Building the Dynamic

Reservoir Model
 Integrated Reservoir Study
DYNAMIC MODEL
Well/Facilities Near Wellbore
Model Performance Models

Upscaled Preliminary Calibrated

Reservoir Dynamic Dynamic
Model Model Model

Optimized
Projected Field
Reservoir
Reservoir Economic
Development
Performance Model
Plan
 Dynamic Reservoir Model
• A mathematical model that describes and
simulates the time-dependent flow processes
active in a hydrocarbon reservoir
• The dynamic model combines
– The static model
– Pressure- and saturation-dependent
properties
– Near-well, wellbore, and facilities properties
to calculate production and pressure vs time
 Importance
Static Model
• Model of reservoir at initial conditions
• Insufficient to predict performance or optimize
development
Dynamic Model
• Models entire life of reservoir
• Can produce reservoir several times under different
operating conditions
• Can optimize reservoir depletion plan
Steps in Model Construction
• Selection of the model type
• Selection of the fluid model and number of
phases
• Selection of the grid coordinate system and
number of dimensions
• Selection of optimal grid block sizes
• Specification of reservoir properties
• Simplification of reservoir geometry
• Selection of the well model
• Specification of well control
• Specification of time step size control
• Selection of the numerical solution method
Factors Affecting Model Design
• Objectives of the simulation study
• Quality of the answer needed to satisfy the
project objectives
• Complexity of the reservoir processes to
be modeled including secondary and/or
tertiary recovery processes
• Budget constraints
• Time available to complete the study
Factors Affecting Model Design
• Availability and quality of reservoir data
• Availability and quality of historical
production data
• Capabilities and ease of use of available
simulators
• Capabilities and availability of computer
facilities
 Data Categories
• Reservoir and well data

• Simulation specific data

– Simulation grid

– Simulation time schedule

– Simulator control data

 Reservoir and Well Data
• Reservoir fluid property data

• Reservoir rock property data

• Strata properties

• Well data

• Wellbore and facilities data

 Location of VLE 196 Field in Block V, Lake
Maracaibo, Venezuela
Caribbean Maracaibo MARACAIBO
Sea
N
Barranquilla

Lake
Maracaibo

COLOMBIA VENEZUELA I

II

IX X VII
XV III

VI
XI
V VII
XIII

LAKE
BLOCK MARACAIBO
V
VLE - 196
AREA
VLE 400 Fault
Base Map for the C4-C5 Reservoir
1078000.

LRF 0033
LRF 0009 LRF 0003 LRF 0028
LRF 0062

LRF 0039

Simulation Study LRF 0006 LRF 0079

214000.
LRF 0078

LRF 0113

LRF 0114
LRF 0016

216000.
LRF 0026

LRF 117

LRF 0035

Block VI
Block V
7 5 12 11
6 4
8 VLE 1215

10
1076000.

VL
9
E4
00
fau

VLE 647
3
lt

1074000.
VLE

1
400
f au

Scale 1:20000.
lt

200. 0. 200. 400. 600. 800. 1000. meters

VLE 738
0.2 0. 0.2 0.4 0.6 0.8 1. kilometers

S.A. Holditch & Associates, Inc.

Maraven, VLE-196 Reservoir
Block V Lamar, Lake
VLE 470 Maracaibo
Fault Areas Map
Scale 1:20000.
1072000.
 Introduction
• VLE-196 field discovered in 1958
• Produced over 200 million STB
• Wells produced with sliding sleeves
to control water production
• About 50 wells completed in the
Misoa sandstones
 Objectives
• Characterize the reservoir for
modeling
• Identify bypassed and untapped
compartments
• Develop a plan to increase
production rates and reserves
Reservoir Fluid Property Data
• 5 basic reservoir fluid types
– Dry gas
– Wet gas
– Retrograde gas (gas condensate)
– Volatile oil
– Black oil
 The Fluid Model Must
• Be consistent with the project objectives
• Properly account for the number of phases
existing at reservoir and surface conditions
(This requirement can be relaxed in cases
where immovable fluids exist which do not
contain appreciable amounts of the movable
fluids or components in solution)
• Accurately predict phase changes which
occur during the simulation
 The Fluid Model Must
• Accurately predict the volumetric properties,
including compressibility and density, of the
reservoir and surface fluids
• Accurately predict the viscosity of reservoir
fluids and any other critical parameters such as
specific heats in cases of nonisothermal
simulations
• When an equation of state is used to predict
phase behavior, the equation must be "tuned"
to match all available special laboratory data
measured on valid reservoir fluid samples
Reservoir Fluid Property Data
• Fluid PVT data
– Black oil

