Reservoir Simulation AND Modeling Updated Version 1 (Auto

Recovered)
Petroluem economics (Madonna University Nigeria)

Scan to open on Studocu

Studocu is not sponsored or endorsed by any college or university

Downloaded by Chinedu Kanu (kanudaniel866@gmail.com)
 nCHAPTER ONE
OVERVIEW OF MODELING METHODS

1.1. Introduction
Generally simulation is the utilization of a model to obtain insight into the
behavior of a
physical process.It is a process or mechanism by which a particular problem
can be
studied in a varying depth of details to obtain answer or to confirm a hypothesis.
However in petroleum engineering,the area of reservoir simulation combines
physics,
mathematics,reservoir engineering and computer programming to develop a
tool for
predicting hydrocarbon-reservoir performance under various operating
conditions by
applying the concept of mathematical modeling.This involves the hydrodynamic
of flow
within the reservoir as well as the total production system which includes the
reservoir,
the surface facilities and any interrelated significant activities.Thus reservoir
simulation
is the process of using a model that is artificially created to represent the
reservoir in
order to predict hydrocarbon reservoir performance with respect to time.The
model is
built by combining several governing equations that describe the physical
processes
occurring in the reservoir during hydrocarbon production.The resulting equation
is a
non linear partial differential equation that is incorporated in a grid cell.
Fig 1.1-The production system

This document is available on

Downloaded by Chinedu Kanu (kanudaniel866@gmail.com)

1.2. Reservoir Modeling Approaches

The conventional methods of forecasting reservoir performance generally can

be divided
into four categories:analogical method;experimental method (analogy,and
physical
model),mathematical and reservoir simulation/numerical model.Fig 2 depicts
the
reservoir modeling approaches.

Reservoir Modeling Approach

Experimental

Analog model Physical model

Mathematical
Reservoirsimulation/Numerical model
Analogical

Analytical mode
Material balance model Decline curve model Statistical model

Fig 1.2-Reservoir Modeling Approaches

1.2.1 Analogical Method.This method uses properties of the mature reservoir

that are either geographical and petro physically similar to the target
reservoir to attempt to predict the performance of a target zone.The
method is used when limited data (production data)or no data are
available for the target reservoir.It can be used to
estimate recovery factor,initial production rate,decline rates,well spacing
and
recovery mechanism.This method is only suitable if the reservoirs are
operating under similar development strategy or recovery
mechanism.The major setback is
if the reservoirs are operating under different recovery mechanism and
what if
sensitivity cannot be investigated.

Downloaded by Chinedu Kanu (kanudaniel866@gmail.com)

1.2.2 Experimental method: this method measures physical properties
(rates,pressure,
and saturation)in laboratory models and scale results to the entire
hydrocarbon
accumulation.This method is divided into two,namely:analog and
physical
models.
· Analog: this method uses similarities between the phenomenon of fluid flow
through porous media and other physical phenomena to simulate reservoir
performance.Analog models based on the governing equations (see table
1.1)are built to represent the reservoir and appropriate quantities
(pressure and flow rate) are measured.These quantities are translated
through the governing equations into
their porous media analogs.Three analog methods-RC
network,potentiometer,
and Hele-Shaw model are commonly used.The RC network uses the
similarity/analog between fluid flow through porous media and electrical
flow to
model reservoir performance.Capacitance (porosity)is used to model
fluid
storage at a point in space while resistance(Permeability)is used to
model the
transmissibility between points.This model is used to model unsteady
state
behavior and is ideal for reservoir under primary depletion.However they
are
discrete models-the capacitance represent the storage at discrete points
in the
reservoir.Contrast to the RC network model,the potentiometer is
constructed
with a continuous electrical modeling material.Voltages are applied at the
well
sites and the voltage measurement can be made at any point.The model
can only simulate steady state flow.The major drawback of electrical
analog models is that
they must be custom built for a particular or individual reservoir,thus
making it
difficult to be used for other reservoirs.Also the performance of the RC-
network
model can be affected by malfunction of the electrical components
(capacity,
meters,and resistors).Furthermore the electrical analog models are bulky
and
occupy very large space.These models are limited to modeling single-
phase flow
in porous media.

