Reservoir Modelling & Simulation

(PDB3053)

Chapter 1 : Introduction to Reservoir Modelling

& Simulation

Juhairi Aris Bin Muhamad Shuhili

Block L-01-01 Please kindly refer to this info
+605-3687113 regarding myself, the info in the
later chapters are obsolete.
[email protected]
Course Background
Lecturers
Juhairi Aris
Reference Material
• Ertekin, T., Abou-Kassem, J. H., & King, G. R. (2001). Basic Applied Reservoir Simulation. Texas:
Henry L. Doherty Memorial Fund of AIME, Society of Petroleum Engineers .
• Khaled, A., & Settari, A. (1979). Reservoir Simulation. London: Applied Science Publisher LTD.
• Fanchi, J. R. (2006). Principles of Applied Reservoir Simulation. Oxford: Gulf Professional Publishing
• Carlson M. R. (2006). Practical Reservoir Simulation. Oklahoma: PennWell Corporation
 Assessment Breakdown Assessment Marks (10%)
Active Learning : Brainstorming 1
Lab Assignment (20%) (Simulator: Eclipse 100)
• A group of 5 members where the students are free to form their own group. Active Learning : 2 minutes paper 1
• Students with no group will be formed into a group or assigned to an existing Active Learning : Think Pair Share 1
group as the 6th member.
• The lab session has no attendance. Please book your lab hours with GAs of the Active Learning : Note Checking 1
semester.
• Book a slot for your lab presentation with your respective GA. Only email Active Learning : One Final Question 1
softcopy of your lab presentation in pptx. format and all your lab work (.data, Student Self Reflection Tool (SSRT Cycle 1) 1
.include, excel) to your respective GA and cc me. There is no need for any
report. Course Outcome Survey (COE Cycle 1) 1
Student Self Reflection Tool (SSRT Cycle 2) 1
Test 1 (15%) (Chapter 1 -3) (1 hour 30 minutes)
• The first questions are theoretical questions worth 25 marks Course Outcome Survey (COE Cycle 2) 1
• The second question is 1-D, Dirichlet, Homogenous with parameters linked to your ID
worth 25 marks Attendance 1

• For the active learning, SSRT and COE please upload your PDF softcopy of
Test 2 (15%) (Chapter 4 -6) (1 hour 30 minutes) the in-class activity. Please scan the doc using Microsoft lens or other
• The first question is derivation question worth 25 marks scanning tool. Please don’t send pic. Name the softcopy with your ID.
• The second question is 1-D, Neuman, Homogenous and 2-phase worth 25 marks • For SSRT and COE, just upload the screenshot of your completion (hide the
score if you wish). Name the pic after your ID.
---------------------------------------------------------- • Each AL activities must have minimum of 100 words handwritten.
Total coursework 60% Final Exam Questions • Students who is willing to be the class rep will be awarded an additional of
1. Theoretical questions with no calculation 2% for coursework. Your main task is to collect attendance only during
Final exam 40% 2. Similar to Test 1 question but Neumann lecture and upload the attendance in the designated google drive link and
3. Derivation question similar to Test 2 Question monitor them.
4. Normalization and denormalization • For the grading, any submission within the criteria will get full mark since
they are considered as the reward for formative assessment. For attendance,
above 90% get the mark.
Ground Rules
• You can eat, sleep, use hand phone or PC during any sessions as long as you don’t make noise.
• Please join the class no matter how late you arrive.
• You can take a break outside if you need. No need to ask my permission.
• Please do not record any audio or video of my lecture. However, you can snap the picture of the whiteboard.
• I shall conduct no pop-quiz. All the assessment dates will be notified.
• All the replacement lectures, tutorials and lab sessions do not have attendance. Attendance is only compulsory for
lectures at designated hours.
• Students with less than 90% attendance (without valid reason) will be warned and barred.
• Students who miss any assessment (without valid reason) will have no make-up assessment.
Ground Rules
• There will be no extension on any assessment deadline (zero mark) without valid reason.
• Plagiarism will result in zero mark.
• Please do not chat during the lecture and put your mobile phone to silent mode.
• Meeting with lecturer can be done at any time from 7 to 4 pm (except 12 to 2 pm).
• Any queries please email or WhatsApp me.
• Error Carry Forward (ECF) will not be considered during assessment evaluation.
• Please never hesitate to request for extra marks.
• Presentation slide must be in PPT format. Group assignment must be submitted by group leader via email only
which you can find in the first page. Submission in PDF format or any other format will result in 0 mark. Do not
submit any hard copy.
• If I am late to lecture for more than 15 minutes, you are allowed to leave the class.
• Test and final questions will only be asked from something that I have thought you. Not something out of blue.
• All the formulae including conversion units will be given for test and final. Feel free not to memorize any formula.
• For explanation questions, no need to memorize my words or words in the slide. As long as your explanation carries
the meaning, you get the mark.
Ground Rules (Test)
• Please read the rules & regulations religiously before the test. Please ask if you do not understand.
• Plagiarism is strictly prohibited. Caught conducting so will result in 0 mark for Test. Plagiarism includes but not limited to bringing notes,
any electronic devices (except for non-programmable calculators), looking at your friend and etc. All the notes and devices must be placed
in your begs and put your bag in front.
• Please do not start writing or even holding your stationery before you are allowed to do so. Failed to abide by this rule will result in 0
mark.
• You are not allowed to leave the hall even to toilet. Please ensure you are physically fit for the 2-hour test.
• Once you are being told to put down your stationery after the test ends, you must put them down or 0 mark will be given.
• I truly mean every words written here. Please write your matric ID (below) once you complete reading the rules and regulations.
Course Outcomes
1.Analyze and validate different inputs required for reservoir simulation including the sources of
input.

