Petroleum Experts

REVEAL

October, 2003

USER GUIDE

The information in this document is subject to change as major improvements and/or

amendments to the program are generated. When necessary, Petroleum Experts will issue
the updated documentation.
The software described in this manual is furnished under a licence agreement. The software
may be used or copied only in accordance with the terms of the agreement. It is against the
law to copy the software on any medium except as specifically allowed in the license
agreement. No part of this documentation may be reproduced or transmitted in any form or by
any means, electronic or mechanical, including photocopying, recording, or information storage
and retrieval systems for any purpose other than the purchaser's personal use, unless express
written consent has been given by Petroleum Experts Limited.
All names of companies, persons or third party products contained in the examples in this
documentation are part of a fictitious scenario or scenarios and are used solely to document
the use of a Petroleum Experts product.

Address:
Petroleum Experts Limited
Spectrum House
2 Powderhall Road
Edinburgh, Scotland
EH7 4GB
Tel: (44 131) 474 7030
Fax: (44 131) 474 7031
Email: [email protected]
Internet: www.petroleumexperts.com

Registered Office:
Petroleum Experts Limited
Spectrum House
2 Powderhall Road
Edinburgh, Scotland
EH7 4GB

REVEAL

Introduction .................................................................................................................. 1
1.1 System Requirements .............................................................................................. 1
1.1.1 Hardware and Software Requirements ............................................................... 1

1.2

Overview of REVEAL ................................................................................................ 1

1.3

Contacting Petroleum Experts Ltd..................................................................................... 2

Step by Step Examples................................................................................................ 1

2.1 Getting Started - Step by Step Example................................................................... 1
2.1.1 Step 1 - Initialise New Case ................................................................................ 2
2.1.2 Step 2 - Wizard Basics ........................................................................................ 3
2.1.3 Step 3 - Control Section ...................................................................................... 4
2.1.4 Step 4 - Reservoir Section .................................................................................. 6
2.1.5 Step 5 - Physical Section .................................................................................... 9
2.1.6 Step 6 - RelPerm Section.................................................................................. 11
2.1.7 Step 7 - Wells Section ....................................................................................... 14
2.1.8 Step 8 - Initialisation Section ............................................................................. 15
2.1.9 Step 9 - Schedule Section................................................................................. 16
2.1.10 Step 10 - Run Simulation .................................................................................. 17
2.1.11 Step 11 - View Results ...................................................................................... 18

Using the Input Wizard ................................................................................................ 1

3.1

Input Wizard.............................................................................................................. 1

3.2 Control Section ......................................................................................................... 2

3.2.1 Solver Options..................................................................................................... 2
3.2.1.1
Both Implicit Solvers - First Screen of Solver Options .............................. 4
3.2.1.2
Both IMPES Solvers - Second Screen of Solver Options......................... 5
3.2.1.3
IMPES Solver ........................................................................................... 5
3.2.1.4
Corant Flux Limit implicit solver................................................................ 5
3.2.2 Model Options ..................................................................................................... 6
3.2.3 Component List ................................................................................................... 7
3.3 Reservoir Section ..................................................................................................... 7
3.3.1 Geometry............................................................................................................. 7
3.3.2 Cartesian Geometry ............................................................................................ 9
3.3.3 Radial Geometry ............................................................................................... 10
3.3.4 Curvilinear Geometry ........................................................................................ 10
3.3.5 Grid Depth ......................................................................................................... 12
3.3.6 Porosity ............................................................................................................. 12
3.3.7 Permeability....................................................................................................... 13
3.3.8 Transmissibility Multipliers................................................................................. 13
3.3.9 Absolute Transmissibility................................................................................... 13
3.3.10 Net to Gross ...................................................................................................... 14
3.3.11 Pore Volume Multipliers .................................................................................... 14
3.3.12 Rock Type ......................................................................................................... 14
3.3.13 PVT Region ....................................................................................................... 15
3.3.14 Equilibration Region .......................................................................................... 15
PETROLEUM EXPERTS LTD

REVEAL

3.3.15 Non-Neighbour Connections ............................................................................. 15

3.3.16 Grid Refinement ................................................................................................ 16
3.4 Physical Section ..................................................................................................... 17
3.4.1 Fluid Properties ................................................................................................. 17
3.4.2 Rock Compressibilities ...................................................................................... 18
3.4.3 Miscellaneous.................................................................................................... 18
3.5 RelPerm Section..................................................................................................... 22
3.5.1 General Options ................................................................................................ 22
3.5.2 Residual Saturations ......................................................................................... 24
3.5.3 Relative Permeabilities...................................................................................... 27
3.5.4 Capillary Pressures ........................................................................................... 28
3.5.5 Endpoint Scaling ............................................................................................... 29
3.6 Aquifer Section ....................................................................................................... 30
3.6.1 Analytical Aquifer............................................................................................... 30
3.6.2 Cater Tracy model............................................................................................. 31
3.7 Mobility Section....................................................................................................... 33
3.7.1 Polymer and Gel................................................................................................ 33
3.7.2 Original Polymer Gel Model .............................................................................. 34
3.7.3 Carreau Polymer Model .................................................................................... 35
3.7.4 Kuparuk Polymer Gel Model ............................................................................. 36
3.7.5 Gelation and Degradation ................................................................................. 37
3.7.6 Inaccessible Pore Volume................................................................................. 38
3.7.7 Foam ................................................................................................................. 38
3.8 Phase Section......................................................................................................... 39
3.8.1 Introduction........................................................................................................ 39
3.8.2 Alcohol/Polymer Partitioning ............................................................................. 41
3.8.3 Surfactant Phase Model (Micro-Emulsion)........................................................ 43
3.8.4 Ternary Diagram ............................................................................................... 51
3.8.5 Surfactant Interfacial Tension............................................................................ 52
3.8.6 Surfactant Viscosity........................................................................................... 53
3.9 Adsorption Section.................................................................................................. 55
3.9.1 Adsorption Properties........................................................................................ 55
3.9.2 Adsorption Isotherms ........................................................................................ 56
3.9.3 Permeability Reduction ..................................................................................... 59
3.10 Water Chemistry Section ........................................................................................ 60
3.10.1 CO2 and H2S Partition Coefficients .................................................................. 62
3.10.2 H2S Souring ...................................................................................................... 63
3.10.3 Scale Inhibition .................................................................................................. 64
3.11 Wells Section .......................................................................................................... 65
3.11.1 Location and Properties .................................................................................... 65
3.11.2 Fracture Properties............................................................................................ 68
3.11.3 Geertsma deKlerk 1D Thermal Fracture ........................................................... 71
3.11.4 3D Thermal Fracture ......................................................................................... 72
3.11.5 Rock Stress and Elasticity................................................................................. 76
3.12 Well-bore Heating Section ...................................................................................... 78

PETROLEUM EXPERTS LTD

REVEAL

3.13 Initialisation Section ................................................................................................ 79

3.13.1 PVT Initialisation................................................................................................ 79
3.13.2 Equilibration Initialisation................................................................................... 79
3.13.3 Trace Component Concentrations .................................................................... 80
3.13.4 Residual Saturations ......................................................................................... 80
3.14 Schedule Section.................................................................................................... 81
3.14.1 Well Schedule ................................................................................................... 81
3.14.2 Thermal Fracture Schedule............................................................................... 83
4

Grid Refinement ........................................................................................................... 1

4.1

Refinement Section .................................................................................................. 3

4.2

Reservoir Section ..................................................................................................... 5

4.3

Aquifer Section ......................................................................................................... 5

4.4

Wells Section ............................................................................................................ 5

4.5

Well-Bore Heating Section........................................................................................ 6

4.6

Initialisation Section .................................................................................................. 6

Menu Commands ......................................................................................................... 1

5.1

File ............................................................................................................................ 1

5.2

Options ..................................................................................................................... 2

5.3

Edit............................................................................................................................ 2

5.4

Input.......................................................................................................................... 3

5.5

Project....................................................................................................................... 3

5.6

Run Simulation ......................................................................................................... 4

5.7

Results...................................................................................................................... 4

5.8

View .......................................................................................................................... 5

5.9

Window ..................................................................................................................... 5

5.10 Help .......................................................................................................................... 5

5.11 Playback ................................................................................................................... 5

PETROLEUM EXPERTS LTD

REVEAL

External Data Import .................................................................................................... 1

6.1

Importing Overview................................................................................................... 1

6.2

Importing from ASCII ................................................................................................ 2

6.3 Importing from Eclipse (binary files) ......................................................................... 5

6.3.1 Data files Required.............................................................................................. 5
6.3.2 Rock Type Arrays and Compressibilities............................................................. 5

1 Introduction
1.1 System Requirements
REVEAL supports all Windows-certified drivers that are shipped with Windows. The
list of devices, software and hardware supported by Windows is included with the
documentation of your copy of Windows.