This document is available on

Downloaded by Chinedu Kanu(kanudaniel866@gmail.com)

 Table 1.1-Physical Phenomena Analogous To Fluid Flow through Porous Media

Fluid Flow Through Fluid Flow Through Electricity Flow Throug

Phenomenon Porous Media Parallel Plates Circuitry Heat Flow by Conduction
Goveming equation Hagen-Poiseuille law Ohm's law, Fourier's law,
1=(1/R)△E

Properties Volumetric rate,q Volumetric rate,q Current,I Heat flow rate,Q

Transmissibility, Hydraulic conductance Electrical conductance, Thermal conductance,
1/8

Electrical conductivity," Thermal conductivity,

Fluid mobility,Hydraulic conductivity,
1/r K

Pressure,p Pressure,p Voltage,E Temperature,T

*Horzontal fow.

Fig 1.2-Electric circuit analog for a simple hydrocarbon reservoir /aquifer system

Fig 1.3-Electric circuit for determining the potential distributions in 2D flow system

Downloaded by Chinedu Kanu(kanudaniel866@gmail.com)

· Physical models: these are models used to make direct measurement of
flow
properties such as porosity and permeability to establish mechanism of
oil
recovery.Core flooding experiment,slim tube and sand packed are used in
this
process.The setback of this type of models is that the experiments are
conducted
at a scale that is not representative of actual reservoir.Consequently,the
result
must be scaled up to more representative scale.Another type of physical
model
uses geometrical-mechanical and thermal-similarity concepts.Here the
area
geometry,thickness,porosity and permeability of the model and fluid
properties are scaled such that the shape and dimension of the model
are the same as those in
the reservoir.Hence,the performance of this scaled model reflects that of
the
reservoir.Sobocinski and Corenelius single well coning model is an
example of
this model.The model can determine critical coning rates,water
breakthrough
time and post breakthrough water cut.However,since the reservoirs are
naturally
heterogeneous,it is impossible to scale all physical properties of the
reservoir,
therefore the use of scaled models is very limited.
 Fig 1.4-A laboratory coning model

This document is available on

Downloaded by Chinedu Kanu(kanudaniel866@gmail.com)

 1.2.3 Mathematical Model. These are one of the modern models use to predict
reservoir
performance.These include material balance,decline-curve,statistical and
analytical
(well test)method.

· Material balance method: this is a tank model that is a mathematical

representative of reservoir or drainage volume.The model is based on the
principles of conservation of mass which simply states that the amount of
material (gas,oil and water)remaining in the reservoir after a production
interval is equal to the amount of material originally in place in the
reservoir,minus the amount of
material removed from the reservoir because of production,plus the
amount of
material added to the reservoir because of injection and
encroachment.The model
can be used to estimate the initial oil in place,predict ultimate recovery
under
various drive mechanism,predict future performance.However,the
reliability of the model depends on the accuracy of available data and the
extent to which the
underlying assumptions are met.The model is often refers as zero
dimensional
because it does not take into consideration the spatial variation of rock
and fluid properties (homogenous reservoir),pressure changes with
space and time during
depletion,and the well location.
· Decline curve analysis: is an empirical model that uses the historical data of
a
reservoir to predict reservoir performance.The method includes
exponential,
hyperbolic and harmonic decline curve.Here the method uses historical oil
production rate versus time to extrapolate for future performance.The
principal assumption of this method is that all process occurring in the
past will continue. The model functions well when operating practices
are not expected to change in the future.If operational practice
changes,this method cannot be used.Thus the
method cannot be used for what if analysis.
● Statistical method:statistical approach uses empirical correlations that
are
statistically derived by use of past performance of numerous reservoirs to
develop
empirical correlations for reservoir prediction performance.The correlation
is
derived with data from mature reservoir located in the same region with
the same
litho logy(sandstone,carbonate)and operating under the same deriving
mechanism(water derive,solution gas derive).The model can be used
with
confidence if the reservoir properties are within the limits of regression
database used to develop the correlation.The model is limited to
estimating the ultimate
hydrocarbon recovery and cannot be used to develop a rate /time
relationship.