• Bo(p), Rso (p), µ o(p), ρ o(p)

• Bg(p), µ g(p), ρ g(p)

– Volatile oil

• Bo(p), Rso (p), µ o(p), ρ o(p)

• Bg(p), Rsg (p), µ g(p), ρ g(p)

 PVT Properties
• Undersaturated oil at initial conditions
• Compositional gradient present
initially
• Extrapolated PVT data from 5
laboratory analyses
• Resulting black-oil PVT properties
used in model
Reservoir Fluid Property Data

• Equilibration data
– Original pressure at datum
– Fluid contact elevations
Reservoir Rock Property Data

• This section contains rock properties usually

derived from core analysis
• Small number of samples, so not enough data to
generate maps, or distributions, of these
properties
• Usually assume these properties are constant
throughout a particular zone, or stratum
Reservoir Rock Property Data
• Capillary pressure
• Relative permeability
• Hysteresis in capillary pressure and relative
permeability
• Pressure-dependent porosity and permeability
functions
• Non-darcy flow (gas)
• Dual-porosity data
 Relative Permeability and
Capillary Pressure Data
• Initially developed 4 sets of data, for
4 different porosity ranges
• Initially, did not adequately model
water production in thin, high-
permeability layers
• Pseudo relative permeabilities were
developed to match observed water
production
Capillary Pressure (J Function) as a
Function of Water Saturation
350.0

300.0

250.0
J(Sw)=Pc*(K/Phi)^0.5

200.0

150.0

100.0

50.0

0.0
0 10 20 30 40 50 60 70 80 90 100
Sw
 Strata Properties
• This section contains rock properties usually
derived from well log data or seismic data
• Larger number of samples, so often enough data
to generate maps, or distributions, of these
properties
• Often input separate maps of these properties
for each zone, or stratum
 Strata Properties
• Structural properties
– Elevation (structure map)
– Gross interval thickness
– Reservoir limits
– Discontinuities (faults)
 Strata Properties
• Formation properties
– Net-to-gross ratio (or net sand thickness)
– Porosity
– Formation permeability
– Pore volume and transmissibility modifications
Reservoir Characterization
Previously Documented
• Martin, et al., 1997
• Geophysical, geological and
petrophysical analyses
• Geological model consisting of 19
layers
• Correlated seismic attributes to
petrophysical parameters to improve
reservoir mapping
 18
L R F 0 1 1 3

18

18
21 18
18

18
L R F 0 0 3 0
18
L R F 0 0 1 7
L R F 0 2 3
B L O C K V I L R F WE L L S 15

21
B L O C K V V L E W E L 1L S 1 0 9
2 F 1

18
VLE 400 Fault

2
V L E 1 1 0 2
9 4 1 1 6 1 2 8

15
1 3 5
B F F V L E 9 8 7
V L E V 1 L 0E 1 63 9 4 VF L E 1 1 2 8

Misoa Formation,
V L E 8 1 0 3 4 V L E 4 2 2
1 1 5 0
F / R

15
V L E 1 1 2 7
18 21
01 4 5 1 1 5
1 2 15
/ RB / RF
F
2 1 7 0 4 1 V L 9EV 1 9 L 0 51 E 8 0 0 1 6 0 4 6
L VE 18 21 21
2 F1 21
F /B B / C / R