1.1.1 Hardware and Software Requirements

Minimum requirements recommended for REVEAL.
Pentium III class PC (Windows 95, 98, NT, 2000 or XP)
500 MHz processor (2GHz recommended)
256 Mbytes of memory (1Gbytes recommended)
5 Gbyte hard disc space for temporary files
1280 by 1024 minimum display size with high colour (16 bit) resolution
CD drive if the software is installed from a CD
Licenses - REVEAL can be run using a single user (stand-alone) license or on a
network.
In either case, a special security key is needed. The security key is called Bitlock for
stand-alone licenses, and Hardlock for network licenses. The security key is
provided by Petroleum Experts.
For a stand-alone license, the security key (Bitlock) must be attached to the parallel
port of the PC. For a network installation, the security hey (Hardlock) can be
attached to any PC communicating with the network. You should refer to the
separate installation procedure for a network Hardlock sent with the Hardlock
license.

1.2 Overview of REVEAL

REVEAL is the fifth member of Petroleum Experts Integrated Production
Management (IPM) suit of software.
REVEAL is a full field reservoir simulator concentrating on modeling mobility and
injectivity issues arising from injection of non-reservoir fluids at non-reservoir
temperatures. Therefore thermal and chemical effects are widely integrated in the
REVEAL models. In addition to standard simulator models, the features present
within REVEAL include:

2-2

CHAPTER 1 INTRODUCTION

PC Environment

Wizard Data Input

2D and 3D Graphics
OpenServer

IPM Integration

Thermal Fracturing

Hydraulic Fracturing
Water Chemistry

Chemical Adsorption
Polymer and Gel

Surfactant

Well-bore Heating
Analytical Aquifer
IMPES Solver

Implicit Solver

Grid Refinement

runs on PC (Windows 95, 98, 2000, NT) with single

interface to all functionality.
simplified data entry, verification and visualisation prior
to calculation.
may be viewed during simulations.
full OpenServer functionality built in, for automation or
batch processing of runs or connection to third party
software.
PROSPER lift curves, Petroleum Experts PVT matching
and GAP optimisation (GAP is able to control, through
choke pressure drops, individual completions in
REVEAL).
integrating thermo- and poro-elastic effects on
reservoir stress with injection profile and fracture
mechanics.
production fractures.
chemical equilibrium of mixed waters predicting
precipitation and dissolution, including H2S souring
and scale inhibition.
retardation and permeability reduction effects.
mobility changes for aqueous phase, including shear
thinning and permeability reduction
4th phase micro-emulsion model using ternary
diagrams.
microwave electrical heating of heavy viscous oils.
Carter Tracy model.
with flux corrected transport (fct) for models where
numerical dispersion is important (e.g. chemical
additives), and material balance iterations.
for high mobility and large throughput models (e.g.
coning).
general hexahedral refinement of master grid for
IMPES and implicit solver options.

An import facility is available to import part data sets from ECLIPSE, VIP or ASCII.
This will be extended to other simulators in due course.

1.3 Contacting Petroleum Experts Ltd

We encourage feedback and if you have problems or questions using REVEAL,
please send an e-mail to [email protected] with the following information.
1

Include the keyword 'REVEAL' in the e-mail subject.

REVEAL version and build number - use the menu option Help|About
REVEAL... to obtain this information.

Description of the problem or question.

Include a REVEAL archive (*.rvl) file where possible. Check that this file is
not too large (>2MB). If it is large, then run the simulation for one timestep
and save the file, this will eliminate potentially large quantities of graphical
output data.

PETROLEUM EXPERTS LTD

2 Step by Step Examples

2.1 Getting Started - Step by Step Example
This section describes how to set up and run a very simple REVEAL (Build 111)
model from scratch. The emphasis is on how the options and wizard can be used to
enter basic information, rather than the engineering content of the model.
The example is a simple 25 by 25 by 15 (X, Y, Z) grid reservoir, with one horizontal
producing well at its centre.
Step 1

initialise new case

Step 2

wizard basics

Step 3

control section

Step 4

reservoir section

Step 5

physical section

Step 6

relperm section

Step 7

wells section

Step 8

initialisation section

Step 9

injection schedule section

Step 10 run simulation

Step 11 view results

2 - 19

2.1.1

CHAPTER 2 STEP BY STEP EXAMPLE

Step 1 - Initialise New Case

Start REVEAL, and open a new project (File|New), or use the icon

Select Options|Units and set the input and output units to Oilfield, then select OK.

PETROLEUM EXPERTS LTD

CHAPTER 2 STEP BY STEP EXAMPLES

3 - 19

Now is a good time to save the file using File|Save Project As..., and enter a file
name (e.g. example1.rvl).

2.1.2

Step 2 - Wizard Basics

The REVEAL input script is divided into several sections. These may be modified
either by directly altering the text in the input script, or by using the project wizard.
The wizard may be used for one section only, or run consecutively through all of the
sections.
The wizard is activated with a right click inside the input script window, or by selecting
Input. Once the wizard ends normally (section(s) completed, or Finish selected), the
output script is updated. To prevent the changes made in using the wizard from
being written to the output script, select Cancel. Once the Wizard is exited, the only
data stored internally by REVEAL is that written to the script and held in datablocks1.
The Next and Previous buttons can be used to navigate forwards and backwards
through the wizard screens. The Validation, Plot and Calculate buttons may be used
on some screens to check data input.
It is obviously advisable to save the REVEAL file archive (*.rvl) using File|Save
Project or File|Save Project As... at regular intervals. The wizard cannot be running
while a save is made.
The REVEAL archive file contains the script and additional project files such as PVT,
lift curve, output graphics etc. The archive files present may be viewed, added to or
deleted using the option Project|Edit / View Project....
Start the wizard using Input|Script Wizard|Run All.

Potentially large arrays of data (e.g. grid coordinates) held in the REVEAL archive but not
displayed in the input script.

OCTOBER 2003

REVEAL MANUAL

4 - 19

2.1.3

CHAPTER 2 STEP BY STEP EXAMPLE

Step 3 - Control Section

This section sets the scope of the REVEAL model being created, and includes major
solver options.
The first screen selects the solver and its options. Two IMPES (implicit pressure and

CHAPTER 2 STEP BY STEP EXAMPLES

5 - 19

The next screen selects additional components (additional to water, oil & gas),
including non-reservoir injection gases. Select Next. This completes the control
section.