Downloaded by Chinedu Kanu (kanudaniel866@gmail.com)

 · Analytical model: this model is developed by combining the physical
processes
or governing equations that describe flow in the reservoir to derive a
theoretical
equation.The solution of the model is based on the exact solution of the
theoretical model.The equations are solved analytically using simplifying
assumptions to reduce the complexity.Examples of these models are the
pressure transient test and Buckley-Leverett.The simplifying
assumptions used in pressure
transient analysis include horizontal reservoir,uniform thickness,single-
phase
flow and laminar floe,while the simplifying assumptions used in Buckley-
Leverett analysis include incompressible linear flow,negligible
capillary,and
gravity effects.

1.2.4 Reservoir simulation/Numerical model.

This is the most modern and advanced method of modeling reservoir or

predicting the
performance of reservoir.The model is developed by combining several
governing
equations that describe flow in porous media.The governing equations can be
conservation of mass,transport equation/Darcy law,equation of state,heat
transfer, energy balance equation and mass transfer.The resulting equation
is a set of coupled
non linear partial differential equation which is solved numerically using finite
difference method and appropriate boundary conditions.A high speed computer
is used to solve the equations.The advantages of this method lie in the fact
that the least number
of simplifying assumptions are used for the reservoir heterogeneity,mass
transfer
between phase and mechanism responsible for flow.In addition,spatial
variation of rock
properties,fluid properties and relative permeability characteristics can be
represented
accurately in the reservoir simulator.Fig 1.5 depicts the major steps involved in
the
development.
 7

This document is available on

Downloaded by Chinedu Kanu (kanudaniel866@gmail.com)

 Formulation: Discretization Well representation
single/multiphase
flow

Linear
Recovery Non linear algebraic
Non linear
process algebraic equation equation
PDE

Validation Linearization

Numerical reservoir Pressure saturation

simulation process distribution/well rates

Fig 1.5-Major steps used to develop reservoir simulator

1.3. Classification of reservoir simulator

Reservoir simulators are classified according to the type of reservoir fluid,the

reservoir,
and the recovering process occurring in the subject reservoir.It can also be
classified
based on the co-ordinate system used in the model and the number of
phases.Reservoir
simulators commonly use in reservoir simulation includes black oil simulator,
compositional simulator,chemical flooding simulator and thermal recovery
simulator.

· Black Oil Simulator: the basic governing equations use to formulate the
simulator are conservation of mass,rate equation/Darcy law,and equation of
state.
The effects of mass transfer among the fluids (oil,gas and water)are
assumed to
be negligible and the fluid properties B.,Bg,and Rs are governed by PVT
behavior.The simulator is used in situation where recovery process is
insensitive
to compositional changes in the reservoir fluids.Black oil simulator is best
suitable for modeling of primary and secondary recovery process where
mass
transfer is negligible.

Downloaded by Chinedu Kanu (kanudaniel866@gmail.com)

· Compositional Oil Simulator: in addition to the governing equations used in
Black oil simulator,the effects of mass transfer and compositional
changes in the
reservoir fluid properties are considered.A cubic equation of state governs
the
PVT(Peng Robinson equation).The simulator is used to model primary
depletion
of volatile-oil and gas condensate reservoir as well as secondary recovery
like
water injection and also tertiary recovery such as miscible gas injection.
· Chemical flooding simulator: this simulator is used to model chemical
recovery of hydrocarbon such as polymer or surfactant flooding etc.The
simulator differs
from black oil and compositional simulator in that additional conservation
equations are used to track the individual chemical species used in the
flooding.
· Thermal recovery simulator: this simulator is used to model thermal
recovery
process such as steam floods and in-situ-combustion.The simulator uses
an
energy balance equation in addition to the mass balance equation.A
compositional approach is used in the simulator.