21
V LV E L E 6 6 1 31 3 9
1 6 2
V L E 9 7 3 18
1 6 2
V 1 L 3 E 8 1 0 0 7 B B / F
F / C V L E 5 1 0 V L E 6 3 1

21
V L E 9 8 5
15

18
C-5 Sand, Layer 7,
21 4 3 18

18
C
V L E 6 7 5
21 1 5 3
1

21
C / F 1 2 6 6
V L E 1 0 6 3
B 18
V L E 1 0 0 4 21
18 1 6 9
8 0 B / F
V L E 1F 0 1 9 V L E 5 1 6 V L E 2 9 7
V L E 4 4 9 21
V L E 6 4 7
V L E 1 0 0 251 1 5 4 21
18

18
Average Porosity
1 2 17B
F 21
2

V L E 18 1 1 3 V 5 VL E
L E
6 7
1
4
1 5 5
1 0 2
F
V L E 1 1 0 1 21
21

V L E 6 5 5 1 4 4 21
V L E 6 3 5 1 1 7

18
C / F

21
F / R
V L E 5 0 6 V L E 5 7 1 18
21

1 5 3
B 18 18
V L E 1 1 3 0
1 2 8
F
V 1L 6E 1 6 2 71 7

21
F
V L E 01 9 2 61 1 0 2
V L E 4 7 1 F / R F / R
V L E 4 0 0
V 1 8L E 1 9 6 V L E 6 5 1
18
1 2 7
21 F
V L E 1 1 4 0
V L E 1 0 0

V L E 9 9 2
V L E 5 0 4 9 9

N V L E 4 6 0
V
F
L E 6 1 9
21
18
V L E 1 1 4 8
21
21

1 7 6 21
Percent V L E 9 6 8
V
F
L E
/ R
4 5 3
V L E 0 9 4 18
21
>21
18 18

21 18
18
V L
18
E 7 3 8 V L E 4 3 2 1 2 6 18
18

V L E 5 F 1 /1 R 1
2

V L E 6 7 1 18
18 21 21
15
V L E 9 9 9
15 21 18
12 18
1 6 6
F
V L E 7 1 5 V L E 0 9 1 V L
V L E 4 7 0 2 E1 4 6 5

18
21
18

0 3000 21 18
ft 18

0 1000
m
 Well Data
• Completion data

• Production and injection data

• Pressure data

• Operational control and constraint

information
 Well Data
• Completion data
– Well bottomhole location and orientation
– Perforation interval
– Wellbore radius
– Permeability-thickness product
– Skin factor
– Hydraulic fracture length
– Hydraulic fracture conductivity
– Well constant
 Well Data
• Production and injection data
– Production vs time for all produced
fluids – oil, gas, condensate, water
– Injection vs time for all injected fluids –
gas, water
 Well Data
• Pressure data

– Static pressure data

– Flowing pressure data

– Pressure transient test data

 Well Data
• Operational control and constraint
information
– How are wells produced?
• Natural flow
• Rod pump
• Gas lift
• Submersible pump
 Well Data
• Are there any limits, or constraints, on production?
– Equipment pressure limitations
– Maximum fluid lifting capacity
– Maximum water handling capacity
– Economic limit production rates
– Allowable production limited by regulatory agencies
 Well Data
• How are injection wells operated?

• Are there any limits, or constraints, on

injection?

– Equipment pressure limitations

– Maximum injection rate capacity

Wellbore and Facilities Data
• Well mechanical configuration data
– Lengths and ID’s of tubulars through
which fluids are produced and injected
• Configuration of surface gathering and
injection system
– Note types and locations
– Lengths and ID’s of tubulars through
which fluids flow
 Simulation Grid
• Primary grid specification