OCTOBER 2003

REVEAL MANUAL

6 - 19

2.1.4

CHAPTER 2 STEP BY STEP EXAMPLE

Step 4 - Reservoir Section

This section enters the reservoir grid and some basic physical properties of the
reservoir.
On the first screen enter a Cartesian grid with X and Y dimension set to 25 and Z
dimension set to 15, then select Next.
We are going to enter a very simple grid with 300 ft square and 20 ft deep blocks.
Enter the values as shown, then select Add to enter the Z direction depths. Select
Plot to view a wire frame mesh. Close the plot screen to return to the Reservoir
Section: Cartesian grid input screen and select Next.

The next screen enters the reference depth at the top of the grid. Select Add, then
All, to set all of the columns of blocks, then enter a value of 10000 ft. This data has
been entered and can be checked using the Validate button. The list of data entered
is not updated on the screen dynamically, but may be seen by clicking left over the
list.

PETROLEUM EXPERTS LTD

CHAPTER 2 STEP BY STEP EXAMPLES

7 - 19

Select Next to enter the porosity data, select Add, then select All grid blocks, and
enter a value of 0.2.
Select Next to enter the X direction permeability, select Add, then select All grid
blocks, enter a value of 100 mD. Select the Y Permeability tab at the bottom of the
screen, and leave the Y direction permeability multiplier at 1, select Z Permeability
tab and set the Z direction permeability multiplier to 0.1.
Select Next to enter transmissibility multiplier data, used for permeability barriers etc.
We wont be using these so select Next to enter absolute transmissibility data. If this
is entered (and the drop-down list at the top of the screen is changed to say 'Yes')
REVEAL will not perform its own transmissibility calculations but will use these
transmissibilities directly. These would normally be imported directly from the results
of another simulator, so we wont be using these, select Next. We wont be setting
net-to-gross ratios so select Next again. We won't be using pore volume multipliers
so select Next again to define rock types.
We wont be setting net-to-gross ratios so select Next again. We won't be using pore
volume multipliers so select Next again to define rock types. We will have only one
rock type so select Add, then All. The same goes for PVT and equilibration regions
on the next two screens.
After completing the PVT and equilibration region screens select Next to enter nonneighbour connections, then Next again to enter grid refinement regions. We wont
be using either of these for this example.
This ends the reservoir section and the data entered should be saved. Select Finish
to end the wizard and write the data entered to the input script. Then select

OCTOBER 2003

REVEAL MANUAL

8 - 19

CHAPTER 2 STEP BY STEP EXAMPLE

File|Save Project to save the REVEAL archive. The data entered so far can be seen
by scrolling up within the input script window.
The on-context help can be invoked by selecting the icon shown below to view a
simple description of the data entered in the script. Move the cursor over the script to
view the on-context help and turn the on-context help off by de-selecting the icon
shown.

PETROLEUM EXPERTS LTD

CHAPTER 2 STEP BY STEP EXAMPLES

2.1.5

9 - 19

Step 5 - Physical Section

This section enters the fluid and rock physical properties, the most important of which
are the PVT for water, oil and gas.
Use the right mouse button from within the input script window to restart the wizard at
the physical section by selecting Start Wizard|Physical.
We are going to enter a simple (unmatched) black oil PVT for an oil that is
undersaturated at 5000 psig and 200 F. Select New to enter a new PVT, then select
Oil for the fluid type and enter a name (e.g. example1), then select OK to bring up the
PVT entry screen. This is the same as other Petroleum Experts PVT entry and
indeed previously created PVT files can be imported directly using the Import option.
Do not do this now since we are creating PVT data from scratch and not importing an
existing file.
Enter the following black oil PVT data and then select Done to save the PVT data.
We are not concerned with complexities of PVT matching and miscibility during this
tutorial.

Thermal conductivity, diffusivity and dispersion data may be entered on this screen,
but we are not including these effects in this model. Enter the following physical data
using the default thermal properties button.

OCTOBER 2003

REVEAL MANUAL

10 - 19 CHAPTER 2 STEP BY STEP EXAMPLE

Note that units may altered dynamically by left clicking over the unit (highlighted
above). Dont alter the units now.
Select Validate to check the data entered. Click right over any data entry cell and
select Validation Range to change the minimum and maximum values. It is not
necessary to alter the validation range for REVEAL to accept data input. Any
changes made to the validation ranges are saved with the file.
Select Next to enter the rock compressibility, enter a value of 1e-5 psi-1 for the
uniform compressibility.
Select Next to enter various (mostly optional) viscosity model parameters. No data
entry is required so select Next to complete the physical data section.

PETROLEUM EXPERTS LTD

CHAPTER 2 STEP BY STEP EXAMPLES

2.1.6

11 - 19

Step 6 - RelPerm Section

The relative permeability and capillary pressure data are entered in this section.
Relative permeabilities may be direction and well dependent, but not for this case so
select Next.
This screen defines which residual saturation data will be applied to which rock type.
This is simple in this case, since we are only going to enter one set of data. Select
Add Region and make sure the Rock Coverage All checkbox is ticked. Enter the
residual phase saturations as shown and select Next. Note that residual saturations
for desaturation and gas hysteresis would be entered on this screen if these models
were active.

Select Add Rel Perm, then tick the Directions(All) tickbox and set the Data Entry to
Parametric. Enter the following data and select Plot to view the data entered. Select
OK, then Next to enter the capillary pressure data.

OCTOBER 2003

REVEAL MANUAL

12 - 19 CHAPTER 2 STEP BY STEP EXAMPLE

We are going to have no capillary pressure in this model, so enter the following
tabular data.

PETROLEUM EXPERTS LTD

CHAPTER 2 STEP BY STEP EXAMPLES

13 - 19

This completes the relative permeability section, so select Finish to write all of the
data entered so far to the input script. Save the REVEAL archive and restart the
wizard at the Wells section. Note that the Aquifer, Mobility, Phase behaviour,
Adsorption and Water Chemistry sections are not required for this simple example.

OCTOBER 2003

REVEAL MANUAL

14 - 19 CHAPTER 2 STEP BY STEP EXAMPLE

2.1.7

Step 7 - Wells Section

Start the wizard in the wells section if it is not already there.

We are going to enter one horizontal well perforated at a depth of 10150 ft (centre of
the reservoir). In the wells list select Create, to add the well, then select Multilateral
description in the Enter well position by: selection box. Give the multilateral
description an identifier name (e.g. horiz) and set the BHP reference to Fixed
elevation and set a value of 10050 ft (remember the top of the reservoir is at 10000 ft
and we have 5 vertical blocks of 20 ft each).

Select the Edit Multilateral button to enter the deviation survey for the well.
Select the Change Reference button to set the first node of the well at coordinates
(2000,3750,10150). Select OK once the reference position has been entered. Select
Add tubing and then OK. Next, enter the deviation survey as shown below, note the
internal diameter units have been change from feet to inches, and finally select OK to
return to the wells input screen.

PETROLEUM EXPERTS LTD

CHAPTER 2 STEP BY STEP EXAMPLES

15 - 19

Select Plot to view the reservoir and completed well (the left mouse button rotates the
reservoir and the CTRL and SHIFT keys in combination with the left and right mouse
buttons all modify the display if it gets too distorted use F5 to return to a default
scaling). Return to the wizard and select Next to complete the Wells section.

2.1.8

Step 8 - Initialisation Section

The first page of this section assigns a PVT description to each PVT region. Since
we only have one PVT region and one PVT description the default assignment has
been made already so select Next.
The next page sets initial reservoir temperature and pressure, in addition to initial
contact depths for each eqilibration region. We will simply set the temperature and
pressure at the centre of the reservoir. Enter a reference depth of 10050 ft and
reference pressure of 5000 psig. Enter an overburden and underburden temperature
of 200 F and a 0 F/ft thermal gradient.
Select Next to enter the last section.

