.24 0.078 0.196 0.0
0.28 0.100 0.150 0.0
0.32 0.126 0.112 0.0
0.36 0.156 0.082 0.0
0.40 0.187 0.060 0.0
0.44 0.222 0.040 0.0
0.48 0.260 0.024 0.0
0.52 0.300 0.012 0.0
0.56 0.348 0.005 0.0
0.60 0.400 0.00 0.0
0.64 0.450 0.00 0.0
0.68 0.505 0.00 0.0
0.72 0.562 0.00 0.0
0.76 0.620 0.00 0.0
0.80 0.680 0.0 0.0
0.84 0.74 0.0 0.0
0.88 0.74 0.0 0.0
0.92 0.74 0.0 0.0
0.96 0.74 0.0 0.0
1.00 0.74 0.0 0.0
C EQUATION OF STATE DATA
C
C
EOS PR
COMPONENTS
P1 P2 P3 P4 P5 P6 P7
PROPERTIES F PSIA
COMP MW TC
PC ZC ACENTRIC
P1
16.38 -119.91 662.80 .2888 .0133
P2
31.77 89.83 752.20 .2894 .1130
P3
50.64 245.90 581.00 .2818 .1724
P4
77.79 407.26 468.40 .2679 .2532
P5 102.03 561.98 410.50 .2624 .3150
P6 150.30 677.63 325.20 .2460 .4138
P7 233.65 843.58 244.40 .23070 .57890
DJK P1
P2
.000947
P3
.007835
P4
.021885
P5
-.160885
P6
-.155665
P7
.248173
DJK P2
P3
.003695
P4
.010541
OMEGAA
.352806
.451923
.459850
.460949
.419216
.551640
.537364
OMEGAB
.071039
.079260
.0784370
.078234
.095091
.089993
.095091
55
Compositional Simulation
www.petroman.ir
P5
.010541
P6
.010541
P7
.010541
DJK P3
P4
.002281
P5
.002281
P6
.002281
P7
.002281
DJK P4
P5
.000000
P6
.000000
P7
.000000
DJK P5
P6
.000000
P7
.000000
DJK P6
P7
.000000
C
ENDEOS
C
C
GASMF
.679300 .099000 .110800 .045000 .034766 .019684 .011451
OILMF
X
X
X
X
X
X
X
C
C
C RESERVOIR DESCRIPTION ARRAYS FOLLOW:
ARRAYS
C
DX CON
293.3
DY CON
293.3
DZ ZVAR
30 30 50 50
MDEPTH ZVAR
7330 7360 7400 7450
POR CON
.13
C
KX ZVAR
130 40 20 150
KY ZVAR
130 40 20 150
KZ ZVAR
13 4 2 15
C
RUN
STOP
END
56
Compositional Simulation
www.petroman.ir
Spe3r.dat
C VIP-EXEC COMPOSITIONAL EXAMPLE #1
C
C DATA SET NAME: spe3r.DAT
C
C FEATURES DEMONSTRATED:
C (1) IMPLICIT
C (2) GAUSS
C
C ----------------------------------------RUN
C
DIM NAMAX
300000
C
IMPLICIT
C
RESTART 0 3
C
TITLE2
DATA SET NAME: spe3r.DAT
TITLE3
SINGLE WELL - RICH GAS CONDENSATE,GAS CYCLING
C
C START CARD DEFINES WHERE DATA PROCESSING BEGINS
START
C
C USE GAUSS AS SOLUTION METHOD
GAUSS
C
C WELL DEFINITION DATA
WELL N IW JW IGC IBAT
1
7 7 1 1
2
7 1 1 1
C
C DEFINE WELL 1 AS A PRODUCER
PROD G 1
INJ G 2
YINJ 2
0.8036 0.1052 0.08 0.0112 0.0 0.0 0.0
FPERF
WELL L IW JW KH
1 3 7 7 1000
1 4 7 7 7500
2 1 1 1 3900
2 2 1 1 1200
C FLOW CHARACTERISTICS FOR WELL 1!
WI 1
1.55
C FLOW CHARACTERISTICS FOR WELL 2
WI 2
1.55
C
C PRESSURE AND RATE CONTROLS FOR WELL 1
BHP 1
500
7400
QMAX 1
6200
QMAX 2
4700
C
C SURFACE SEPARATION DATA FOR WELL 1
SEPARATOR 1
57
Compositional Simulation
www.petroman.ir
58
Compositional Simulation
www.petroman.ir
REFERENCES:
1. Coats, K.H. and Smart, G.T.: " Application of a Regression Based EOS PVT Program to
Laboratory Data," SPERE (May 1986) 277-299.
2. Peng, D.-Y. and Robinson, D.B.: "A New Two-Constant Equation of State," Ind. Eng. Chem.
Fundam. (1976) 15, 59.
3. Soave, G.: "Equilibrium Constants From a Modified Redlich Kwong Equation of State," Chem.
Eng. Sci. (1972) 27, 1197-1203.
4. Coats, K.H., Dempsey, J.R., and Henderson, J.H.: "A New Technique for Determining Reservoir
Description from Field Performance Data," SPEJ (March 1970) 66-74.
5. Whitson, C.H.: "Characterizing Hydrocarbon Plus Fractions," SPEJ (Aug. 1983) 683-94.
6. Coats, K.H.: "Simulation of Gas Condensate Reservoir Performance," JPT (Oct. 1985) 1870-86.
7. Kenyon, D.E. and Behie, A.: "Third Comparative Solution Project: Gas Cycling of Retrograde
Condensate Reservoirs," SPE 12278, JPT (August 1987), 981-997.
8. Abel, W., Jackson, R.F. and Wattenbarger, R.A.: "Simulation of a Partial Pressure Maintenance
Gas Cycling Project with a Compositional Model, Carson Creek Field, Alberta", Numerical
Simulation, SPE Reprint Series No. 11, 285-293.
9. Thompson, F.R.: "Compositional Simulation of a Gas Cycling Project, Bonnie Glen D-3A Pool,
Alberta, Canada", Numerical Simulation, SPE Reprint Series No. 11, 314-329.
10. Field, M.B., Givens, J.W. and Paxman, D.S.: "Kaybob South - Reservoir Simulation of a Gas
Cycling Project with Bottom Water Drive", Numerical Simulation, SPE Reprint Series No. 11, 294305.
11. Course notes, PETE 605, PETE 607 AND PETE 611, Texas A&M University.
12. Lake, L.W.: Enhanced Oil Recovery, Prentice Hall, 1989.
13. Stalkup, F.I.: Miscible Flooding Fundamentals, SPE Monograph Series, 1983.
59
Compositional Simulation
www.petroman.ir