• Local grid refinement specification

 Simulation Grid
• The simulation grid is the definition of how
we divide, or discretize, space in order to
solve the differential equations numerically
• Although we use reservoir and well properties
in designing the simulation grid, the
simulation grid is independent of the
reservoir and well properties.
Primary Grid and Local Grid
Refinement
Common Grid Coordinate
Systems Include
• Cartesian
• Cylindrical
• Curvilinear (including stream-tube)
• Corner point
 Selection of Optimal
Grid Block Sizes
• Optimal grid
– Results in the desired level of accuracy
– Properly represents the reservoir geology,
and
– Has the lowest computer memory and time
requirements to solve the problem.
Effects of Grid Block Sizes on
Simulation Results
• FD solution approaches PDE solution as
∆ x approaches 0
• Use small blocks where convergent flow
• Use small blocks along displacement
fronts
 Common Rules for
Constructing a Grid
• Logarithmic spacing in cylindrical grids.
• Adjacent blocks increase in length by no more
than factor of 3.
• No more than 10 to 20% of the total pressure
drop should be between any two adjacent grid
blocks.
• Large changes in elevation or thickness
should be distributed over multiple grid
blocks.
 Common Rules for
Constructing a Grid
• Reservoir flow units should, in general, be
separated by grid block boundaries.
• Vertical discretization should be fine enough for
accurate accounting of gas percolation and
migration and/or gravity over-ride or under-ride.
• If in doubt, halve the grid blocks (double the
number of blocks) in one or more dimensions.
Effect of Grid Block Sizes on a Linear Buckley-
Leverett Displacement Simulation With a Sharp
Displacement Front and Mobility Ratio of 50.
1000

100
DX=22.5
DX=180
10
Water/Oil Ratio

0.1

0.01

0.001
0.7 0.8 0.9 1 1.1 1.2 1.3 1.4

Pore Volumes of Water Injected

Structure Map and Reservoir Simulation Grid
X

0
20
1234567 8910
1112
1314
116
5 171819 2021 22 23 24 25 26 27

12
Y 1 1
2
7 5 Block VI 2
3 3
4 Block V 4
5 6 4 5
6 1215 6
7
8 8 7
8
9 1107 9
10 3 10
11 9 10 11
12 12
13 13
14 1263 14
15 15
16 16
17 17
18 18
19 19
12
20 20
21 21 700
22 22

OW
23 23
24 24

12
25 25

80
26 2 26

0
27 27
28 28
29 29
30 30
31 31
32 32
33 1 33
34 34
35 35
36 36

OW
37 37

C
38 38
39 39
40 40
41 41
42 42
13000
43 43
44 44
45
45
46
46
OW

47 Legend 47
C

existing production wells

3
true fault traces 48
48 fault traces in simulation model

1234567 8910
1112
1314
116
5 171819 2021 22 23 24 25 26 27
Simulation Model Horizontal Permeability
Distribution for Simulation Layer 1
hist33k - X-Perm eability (m d )
700

600

675 : 5 1
1101 : 8449
1 : 6 1 394 : 6 1

1063 : 11 1

677 : 6 1 506
1222
674 : 3 1
: 8: 11 1
510 : 4 1
1006 : 3 1
500
1140 : 1 1 1155 : 2 1
671 : 2 1 516 : 5 1 973 : 4 1
465 : 4 1 453 : 3 1 422 : 7 1
196 : 1 1 1004 : 6 1

619 : 4 1 571 : 5 1 631 : 41254

1 : 1 1
987 : 1 1 400
297 : 3 1

300

200

100

0
12/01/1957 00:00:00 0.0000 days
Simplification of Reservoir
Geometry
• Example situations

• Using symmetry to simplify the grid

• Using pseudofunctions

• Checking validity of assumptions

Using Symmetry to Simplify the Grid

• 1/4th of the drainage area containing a well with

a hydraulic fracture can be simulated
• Repeated 5-spot water flood pattern can be
modeled using 1/8th of the pattern
• Any time a well is centered in a
homogeneous drainage area and forces on
either side of the lines of symmetry are
identical.
Using Pseudofunction to Model
3D Systems With 2D Model
• Pseudo-relative-permeability and pseudo-
capillary-pressure functions can be used to
simplify simulations of displacement
process in layered reservoirs with no
crossflow
• Dynamic pseudo-functions can be generated
from cross-sectional simulations.
Checking Validity of Assumptions
Made in Simplifications
• Any time a simplification is made, it
should be validated by run(s) without
simplification.
Representing Wells in Simulation

• Well representation should

– Account for near well conditions
– Allow the necessary degree of well
control
Well Constant Equations

Pwell

Pcell

 b kr 
q = WC   ( p well − pcell )
 µ 
 Well Constant Equations

• The general formula presented by

Peaceman for the well constant of a well
located in the center of a rectangular grid
block is given by