OCTOBER 2003

REVEAL MANUAL

16 - 19 CHAPTER 2 STEP BY STEP EXAMPLE

2.1.9

Step 9 - Schedule Section

This section sets the well constraints and time-step control data. We will run this
case for 1000 days with a constant production rate of 10000 STB/d oil with a
minimum Pwf of 2000 psig.
Select Produce for the well type and Fix rate for the well control, and enter a rate of
10000 STB/d and set the Rate type to Total (oil+water). Select the green Constraints
button to enter the minimum Pwf of 2000 psig.
Set the initial time step size to 1 day. We recommend not using other time-step
controls unless it is shown to be required. Set the schedule to run for 1000 days as
shown. Finally select Next to complete the data input.

Save the REVEAL file using File|Save Project.

PETROLEUM EXPERTS LTD

CHAPTER 2 STEP BY STEP EXAMPLES

2.1.10

17 - 19

Step 10 - Run Simulation

Select Run Simulation|Select Properties... to select which data and how often it
should be saved. The grid data (arrays over active grid blocks) can be saved at
different intervals to the tabular data (well, average reservoir properties, wellbore
results). Modify this as appropriate and select OK when this has been done.
Start the calculation (

OCTOBER 2003

) and save the REVEAL archive once it has completed.

REVEAL MANUAL

18 - 19 CHAPTER 2 STEP BY STEP EXAMPLE

2.1.11

Step 11 - View Results

Use the
buttons to view the results. The results may also be viewed
interactively during a calculation.
The
button provides average reservoir properties such as mean phase pressures,
saturations and volumes. These data may be viewed, plotted and printed, or copied
to the clipboard for import into another application.
The
button provides well results data, including instantaneous and cumulative
rates, GOR, water cut and phase pressure and saturation data at well blocks.
The
button provides a three dimensional display of the fluid properties that were
selected for display (Run Simulation|Select Properties...). It includes graphical
representation of wells, completions and fractures.
The following is a list of recommended control functions within the 3D display
window.
left mouse
Ctrl+right mouse

Shift+left mouse

Ctrl+left mouse
F5
W key
S key

rotates the image in wire-frame mode about the X and Y axes.

rotates the image in wire-frame mode about the Z axis.

pans the image. This may also be achieved by using the left
mouse button near the edge of the display window.
selects a region to be enlarged.
returns the scaling to its original size.
sets the display to wire-frame mode.
sets the display to surface mode.

The following is a list of recommended options available using the right mouse button
within the 3D display window.
right mouse|3D Options

right mouse|Scale

right mouse|Object
Properties

right mouse|View
right mouse|Playback
Options

set block shrink factors, surface transparencies and

select regions of blocks (usually a plane of blocks) to
view.
set the scale range for colour tables; set linear and
logarithmic scaling and the end points (minimum and
maximum values to scale between).
set visibility (presence or absence), transparency, wireframe or surface mode for individual elements within the
display (e.g. set fractures to be wire-frame). Some
element types may be selected individually (wells,
completions, fractures) by selecting the appropriate
element type with the left mouse button in the Viewing
Objects column of the table.
set the variable to view.
set the range, interval and speed of 3D playbacks. The

forward and reverse playback is initiated and

controlled using the
buttons.

Once a view has been created, its properties may be saved and reopened later. Use
Options|Preset Options|Save File and Options|Preset Options|Load File to achieve
this. Enter a name for the view, select the properties to be saved (usually All) and

PETROLEUM EXPERTS LTD

CHAPTER 2 STEP BY STEP EXAMPLES

19 - 19

select OK. Note that the REVEAL archive2 must be saved for the view to be saved
permanently. Many of the REVEAL example cases have default views saved.
To view grid block properties evolving with time on a 2D plot, select the blocks to be
viewed, either using Edit|Select Multiple Blocks, or by selecting the
button to
button
select and unselect individual blocks using a left mouse click. Deselect the
once the required blocks have been selected. Select Results|Results of Selected
Cells to view the cell results. Use Edit|Unselect All Blocks to deselect all blocks.

The REVEAL archive file (*.rvl) contains the script and additional project files such as PVT,
lift curve, output graphics etc. The archive files present may be viewed, added to or deleted
using the option Project|Edit / View Project....

OCTOBER 2003

REVEAL MANUAL

3 Using the Input Wizard

A REVEAL input data set is divided into several sections. These may be modified either
by directly altering the text in the input script, or by using the project wizard. The wizard
may be used for one section only, or run consecutively through all of the sections.
The wizard is activated with a right click inside the input script window, or by selecting
Input from the menu. Once the wizard ends normally (section(s) completed, or Finish
selected), the output script is updated. To prevent the changes made in using the wizard
from being written to the output script, select Cancel. Once the Wizard is exited, the only
data stored internally by REVEAL is that written to the script and held in datablocks1. The
Next and Previous buttons can be used to navigate forwards and backwards through the
wizard screens. The Validation, Plot and Calculate buttons may be used on some
screens to check data input.
It is obviously advisable to save the REVEAL 2archive file using File|Save or File|Save As
at regular intervals. The wizard cannot be running while a save is made.

3.1 Input Wizard

The input wizard is divided into 13 sections, each containing several related pages. The
sections and pages displayed are context sensitive and will depend on the options
previously selected, the principal options are set in the control section.
General help starting and using the input wizard is available, along with specific page-bypage help.
Control section
Reservoir section

Physical section

Relperm section

Aquifer section
Mobility section

Phase section

Solver options, model options, component selection

Geometry, grid depth, porosity, permeability,
transmissibility multipliers, absolute
transmissibilities, net to gross, pore volume
multipliers, rock type, PVT region, equilibration
region, non-neighbour connections and grid
refinement
Physical properties, rock compressibities and
miscellaneous models (wax dropout, interfacial
tension, miscibility, wettability, non-Newtonian oil)
Options, residual saturations, relative permeabilities,
capillary pressures, endpoint scaling
Analytical aquifer, Carter Tracy model
Polymer/gel, original Carreau Kuparuk viscosity
models, gelation & degradation, inaccessible pore
volume, foam, partition coefficients
Introduction, alcohol/polymer partitioning, surfactant

Potentially large arrays of data (e.g. grid coordinates) held in the REVEAL archive but not
displayed in the input script.
2

The REVEAL archive file (*.rvl) contains the script and additional project files such as PVT, lift
curve, output graphics etc. The archive files present may be viewed, added to or deleted using the
option Project|Edit / View Project....

2 - 84

CHAPTER 3 USING THE INPUT WIZARD

Adsorption section
Water chemistry section

Wells section
Well-bore heating section
Initialisation section

Schedule section

micro-emulsion phase, ternary diagram, interfacial

tension, surfactant viscosity
Adsorption properties, permeability reduction
Water chemistry options, H2S souring options, scale
inhibition options, chemical database
Location and properties, fractures, stress
Well-bore heating model
PVT initialisation, equilibration initialisation,
component concentrations, residual saturations
Well schedule, thermal fracture schedule

3.2 Control Section

3.2.1 Solver Options
Four principal solver options are available, two IMPEC (implicit in pressure and explicit in
concentration) solvers and two implicit solvers. The term IMPES (implicit in pressure and
explicit in saturation) is equivalent to IMPEC in this context.
Hints and tips - always start with the default options. Some experimentation is
sometimes required for difficult cases.
By checking the text debug output file, identify where cpu time is being spent (end of
debug file).
If the preconditioner and linear solver times are very different, then increase or decrease
the power of the preconditioner appropriately. The more time spent in the preconditioner
will speed the linear solve and vice-versa, so an optimum is usually when a simular cpu
time is spent in the preconditioner and linear solver. To increase the preconditioner
power, reduce the preconditioner drop and/or permulation tolerances. This increases the
number of non-zero elements in the preconditioner.
If the linear solver is having problems, then increase the preconditioner power or change
the row/column ordering. Usually try ZD4 or D4Z orderings.
If the newton steps are not converging, then this is often a consequence of problematic
derivatives. Try using a known simple PVT, then simple relative permeability and
capillary pressure curves. Sometimes the linear solver should be pushed further to
improve Newton convergence. This is typically seen when the Newton convergence
looks good for the first few iterations, then fails to converge. The linear solver may be
pushed further by reducing the relative residual reduction. Also reduce the maximum
saturation change on the solver screen.
Check for very small blocks or very large permeability and gas flow.
Implicit pressure explicit concentration - The IMPEC solvers solve each time-step
faster than the implicit formulation and uses iterations to ensure material balance3. Its

Material balance means that the mass and volume of components/phases within each grid block
at the end of the time-step are consistent with the phase densities, following the volumetric Darcy
flow during the time-step.