2.Apply spatial and temporal discretization in diffusivity equation applied to various type of
reservoir simulation calculation.

3.Apply different type of matrix solver computation in simulation of Black Oil model including IMPES.

4. Conduct full-field simulation study using a commercial simulator

Pre-Requisite
• Computational Method (Compulsory)(Matrix Solvers)
• Reservoir Rock & Fluid Properties (Recommended)
• Reservoir Engineering I (Recommended)
• Reservoir Engineering II (Recommended)
Academic Week List of Activities
1 Chapter 1
2 Chapter 1, 2 & Commencement of Lab Session
3 Chapter 2 & 3
4 Chapter 3
5 Test 1 (During class hour)
6 Chapter 4
7 Chapter 4 & 5
8 Chapter 5 & 6
9 Chapter 6
10 Test 2 (During class hour)
11 Lab Project Presentation (Submission of PPT slide and simulation data file)
12 Lab Project Presentation (Submission of PPT slide and simulation data file)
Lesson Plan (Active Learning Activities)
Activities Week
Brainstorming (Advanced Organizing) 1
Two-Minutes Paper (Closure Focused Discussion) 1
Think Pair Share (Intermittent Discussion) 1
Note Checking (Intermittent Discussion) 1
One Final Question (Closure Focused Discussion) 1
Brainstorming (Reservoir Simulation Input &
Output)
• No introductory lecture will be delivered on the topic
• Group of 5 students will be formed randomly based on seating position
(pic)
• Students are required to brainstorm on the topic
• 3 minutes will be given for students to brainstorm on the topic (pic)
• Any group will be called forward to explain their thoughts (pic)
Two-Minutes Paper (FDP Overview)
• Introductory lecture will be delivered on the topic
• Every student will be required to bring out a sheet of paper
• 3 minutes will be given for students to summarize the important aspect of
the topic (pic)
• Based on the summary, rectify any misunderstanding of students in the
next class.
Think Pair Share (Heterogeneity vs. Anisotropy)
• Introductory lecture will be delivered on the topic
• A question on the topic shall be posed to students
• Students have to write down their response individually for 2 minutes
• Students who sit next to each other shall share their thoughts for 3 minutes
and arrive at a conclusion (pic)
• Two of the students will be called forward to share their learning (pic)
Note Checking (Models Selection)
• Introductory lecture will be delivered on the topic
• Every student shall be instructed to jot down notes from lecture (delivery)
and presentation slide.
• After 10 minutes,
• 3 minutes will be given for students to summarize the important aspect of
the topic (pic)
• Based on the summary, rectify any misunderstanding of students in the
next class.
One Final Question (Boundary Conditions)
• Introductory lecture will be delivered on the topics
• In 3 minutes, ask for 3 students who would like to ask any questions
pertaining to the topic and instruct them to write down the questions in a
piece of paper.
• Explain the questions to the whole class
Reservoir Simulation
• Reservoir simulation is an important and useful tool in reservoir management. Reservoir
engineer can utilise it under different operating conditions to predict the reservoir
performance. This is the main advantage of reservoir simulation. Prediction of performance
accuracy is really important because oil and gas project usually costs huge capital.
• Reservoir simulation is a perpetual process which must be done at all stages.