0.00708kh
WC =
ln( ro rwa )
 Well Constant Equations
• The apparent wellbore radius, rwa , includes
the effect of the skin factor due to damage,
stimulation, partial penetration, and
wellbore inclination. The equation for rwa is

rwa = rw e-s
 Spacing of Wells in the
Simulation Grid
• Spacing dependent on objectives, accuracy
and flow mechanism

• Rule of thumb is 3 to 5 blocks between wells

• In simulation of waterflood, may need 10 or

more

• Interference tests need finer grid spacings

 Spacing of Wells in the
Simulation Grid
• In early stages of study, perform a grid
sensitivity study and select the most
coarse grid
• As study progresses, refine grid
appropriately
 Well Rate and Pressure
Specifications and Constraints
• The well constraints used in simulations
should represent actual field operating
conditions as closely as possible.
• Most simulators will allow specification of
production targets and constraints enabling
the rate and minimum flowing pressure to
be specified.
• The simulator will automatically switch from
rate-controlled production to pressure-
controlled production.
 Timestep Schedule
• The timestep schedule is the definition of how we
divide, or discretize, time in order to solve the
differential equations numerically
• Although we use reservoir and well properties in
designing the timestep schedule, the timestep
schedule is independent of the reservoir and well
properties.
Scheduling Changes in Production
Rates and Other Dynamic Data
• Schedule of times to honor changes in
production and injection rates

Rate

Time
Scheduling of Individual
Timesteps

Rate

Time
 Effects of Time Step Size on
Simulation Results
• Several factors must be considered when
selecting time step sizes for a simulation.
These factors include:
– Data requirements,
– Numerical stability and
– Time-truncation errors.
Common Rules for Time Step
Size Control
• Recommended procedure:
– Start with small time step sizes after
significant rate changes
– Use large time step size multipliers to
build time step sizes quickly
– Use ∆ p, ∆ S and ∆ t limits to restrict
time step sizes to reasonable levels
 Simulator Control Data
• Numerical solution parameters

• Output control data

• Run control
 Output Control Data
• Moderate to large simulations require
judicious specification of desired output.
Output files sizes can quickly become
unmanageable.
• Type of output desired
• Frequency of node output
• Frequency of map output
 Run Control

Rate

Time
 Model Initialization
Pg = f(ρ g)

Pcgo = Pg - Po
Subsea Elevation

Pcgo = 0
GOC

Po = f(ρ o)
Datum
Pcow = Po - Pw

Pcow = 0
OWC

Pw = f(ρ w)

Pi

Pressure
 Model Initialization
1 Sg 0

Pcgo
Pcgo
Subsea elevation

0 Sg 1
GOC
GOC Pceow

OWC OWC
Pcow

0 Sw 1
Pc
0 Sw 1
Complete 3D View of the Simulation Model
(Mid-Point Elevations of Simulation
Gridblock, ft Sub-Sea)
hist33k - Midpt Elevations (ft ss)
-11500

-11600

465
465 :: 4
3 1
3
671 : 2 1
677 : 6 1
-11700
1140 : 1 1 1130 : 3 3 1101 : 8 1
453
453 :: 3
2 1
3 506s
506 :: 48 31 449 : 6 1
1222 : 1 1 : 3 1
506 : 4 6743 1063 : 11 1
196 : 1 1 677 : 7 9 1155 :1101 2 1: 1 9 : 2 5
449
619 : 4 1
619
619 :: 3
2 3
5 196 : 3 5 506s : 1 9 1063 : 9 5 -11800
1 506
96s: 9: 13
677 13 : 1 9 510 : 4 1
506 : 7 11 674
674s : 4: 19 91063 : 1 9
196 : 8 11 677: 5: 81 15
571 510 : 1 :5 3 1
506 : 1155
6 13 : 3 9
516 : 5 1
1006
571 : 3 5
506 : 5 15 516s : 2 973
5 : 4 1