PETROLEUM EXPERTS LTD

CHAPTER 3 USING THE INPUT WIZARD

3 -84

time-steps are limited to the Courant limit4. It has various upstream weighting options to
minimise numerical dispersion5 and reduce grid orientation6 effects. These are the only
solver options available for micro-emulsion surfactant models, and should generally be
selected for models where numerical dispersion of trace components is important (e.g.
water chemistry) and the Courant time-step limit is acceptable.
Implicit pressure and saturation - The implicit solvers are implicit in three phases and
three components (water, oil and gas), using a one-point upstream weighting for
mobilities. If more than the three principal components are present, then a Courant limit
is required on the time-step size to update the additional component concentrations
explicitly. Thermal models are not subject to the Courant time-step limit in the implicit
formulation, but temperature may be subject to numerical dispersion at larger time-steps.
These options should be selected for three component models where high velocities and
small grid blocks are present (e.g. gas coning).
Old IMPES - this is the original IMPES solver and has now been superceded by a more
general IMPES solver (see next option). This option is required if the thermal fracture
DLL is to be used, but does not support grid refinement. If neither grid refinement or a
thermal fracture DLL are present then it should produce virtually identical results the
default IMPES solver below.
IMPES - this is the default IMPES solver, which will be maintained to include future
models. It is effectively a subset of the fully implicit solver.
Fully Implicit - this solver option should be used for three component systems (water, oil
and gas) and other chemical models, where an implicit solver is required. It is not
available for surfactant models, where the IMPES solver should be used.
CFL Implicit - this is the fully implicit solver, where the Courant timestep limit is imposed
to minimise numerical dispersion for chemical models.
Depending on the solver option chosen, various solver options are available on two
screens. Defaults are provided, and altering them must be performed with care. Some
models may be significantly speeded up by careful choices.

Courant Flux Limit (CFL) ensures that no more than one pore-volume throughput through any grid
block occurs within a time-step.
5

Spatial averaging (smearing out) within regions of high concentration gradients (e.g. flood front or
trace component) due to linearisation of equations over the grid and in time. Made worse by using
large grid blocks and large time-steps. Upstream weighting schemes are used to minimise the
effect by extrapolating mobilities between grid blocks to maintain large concentration gradients
where possible. Also known as numerical diffusion.
6

Flow parallel and diagonal to the grid blocks' orientation may produce significantly different
breakthrough times. This may be automatically minimised in REVEAL, using a method that alters
the grid block phase transmissibilities to effectively increase diagonal flow and reduce parallel flow.

OCTOBER 2003

REVEAL MANUAL

4 - 84

CHAPTER 3 USING THE INPUT WIZARD

3.2.1.1 Both Implicit Solvers - First Screen of Solver Options

Maximum variable changes - limits for pressure and saturation are entered. Generally
the defaults should be applied. The saturation limit may be reduced to 0.1 to reduce timesteps and make the solve easier if convergence is slow.
Convergence limits - convergence limits for pressure, saturation and relative rate for
fixed rate wells. Generally the defaults should be applied. The saturation and implicit
well limits may be reduced to 0.001 to increase the number of Newton iterations and
reduce the material balance error, reducing oscillations.
Row/Column ordering - this option is probably the most important and should be set
appropriately. It includes various matrix ordering schemes, the cycling is from left to right
(e.g. XYZ cycles the in the X direction, then Y direction, then Z direction, with the cycling
fastest in X direction). Some D4 (chess board) ordering schemes are also included. In
general the fastest ordering should correspond to the highest transmissibility direction
(often Z) or the smallest bandwidth (direction with smallest number of blocks). For many
areal models, ZD4 or ZYX would be good choices.
Horizontal point scheme - either a 5 point or 9 point horizontal flow scheme is available
for the fully implicit solver. The 9 point scheme is recommended since it significantly
reduces grid orientation effects.
Newton iteration options - these options should not be altered for most models. The
maximum number of Newton iterations is entered. If a converged solution is obtained, but
it took more than the hold limit, then the next time-step will not be increased. If a
converged solution is obtained, but it took more than the reduce limit, then the next timestep will be reduced. A time-step is retaken with a reduced time-step if the current
residual is lager than the divergence condition multiplied by the minimum residual
calculated (i.e. detects Newton divergence). The maximum Newton step size is 1.0 for a
full Newton step; for cases where Newton convergence is not occurring (oscillations or
divergence) this value may be reduced to 0.5 to improve convergence at the expense of
the number of Newton iterations required.
Pre-conditioner options - the options for linear solver pre-conditioner should not
generally be altered from the default values. The fill in parameter is the bandwidth of the
pre-conditioning matrix, and may be increasedto help convergence of the linear solver;
alternatively it may be lowered to reduce the size of the pre-conditioning matrix and hence
memory requirements. The drop and permutation tolerances refer to the relative size of
off-diagonal matrix elements included in the pre-conditioning matrix and column
swapping.
Solver options - the options for linear solver should not generally be altered from the
default values. A stabilised bi-conjugate gradient (BCGSTAB) and generalised minimum
residual solver (GMRES) are used. The residual reduction is a relative residual reduction
defining convergence, and the residual maximum is an absolute residual limit.
Sometimes reducing the residual maximum may help reduce oscillations between Newton
iterations. The residual is defined as the r.m.s. mass rate error. The Krylov space size is
used for the GMRES solver. If this value is increased, the solver may become more
robust, but may require significantly more memory and become much slower, it may be
reduced to a minimum value of 2 to minimise memory requirements.

PETROLEUM EXPERTS LTD

CHAPTER 3 USING THE INPUT WIZARD

5 -84

3.2.1.2 Both IMPES Solvers - Second Screen of Solver Options

Upstream weightings define how properties required at grid block faces are calculated,
and are used to minimise numerical dispersion effects. Weightings are required for
component concentration, temperature, permeability (mobility) and density. The
permeability and density weightings are used prior to the implicit pressure solve, while the
component concentration and temperature weightings are used during the explicit phase
of the solve, after the implicit pressure solve has been performed.
The phase velocities are determined by the implicit pressure solve. The component
concentrations are the concentration of components within the phases present, and
therefore component concentration weightings only effect trace components within the
phases.
1-point - block face value is taken as the value in the centre of the upstream block. This
is recommended for component concentration where trace components are not present,
and also recommended for the surfactant micro-emulsion model, to ensure components
are transported in the ratios present within the upstream block.
2-point - block face value is taken as a value extrapolated from the two upstream blocks.
This method reduces numerical dispersion but generally not recommended.
fct/2-point - similar to the 2-point scheme, but extended fct (flux corrected transport)
algorithm [17] acts to minimise numerical dispersion. Recommended for temperature and
component concentrations where trace components are present.
fct/3-point - higher order, three block (two upstream and one downstream) fct method.
This method should only be used when all blocks are identical cubes. Not generally
recommended.
Original - for mobility weighting, this method uses a 1-point upstream weighting for
relative permeability and centred value for viscosity. Either this method or the 1-point
upstream weighting are recommended for permeability weighting.
Centred - for density weighting (used for gravitational term in implicit pressure solve), this
method is a mean upstream/downstream value. Either this method or the 1-point
upstream weighting are recommended for density weighting.
Diagonal transmissibility - used to minimise grid orientation effects in the horizontal
(constant Z) plane. A value of 0 turns this model off, while a value of 0.5 is recommended
and is equivalent to allowing up to 1/3 of flow to be diagonal rather than parallel to block
faces. This model only has significant effect near regions of large mobility variations (e.g.
flood front).