• Reservoir simulation is done by solving the partial differential equations for single or
multiphase flow using complex numerical method.
• Reservoir simulation is divided into 2 main branches which are history matching and
performance forecasting.
• Reservoir simulation is defined as the process of using the behaviour of a model of the
reservoir to represent or approximate the behaviour of the true reservoir.
Reservoir Simulator
Components of a Reservoir simulator
1. Type of Simulation Model
2. Fluid Model
3. Rock physics function
4. Static Model
5. Initial Condition
6. Dynamic Model
Reservoir Simulator
Reservoir Simulator
Reservoir Simulator
Mathematical Model
• A mathematical model is described by a set of partial differential
equations (PDEs) which describe mass transport in region occupied
by the reservoir together with initial and boundary conditions.
• The set of PDEs plus the initial and boundary conditions is referred to
an initial boundary value problem.
• If the model is sufficiently simple, we may be able to solve the IBVP
analytically
• For most problems, of interest, however, the IBVP cannot be solved
analytically and thus we use a numerical model (or simulator),
which is based on the application of numerical methods to obtain an
approximate solution of the IBVP
 Input & Output (Active Learning: Brainstorming)
1. Reservoir properties 1. Production profile
• Permeability & porosity
• Net to gross • Water oil ratio
• Dimensions & geometry • Gas oil ratio
• Rock compressibility
• Production rate
2. Fluid properties (PVT)
• Oil and gas viscosity, oil density, gas gravity, & Z-factor 2. Saturation profile
• Fluid Formation volume factor • Oil saturation
• Solution gas ratio
• Fluid compressibility • Water saturation
3. SCAL Properties • Gas saturation
• Relative permeability
3. Flowing bottom pressure
• Capillary pressures
4. Past production history 4. Tubing head pressure (if lift
• Reservoir pressure curve is provided)
• Water oil ratio
• Gas oil ratio 5. Composition of reservoir fluid
• Production rate 6. Reservoir and fluid properties
5. Fluid contacts
• Oil water contact
• Gas oil contact
• Initial fluid saturation
• Initial reservoir pressure
6. Exploration, appraisal or development (brown field) trajectory and coordinate
Supremacy of RMS!!!!!!!!!!!!!!!!
RMS in Field Development Plan
 FDP (Overview) (Active Learning : Two-Minutes Paper)

Well logging Core analysis data Poro- Detail fluid Well Test Report (Fault
RFT/MDT
interpretation Per, Rock Typing, Rock studies distance, skin, permeability)
Contacts
Compaction

Top structure Planimetering (Volume Volume from Input data for dynamic
map calculation) Static model model

• Facies Studies Revise static History-match in

• Depositional model Dynamic modeling
Environment Relative permeability
• Petroleum System Well stratigraphic curves, capillary
correlation (exploration pressure, SCAL report Surface Health, Safety
Facilities and Environment
& appraisal well)
Production
• DST Matching
Drilling program
• Upper Completion
• Offset Well Analysis Development
(VLP, Artificial Lift,
• Well Planning Strategy
Lift Curve)
• Drilling Cost & Time
• Lower Completion
• Casing Seat Selection & Casing Design
(IPR, Sand Control)
• Drilling Fluid and Cementing
• Completion Cost &
Economic
Time
Compare in-place Evaluation
Dynamic Model
volumes with
model Initialization
static model