HYDROCARBON-RECOVERY
METHODS

PRIMARY
RECOVERY
Solution-Gas Gas-Cap Gravity Rock
Drive Expansion Waterdrive
Dralnage Expansion

SECONDARY
RECOVERY

Waterflooding Pressure
Maintenance|

TERTIARY
RECOVERY
Chemlcal Thermal Misclble
● Polymer Flooding ●Steam Injectlon ●Vaporizing-Gas Drive
●Surtactant Flooding ● Hot-Water Injection ● Enriched-Gas Drive
●Rovorse Wotting Agent ● In-Situ Combustion ● Alcohol Flooding
●Carbonated Waterflood ●Wellbore Heating ● CO₂Flooding
● Misclble
Hydrocarbon
Flooding

Fig 1.6-Hydrocarbon recovery methods

 9

This document is available on

Downloaded by Chinedu Kanu (kanudaniel866@gmail.com)

1.4 Basics Step of A Simulation Study
1.Set the objectives:this is the first step of any successful simulation study
-set
clear achievable objectives.These objectives must be compatible with
available
data and production history.The objectives are used to set goals,define
basic
strategy,identify available resources and determine what is to be learned from
the
study.
2.Acquire and validate all reservoir data: after the objectives have been
defined,
reservoir and production data are gathered.Only the data required to meet
the
objectives of the study should be incorporated in the simulation model.
Incorporating additional detail/data that does not add to understanding
the
objectives leads to overkill.
3.Construct the reservoir model:once the data have been gathered and
validated.
the simulator is built.Here,the reservoir is divided into grid block.All the
physical processes occurring in the reservoir or governing equations that
describe the physical processes now take place inside the grid
cells.Formation properties such as porosity,directional permeabilities
and net pay thickness are assigned to
these grid blocks.Each grid cells are unique in that they may have
different
reservoir properties;however,reservoir properties are assumed to be
homogenous
within a grid cell.All data are then properly scaled to the simulation grid
4.History match the reservoir model: after the simulation model has been
built,it must be tuned or history matched with available production data
because much of the data in the simulation model might be subject to
error during interpretation by
the engineers and geologists,hence they are not known for certain.
5.Run prediction case: this is the final step in the simulation process in
which various production schemes are evaluated and sensitivity
analysis of various
production and reservoir parameters are performed.
 Continuous Assessments
1.Define simulation in terms of petroleum engineering discipline
2.What are the different ways a reservoir can be modeled?
3.What are the differences between analytical,numerical and statistical
model?
4.Match the following:

Physical model simulator

conceptual model partial differential equation
10

Downloaded by Chinedu Kanu (kanudaniel866@gmail.com)

geological model material balance equation
mathematical model laboratory sand pack
computer model potentiometric model
analog model depositional model
numerical model empirical model
statistical model finite difference equation

5.Distinguish between black oil and compositional simulator.Explain the

governing
equations that apply in each case and the type of reservoir/recovery that
can be
modeled with the simulators.
6.Put the following in sequential order to summarize the basic steps of a
simulation
study:

Preparing data
constructing the geological data
defining the study objectives
history match
analyzing the results
reporting
predicting

7.The equation below represents the steady-state radial flow of a fluid in a

cylindrical porous medium.What are the analog representations
describing the
heat and current flows in similar cylindrical system?Identify the
analogous terms
and/or groups.

Determine the transmissibility of the reservoir if the thermal conductivity of the

material
;the thickness and area of the material are 45mm and
50m²respectively.

8.Comment on the accuracy of the statement"The material balance

equation is
considered to be a zero dimensional model because time dependency is
not
incorporated into it”.

11

This document is available on

Downloaded by Chinedu Kanu (kanudaniel866@gmail.com)