3.2.1.3 IMPES Solver

Row/Column ordering - See Section 3.2.1.1
Pre-conditioner options - See Section 3.2.1.1
Solver options - See Section 3.2.1.1

3.2.1.4 Courant Flux Limit implicit solver

Concentration and temperature upstream weighting - see above.

OCTOBER 2003

REVEAL MANUAL

6 - 84

CHAPTER 3 USING THE INPUT WIZARD

3.2.2 Model Options

This wizard page initialises the models that will be present, and consequently the
subsequent sections and screens that will be displayed by the input wizard.
Fracture model - includes thermal injection fractures and hydraulic production fractures.
Aquifer model - is based on the Carter Tracy analytical model.
Component models - include simple (water, oil and gas), surfactant (4th phase microemulsion), polymer/gel (various aqueous, oleic and gas phase mobility models) and water
chemistry (aqueous phase precipitation model). The components that can be used later
will depend on the component model selected on this screen, and as such these models
are mutually exclusive. Chemically and physically inactive tracer components are
available with all component models.
Souring - H2S souring model, only available if the component model is water chemistry.
Scale inhibition - precipitate scale inhibitor model, only available if the component model
is water chemistry and the component Scale Inhibitor is included in the component list.
Wettability, miscibility and well-bore heating - if these models are selected, then the
various wizard screens for their data input will become active as the wizard is progressed.
Transmissibility model - select 'normal' or 'extended'. The extended model is
essentially that implemented in Eclipse as 'NEWTRAN'. In the normal model, the
interface block area, rather than projections of mutual interface areas, is used in the
transmissibility calculations.
Reference conditions - The reference temperature is used for some correlations (e.g.
adsorption or partitioning relations). The reference depth is used for reported reservoir
average pressures and also used to reference all well Pwfs back to a common datum.
Inactive block definition - there are two criterea that may be used to flag inactive blocks.
1. The porosity*net-to-gross is less than a defined limit, or
2. The grid cell pore volume is less than a defined limit.
User Fracture DLL - a DLL (Dynamically Linked Library) may be supplied by a third party
to perform thermal fracture calculations. Contact Petroleum Experts for more information
on the DLL interface for this option.

PETROLEUM EXPERTS LTD

CHAPTER 3 USING THE INPUT WIZARD

7 -84

3.2.3 Component List

Reservoir components - the components to be included in the model are added by
selecting the component and pressing Add, components may be deleted by selecting the
component to be deleted in the list and using the Delete button. Water, oil and gas are
always present. The components available will depend on the component model selected
(previous wizard screen).
Injected component - A non-reservoir gas for injection may be added. This gas will not
modify the underlying oil PVT model and the mixing of the injection gas with reservoir gas
is modelled approximately, since this is not a fully compositional model.

3.3 Reservoir Section

3.3.1 Geometry
All REVEAL models consist of hexahedral grid blocks connected in a regular structure
(topologically identical), with each block identified by its I,J,K numbers. I runs from 1 to
NX, J from 1 to NY and K from 1 to NZ, where NX, NY and NZ are the number of grid
blocks in each principal direction, with the Z direction being vertically down. NX must be
greater than one.
Select the grid coordinate system and the number of blocks in each principal direction
NX, NY and NZ.
The Update Globally button will update all other sections where possible to reflect
changes made to NX, NY or NZ. The update globally option is only required when the
grid size is being changed, and performs no function the first time a reservoir geometry is
defined.
Corner point grid - most large models will use the corner point grid coordinate system.
The X, Y and Z coordinates of each block corner must be entered layer by layer
(increasing Z) into a data array, by selecting the Data Array button. The data array must
also be given a name. The X, Y and Z coordinate values should all be increasing in their
respective positive directions. The corner point data may be copied and pasted from
another application such as EXCEL, or imported directly from another simulator, where a
filter is provided. See the Input option on the main bar for grid geometry import options.
Once potentially large volumes of corner point data have been entered it is advisable to
select Finish to leave the wizard and save the REVEAL archive7.
Corner point (simplified) - in this scheme adjacent blocks share coordinates, there are
therfore (NX+1) * (NY+1) * (NZ+1) coordinates in total. These are entered as X, Y and Z
coordinates for each layer, with (NX+1) * (NY+1) enties per layer for each spatial
direction.
Corner point (full) - in this scheme ach block has 8 independent coordinates, there are
therefore (2*NX) * (2*NY) * (2*NZ) coordinates in total. These are entered as X, Y and Z
7

The REVEAL archive file (*.rvl) contains the script and additional project files such as PVT, lift
curve, output graphics etc. The archive files present may be viewed, added to or deleted using the
option Project|Edit / View Project....

OCTOBER 2003

REVEAL MANUAL

8 - 84

CHAPTER 3 USING THE INPUT WIZARD

coordinates for each layer, with (2*NX) * (2*NY) entries per layer for each spatial
direction. Each block is considered in turn and its two corner point coordinates are
entered before the next block data is entered.
Map Axes - may be entered for cartesian or corner point grids. Three coordinates in the
horizontal plane (X,Y coordinates only) defining an origin and directions for the X,Y axies
of a map coordinate system are required. The coordinates are entered in the grid
coordinate system and define a second coordinate (map) system that mey be used to
enter well trajectories. Note that the map axes must be orthogonal (90 degrees between
the axes).
Cartesian grid - the Cartesian grid coordinate system generates a regular grid with
constant grid spacing.
Radial grid - the radial grid coordinate system generates a grid with radial grid blocks
forming a sector away from a central well. For the radial grid option a central well radius
must be input. The sector angles may be constant; if this is required then select the
Uniform Sector Angle button and enter an angle. The units for the angle may be toggled
between radians and degrees by clicking right within the angle data entry cell. Note that
there is no flow between the grid blocks at Y=1 and Y=NY, even if the radial model is 360
degrees.
Curvilinear grid - the curvilinear grid coordinate system generates a grid with 2 wells as
one eighth of a repeated five-spot symmetry, reflecting the streamline flow between two
wells, one injector and one producer. For the curvilinear grid option, the spacing between
the well must be entered, and an excluded fraction. The excluded fraction is the fraction
of the distance between the two wells excluded from the model, which prevents stability
problems near the wells. It should typically be set to about 0.05.
An additional page of data input is required for the Cartesian, radial and curvilinear grid
options.