Sensitivity Analysis Maintained

• Properties distributions. Modified
• Upscaling effect
Not used
• PVT Properties
Heterogeneity vs. Anisotropy (Active Learning :Think Pair
Share)
• Heterogeneity refers to the variation of rock and fluid properties with
respect to location.
• Anisotropy refers to the variation of permeability with respect to
direction. The vertical permeability is usually lower than horizontal
permeability due to variation in stratigraphy is more prominent
vertically. The usual permeability anisotropy, kv/kh is ranging from 0.1
to 0.3
• In real case, a reservoir is always heterogeneous and anisotropic.
• Reservoir simulation is the only technique in modelling fluid flow
which considers heterogeneity and anisotropy. This is another
advantage of reservoir simulation.
Heterogeneity vs. Anisotropy (Active Learning
:Think Pair Share)
Material Balance vs. Reservoir Simulation (1-6)
Pressure vs. Potential
• In 1-D flow horizontal flow, the effect of gravity is often neglected.
• However, in vertical flow, the effect of gravity plays an important role
in determining the direction of flow.
• The resultant direction of flow can be computed by taking into
consideration of pressure difference and gravitational force. The
combined term of pressure difference and gravitational force is
known as potential term.
Key Steps in a Simulation Study
1. Clear Objectives (What to simulate)

2. Reservoir Characterization

3. Model Selection (Black

Compare & Adjust

or Compositional)

4. Model Construction (Number of

grid cells, size of grid cells and
no of layers)

5. Model Validation (History matching)

6. Predictions (Forecasting)

7. Documentation 28
 Model Selection (Active Learning: Note Checking)

The Black Oil Model: • The Compositional

Model:
Reservoir processes that can be modelled
using the black oil model include:

1. Modelling dry gas, black oil and wet gas. • Reservoir processes that can be
2. Immiscible displacement modelled using a compositional model
3. When there is no significant change of include:
composition in a reservoir.

1. Retrograde condensate reservoir &

volatile oil
2. Miscible injection
3. Enhanced oil recovery
4. When there is no significant change of
composition in a reservoir. 2
 Model Validation Predictions (Forecasting)
• History Matching must not be achieved at the • Prediction cases shouldn’t exceed capabilities of
expense of parameter modifications that are the simulator.
physically and/or geologically wrong.
• Predictions need to be consistent with field
practices (based on PSC/cannot be forever).
• Even when a model is fully validated,
simulation results will still have some degree • Error in simulation can be due to:
of uncertainty. Thus, the process of model
validation is continuous throughout the  Input error
production period.  Modeling or mathematical constraints
because of compromises made in model
selection.
 Scaling-up

3
Boundaries Condition (Active Learning: One Final Question)
• Dirichlet boundary condition refers to constant pressure at the boundary.
This type of boundary condition usually arises when there is a continuous
support from aquifer in maintaining reservoir pressure.
• Neumann boundary condition refers to the condition of constant
pressure gradient (C) across the boundary. Usually, the C = 0. This means
there is no flow across the boundary. This happens when the reservoir is
bounded by sealing fault.
Also
known as
Robin’s
Boundary
Condition
Spatial Discretization
• The reservoir static model built must be first discretized before simulation.
• Discretization of a reservoir refers to dividing the reservoir into a number of grid blocks.
The range for the number of grid blocks can be thousands to millions. Grid blocks are
also known as cells.
• The number of grid blocks is limited by the simulator capacity, time constraint and cost
limitation. The smaller the size of grid block, the larger the number of grid blocks. The
larger the number of grid blocks, the longer the time needed to complete the
simulation.
• The number of grid size is often related to accuracy (not always).
• The discretization of reservoir often considers the simulator capability, time, cost and
also the necessity of the gridding size.
• Gridding size is usually made smaller nearby a fault and nearby a source/sink because
the pressure change in the vicinity of a the regions is prominent. Thus to accurately
simulate the pressure change, the gridding must be smaller to attain higher accuracy.
Spatial Discretization