571 : 1 9
516 : 3 5
1004 : 6 1 1063973
510 : 5 9
510
: 4 17: 9 11 -11900
510: :2 8513
675s : 2 15
1004
516 : : 41 59 510 : 7 15 394 : 6 1
675s : 1 17 394 : 5 3
631 : 4 1 71215
422 : 1127 : 19 9
1 : 4
297 : 3 1 995 : 1 17
631 : 2 5 394 : 2 9 -12000
1254 : 1 1 422 : 1 91102 : 2 5
1102 : 1 5

987 : 1 1
-12100

-12200

-12300

-12400

-12500
12/01/1957 00:00:00 0.0000 days
Simulation Model Initial Water Saturation
Distribution for Layer 1
hist33k - W ater Saturation
1.0

0.9

675 : 5 1
1101 : 8449
1 : 6 1 394 : 6 1

1063 : 11 1
0.8
674 : 3 1 510 : 4 1
677 : 6 1 506
1222
: 8: 1 1 1006 : 3 1
1140 : 1 1 1155 : 2 1
671 : 2 1 516 : 5 1 973 : 4 1
465 : 4 1 453 : 3 1 422 : 7 1
196 : 1 1 1004 : 6 1 0.7
619 : 4 1 571 : 5 1 631 : 41254
1 : 1 1
987 : 1 1
297 : 3 1

0.6

0.5

0.4

0.3

0.2
12/01/1957 00:00:00 0.0000 days
Reviewing Simulation Output to
Ensure Valid Results
• We recommend using the following
procedure to systematically review
simulation output to ensure that results
are valid.
• The full review will normally be required
following major changes in the model.
Some items should routinely be reviewed.
 Review Warning Messages
• Review output for warning or error
messages printed out by the simulator
– Fluid property, relative permeability,
and capillary pressure table trend
checking
– Grid blocks with zero permeability or
porosity
– Wells located in inactive grid blocks
– Other simulator specific warning/
error messages
 Review Input Data Printout
• Review output to ensure that the simulator
is correctly reading the input data.
– Rock properties
– Fluid properties
– Multiphase properties
 Review Initialization
• Review reservoir initialization/equilibration as
calculated by the model
– Compare original fluids in place in the
model to values estimated using
volumetrics
– Pressure
– Saturation
– Position of gas-oil, oil-water, gas-water
contacts
Review Numerical Performance
Statistics
• Material balance error
• Outer iterations
• Inner iterations
• Cutbacks
• Maximum pressure and saturation
changes
• Is the model oscillating?
 Review Production Statistics
• Ensure that wells are being operated in the
desired manner
• Check
– That each well is in the desired location
– That correct algebraic sign is used for
production and injection
– For wells changing from constant rate to
constant pressure
 Review Production Statistics
• Check
– For wells which have been shut in due to
excessive gas or water production, GOR or
WOR
– Total production from multiply completed
wells
– For wells drilled or completed by an
automatic well management scheme
 Integrated Reservoir Study
DYNAMIC MODEL
Well/Facilities Near Wellbore
Model Performance Models

Upscaled Preliminary Calibrated

Reservoir Dynamic Dynamic
Model Model Model

Optimized
Projected Field
Reservoir
Reservoir Economic
Development
Performance Model
Plan
 References
1. Mattax, C. C., and Dalton, R. L.: Reservoir
Simulation, SPE Monograph Series No. 13,
1990.
2. Aziz, K., and Settari, A.: Petroleum Reservoir
Simulation, Applied Science, 1979.
3. Odeh, A. S.: "Reservoir Simulation...What is
it?" JPT (Nov. 1969) 1383-1388.
4. Coats, K. H.: "Use and Misuse of Reservoir
Simulation Models," JPT (Nov. 1969) 1391-
1398.
 References
5. Coats, K. H.: "Reservoir Simulation: State of
the Art," JPT, (Aug. 1982) 1633-1642.
6. Satter, A., Frizzell, D. F., and Varnon, J. E.:
"The Role of Mini-Simulation in Reservoir
Management," paper presented at the
Indonesian Petroleum Association
Nineteenth Annual Convention, Oct. 1991
7. Kyte, J. R., and Berry, D. W.: "New Pseudo
Functions to Control Numerical Dispersion,"
SPEJ (Aug. 1975) 269-76.