PETROLEUM EXPERTS LTD

CHAPTER 3 USING THE INPUT WIZARD

Cartesian grid

Corner point grid

Radial grid

Curvilinear grid

9 -84

3.3.2 Cartesian Geometry

Block sizes - the X and Y grid block dimensions should be entered for rows 1 to NX and
columns 1 to NY. Consecutive rows or columns with the same grid block dimension may
be entered as a group (e.g. if the first three dimensions are 100 ft and then next two
dimensions are 50 ft, then only two rows of the table require to be entered).
The Z dimension of the grid blocks is entered in one of three ways.
Z range - the most common method, where the Z dimension of the grid blocks are
entered for each layer (or groups of layers).
Single layer range - for a single layer, the Z dimensions for blocks may vary in the X and
Y directions. Care must be taken with this method to ensure that full coverage of the
model is achieved (all blocks have a Z dimension).
Layer n multiple - where one layer Z dimensions are a multiple of a previously defined
layer.
For each method of Z dimension entry, the range and Z dimension should be entered in
the appropriate data entry cells and the Add button selected to add the data to a table of
entries. An entry in the table may be modified by selecting and highlighting the required
entry in the table and then modifying the data entered; the Add button should not be used
to update modified data, since it will add another row to the table. The modified data can
be viewed in the table by re-selecting the row in the table. Table entries may be deleted
by using the Delete button.
OCTOBER 2003

REVEAL MANUAL

10 - 84 CHAPTER 3 USING THE INPUT WIZARD

Full coverage of the model may be checked by selecting the Validate button, and the
resulting grid checked by selecting the Plot button.

3.3.3 Radial Geometry

Block sizes - the X and Y grid block dimensions should be entered for rows 1 to NX and
columns 1 to NY. Consecutive rows or columns with the same grid block dimension may
be entered as a group (e.g. if the first three dimensions are 100 ft and then next two
dimensions are 50 ft, then only two rows of the table require to be entered). The X grid
block dimensions have units of length, while the Y grid block dimensions have units of
angle. If constant angles were set on the previous wizard screen, then no Y dimensions
are required.
The Z dimension of the grid blocks is entered in one of three ways.
Z range - the most common method, where the Z dimension of the grid blocks are
entered for each layer (or groups of layers).
Single layer range - for a single layer, the Z dimensions for blocks may vary in the X and
Y directions. Care must be taken with this method to ensure that full coverage of the
model is achieved (all blocks have a Z dimension).
Layer n multiple - where one layer Z dimensions are a multiple of a previously defined
layer.
For each method of Z dimension entry, the range and Z dimension should be entered in
the appropriate data entry cells and the I button selected to add the data to a table of
entries. An entry in the table may be modified by selecting and highlighting the required
entry in the table and then modifying the data entered; the Add button should not be used
to update modified data, since it will add another row to the table. The modified data can
be viewed in the table by re-selecting the row in the table. Table entries may be deleted
by using the Delete button.
Full coverage of the model may be checked by selecting the Validate button, and the
resulting grid checked by selecting the Plot button.

3.3.4 Curvilinear Geometry

Block sizes - the X grid block dimensions should be entered for rows 1 to NX.
Consecutive rows with the same grid block dimension may be entered as a group (e.g. if
the first three dimensions are 100 ft and then next two dimensions are 50 ft, then only two
rows of the table require to be entered). The Y grid block dimensions are not required,
since they are calculated by the model.
The Z dimension of the grid blocks is entered in one of three ways.
Z range - the most common method, where the Z dimension of the grid blocks are
entered for each layer (or groups of layers).
Single layer range - for a single layer, the Z dimensions for blocks may vary in the X and
Y directions. Care must be taken with this method to ensure that full coverage of the
model is achieved (all blocks have a Z dimension).

PETROLEUM EXPERTS LTD

CHAPTER 3 USING THE INPUT WIZARD

11 -84

Layer n multiple - where one layer Z dimensions are a multiple of a previously defined
layer.
For each method of Z dimension entry, the range and Z dimension should be entered in
the appropriate data entry cells and the Add button selected to add the data to a table of
entries. An entry in the table may be modified by selecting and highlighting the required
entry in the table and then modifying the data entered; the Add button should not be used
to update modified data, since it will add another row to the table. The modified data can
be viewed in the table by re-selecting the row in the table. Table entries may be deleted
by using the Delete button.
Full coverage of the model may be checked by selecting the Validate button.

OCTOBER 2003

REVEAL MANUAL

12 - 84 CHAPTER 3 USING THE INPUT WIZARD

3.3.5 Grid Depth

For all grid geometries except corner point (where the coordinates of all blocks are
entered explicitly), the depth of the top surface of the grid must be defined. The depth
must cover the entire top layer (Z=1) of grid blocks, and is defined as the depth from a
zero datum to the top of the grid blocks.
For Cartesian grids, two methods are available, range/array and dip angles. For the
radial and curvilinear grids, only the range/array method is available.
Select Add to add an entry to the table of values.
Range - each table entry may cover a range of blocks with a constant value (e.g. a flat
top surface at 5000 ft would be achieved by selecting range, then the All button and
entering a value of 5000). This data has been entered to the table, but the table display is
not updated until an entry in the table is selected with the mouse. Several entries to the
table may be added, modified or deleted. The coverage of the table entries must cover all
blocks on the top layer (X=1,NX and Y=1,NY). The Validate button may be used to check
coverage. Select entries in the table with the mouse to modify or delete them.
Array - an array of depths are entered, one for the centre of each top surface grid block.
The array must be given a unique name before data is entered. Data may be copied and
pasted into the array from another application.
Dip angles - the depth of the first block (X=Y=1) is entered, and the dip angles in the X
and Y direction are entered.

3.3.6 Porosity
The porosity of each grid block is required. This is the porosity at the reference pressure,
with porosities being calculated as a function of rock compressibility for grid blocks with
pressures below or above the reference pressure.
A zero porosity is used to identify inactive grid blocks, which will take no part in the flow
calculations.
Two methods are available to enter porosities, range and array.
Select Add to add an entry to the table of values.
Range - each table entry may cover a range of blocks with a constant value. Click the
right mouse button within the value entry cell to toggle between fraction and percentage
porosity. Several entries to the table may be added, modified or deleted. The coverage
of the table entries must cover all grid blocks. The Validate button may be used to check
coverage. Select entries in the table with the mouse to modify or delete them.
Array - an array of porosities are entered, one for each grid block, entered by layer with
increasing depth (increasing Z). The array must be given a unique name before data is
entered. Data may be copied and pasted into the array from another application.
Use parent grid values - only available if the current grid is a refined grid (see grid
refinement for general information on grid refinement and refined reservoir properties for
information on entering refined grid reservoir properties), then use porosities defined for
master (parent) grid.
PETROLEUM EXPERTS LTD

CHAPTER 3 USING THE INPUT WIZARD

13 -84

The Plot button may be used to view the porosity distribution entered.

3.3.7 Permeability
The permeability in the X, Y and Z directions is required for each grid block in the model.
Use the tabs at the bottom of the screen to enter the Y and Z permeabilities, which can be
entered explicitly, or as a multiple of the X direction permeabilities.
See the porosity help screen for details of how to add ranges or arrays of permeabilities.

3.3.8 Transmissibility Multipliers

This screen is optional and no data is required.
The default transmissibility between two blocks is calculated using a harmonic average8
of the two adjacent grid block permeabilities. Transmissibility multipliers may be used to
modify the transmissibilities between blocks to model transmissibility barriers (or
enhancement).
Transmissibility multipliers (default value of 1.0) may be given in the X, Y and Z directions
for any or all grid blocks. e.g. an X direction transmissibility multiplier of 0.1 for grid block
(3,4,2) reduces the transmissibility between grid blocks (3,4,2) and (4,4,2) by a factor of
10.
See the porosity help screen for details of how to add ranges or arrays of transmissibility
multipliers.

3.3.9 Absolute Transmissibility

This screen is optional and no data is required.
In normal use, the inter-block transmissibility is calculated internally by REVEAL from the
block permeabilties, the block geometries (corner points), and the transmissibility
multipliers. This screen can be used to overwrite those values with externally calculated
values in the x-, y-, and z-directions. The values here will only be used if the 'Use external
transmissibilities' option is set at the top of the screen.
This data would normally be imported directly from an ASCII file or from another
simulator.
See the porosity help screen for details of how to add ranges or arrays of
8

The harmonic average is the 'resisters in parallel' average. The following averages are defined for
a variable x, with weightings w.
arithmetic average

x=

xi wi
w

OCTOBER 2003

harmonic average

x=

w
w
x

i
i

REVEAL MANUAL

14 - 84 CHAPTER 3 USING THE INPUT WIZARD

transmissibilities.