• Each grid-block in a model has only one set of properties; there is no variation
in any property within a block.
• For example, pressures and/or phase saturations in a model block will be
volumetric averages of the saturations in that part of the reservoir represented
by the block.
• A model block can be visualized as a well-stirred tank (i.e.., its contents are
homogeneous) connected to adjacent tanks with pipes whose flow capacities
are determined by reservoir flow properties.
Gridding System (Point vs. Block)
• There are 2 types of gridding system which are known as block-centered and point-
centered. In block-centered, grid blocks with known dimensions are superimposed over the
reservoir. For a block-centered system, the grid points are defined as the centres of the grid
blocks.
• In a point distributed grids, grid points are distributed over the reservoir before block
boundaries are defined. For point, block boundary is placed halfway between 2 adjacent
pressure points. Historically, reservoir simulators have used block-centered grids because
the volume associated with each grid point is clearly defined.
• Point-centered (or point-distributed) grid has its boundaries coincident with the exterior
boundaries of the system. Point-centered gridding system is also known as lattice-centered.
Block-centered grid has its boundaries one-half grid block away from the exterior
boundaries.
• The two grid systems do not produce the same grid points unless we choose equally spaced
grid points separated by a uniform space ∆x. Note that we do not need to consider equally
spaced grid-points or grid blocks.
• Only block centered grid abides by material balance.
Gridding System (Point vs. Block)
• Actually, it is more appropriate to use smaller grid spacing near the source/sink
locations. The similar reason as in near fault region, more accurate pressure
monitoring.
• Block-centered & point-centered system are suitable to handle Neumann and
Dirichlet problem respectively. Point distributed grid systems have numerical
advantages when non-uniform spacing is used. In particular, a point-distributed-grid
results in a consistent finite difference operator on an non-uniform grid whereas a
block-centered grid does not.
• Historically, the block-centered grid system has been the most commonly used grid
system in petroleum reservoir simulation. This is because the block-centered grid
system adheres more closely to the material-balance concept used in reservoir
engineering.
Up-scaling/Averaging
• The severity of heterogeneities needed more number of cells to represent the static model.
Thus, the variation in the properties of static model such as permeability can be catered for.
To combine a set of blocks with different permeability into a single block with a single
permeability. Usually, arithmetic and harmonic averaging are used.
• As aforementioned, the number of cells is usually proportional to the cost and time. Thus,
this set up a barrier to exactly imitate the static model.
• Thus, permeability from different locations (laterally or vertically) must be averaged to find
the most representative value for the whole reservoir. Usually, with well test analysis, a
representative permeability value can be obtained. However, if the permeability data is
obtained from core sampling, then averaging process must take place.
• There are 5 types of permeability averaging such as arithmetic, harmonic, geometric, log
mean average (only end points are known) and upstream relative permeability (multiphase
flow). The type of averaging depends on the bedding of the formation. Arithmetic and
harmonic are more common compared to the rest.
Up-scaling
Temporal Discretization
• A simulator also divides the production period into smaller time steps.
• Changes in a reservoir (pressure, saturation, etc.) are computed over
each time step.
• Conditions are defined only at the beginning and end of each time-step;
nothing is defined at any intermediate time within a time interval.
• The accuracy with which reservoir behaviour can be calculated generally
will be influenced by the length of the time steps as well as the number
of grid blocks.
Temporal Discretization
• For any size grid-block and any length of time step, there will always be
abrupt changes in reservoir conditions from one block to the next and
from one time-step to the next.
Temporal Discretization
• A plot of saturation vs. time in any block in the model would be a "stair-
step" curve
–A time/saturation plot for the corresponding location in the reservoir
would be a continuous, smooth curve.
–The stair-step approximation approaches actual reservoir saturation
distribution as the sizes of grid-blocks and time-steps decrease.
• An important step in simulation clearly is the selection of the number of
blocks in a model and the time-step size to use in calculating the
reservoir response.
Learning Outcome
• Students are able to discuss the importance of reservoir simulation
• Students are able to understand the workflow in reservoir simulation
• Students are able to perform spatial and temporal discretization
• Students are able to derive diffusivity equation
 Arigato
Hozaimas
Minasan