3.3.10 Net to Gross

This screen is optional and no data is required.
The default horizontal permeabilities (XY plane) may be modified by entering a net to
gross ratio for any or all grid blocks (default value of 1.0). e.g. a net to gross of 0.5
reduces the X and Y direction permeabilities by a factor of 2.
See the porosity help screen for details of how to add ranges or arrays of net to gross
ratios.

3.3.11

Pore Volume Multipliers

This screen is optional and no data is required.

The pore volume multipliers are multipliers for block volumes as calculated from the grid
geometry. The porosity and inter-block transmissibilites are not affected by these
multipliers.
See the porosity help screen for details of how to add ranges or arrays of pore volume
multipliers.

3.3.12

Rock Type

A rock type must be assigned to each grid block in the model. Rock types are used to
define regions of the model with the same physical or dynamic properties (e.g. relative
permeability tables or rock heat capacity) later in the input wizard.
To add a rock type select the Add button and define its coverage. To modify or delete a
rock type select the rock type from the table and modify its coverage or select the Delete
button.
The coverage of a rock type is defined by a table of selected regions. Define a region by
selecting ranges of X, Y and Z grid blocks, or the entire grid by using the All button.
Multiple regions may be added, modified and deleted to define the coverage for each rock
type.
Check the coverage is complete by using the Validate button.

PETROLEUM EXPERTS LTD

CHAPTER 3 USING THE INPUT WIZARD

3.3.13

15 -84

PVT Region

A PVT region must be assigned to each grid block in the model. PVT regions are used to
define regions of the model where different PVT may occur. Therefore different PVT
regions must be non communicating.
To add a PVT region select the Add button and define its coverage. To modify or delete a
PVT region select the region from the table and modify its coverage or select the Delete
button.
The coverage of a PVT region is defined by a table of selected regions. Define a region
by selecting ranges of X, Y and Z grid blocks, or the entire grid by using the All button.
Multiple regions may be added, modified and deleted to define the coverage for each
region.
Check the coverage is complete by using the Validate button.

3.3.14

Equilibration Region

An equilibration region must be assigned to each grid block in the model. Equilibration
regions are used to define regions of the model where different equilibrations may occur.
Equilibration includes initial pressure and reference depth and also include any contacts,
these are defined in the initialisation section. Each equilibration region must only have
one PVT region associated with it.
To add an equilibration region select the Add button and define its coverage. To modify
or delete an equilibration region select the region from the table and modify its coverage
or select the Delete button.
The coverage of an equilibration region is defined by a table of selected regions. Define a
region by selecting ranges of X, Y and Z grid blocks, or the entire grid by using the All
button. Multiple regions may be added, modified and deleted to define the coverage for
each region.
Check the coverage is complete by using the Validate button.

3.3.15

Non-Neighbour Connections

This section is optional.

Transmissibility connection between pairs of non-neighbour blocks may be entered.
These connections can only be applied to blocks within the current refinement level.
Enter the I,J,K coordinates of the blocks to be connected and a transmissibility.
If the model has inactive (pinchout layers), then active vertical layers may be connected
automatically for blocks separated by less than the 'maximum length between cells for
auto-generation'. If this field is left empty, then no automatic non-neighbour connections
will be set, unless the MinPV option is selected. Other options are available for pinch-outs
as follows (the headings are the keywords that are written to the script):
MinPV: this will generate a pinch out between blocks that have intermediate blocks
removed because of their low pore volume (the pore volume is lower than the value set
for 'minimum grid pore volume' in the Control section.

OCTOBER 2003

REVEAL MANUAL

16 - 84 CHAPTER 3 USING THE INPUT WIZARD

MinTz: the transmissibility multiplier in the z-direction for the topmost cell in a pinch out
will be set to the minimum of the multipliers of the pinched out cells.

3.3.16

Grid Refinement

This screen is optional and no data is required.

Grid refinement is not available for radial or curvilinear grids, or using the original IMPEC
solver.
Refined regions within the main grid may be identified within this section. Only the scope
and names of refined region(s) are entered at this stage. The scope of a refined region
may contain any set of contiguous blocks within the main grid. Different regions of
refinement must not overlap or be contiguous (contact neighbouring refined regions). Any
well of fracture present within a simulation model must be either fully contained within the
main grid (unrefined region) or fully contained within a refined region (i.e. no well or
fracture may pass through a refinement boundary).
To Add or delete a refined region use the New and Delete buttons, and enter a unique
identifier name for the refined region.
The coverage of a refined region is defined by one or more hexahedral sub-regions which
are added or deleted using the Add and Delete buttons within the coverage section of this
screen.
Use the Validate and Plot buttons to check the coverage of refined regions.
See the grid refinement section for more information.

PETROLEUM EXPERTS LTD

CHAPTER 3 USING THE INPUT WIZARD

17 -84

3.4 Physical Section

3.4.1 Fluid Properties
The physical properties of the fluids and reservoir rock are entered. These include the
PVT for water, oil and gas components, density and compressibility for additional
volumetric components and additional thermal data (heat capacity and thermal
conductivity).
PVT - the PVT for water, oil and gas are entered using the standard Petroleum Experts
PVT screens, select the New or Edit buttons. If a new PVT is created it must be given a
name and be either oil, gas or condensate. A previously generated PVT file (*.PVT
generated by PROSPER) may be imported from the PVT screens using the Import facility.
Matched PVT may be converted into tables (up to 5 temperatures) using the Calc Tables
option from the PVT screens, speeding up most REVEAL calculations. If use tables is
selected, then the tables must be complete (all columns completed) and have sufficient
coverage over a range of temperatures and pressures to cover all expected values during
the REVEAL simulation.
Trace components - trace components are either volumetric or non-volumetric. The
mass or volume of the non-volumetric trace components take no part in the pressure
solve (for IMPEC or implicit formulations) so the density of trace components is only used
to convert from mass to volume fractions. For this reason, non-volumetric component
densities are set equal to the phase they are partitioned within (e.g. water chemistry
components are partitioned in the aqueous phase). This means that component mass
fractions within the phase are equivalent to volume fractions.
The volumetric trace components (injected non-reservoir gas, surfactant and the two
alcohol components) are included within the calculation of the phase densities and
therefore take part in the pressure solve. For these components, a reference density and
compressibility (c) are required to calculate the component densities.

= ref exp(c(P Pref ))

Thermal data - all REVEAL calculations are thermal, so heat capacity is required for all
volumetric components. Non-volumetric components take no part in the heat transport
equations. Scroll right on the component data form to enter heat capacities for water, oil
and gas. Thermal conductivities are required for volumetric components if the thermal
conductivity option is selected.
Rock data - thermal data for each rock type is required.
Over/underburden - the overburden/underburden rock density and thermal properties
are also entered. Heat flux from the underburden and to the overburden is calculated
using a model by Vinsome and Westerveld [10].
Diffusion and dispersion - select the options for component diffusivity and phase
dispersivity if required. Additional data (diffusivity and dispersion lengths) will be required
for these options. The diffusivity and dispersivity options are only available for the IMPEC
solver formulation.
Defaults - default thermal properties (specific heat capacity, thermal conductivity and rock

OCTOBER 2003

REVEAL MANUAL

18 - 84 CHAPTER 3 USING THE INPUT WIZARD

density) may be selected. These data are required even for nominally isothermal
simulations, where no thermal gradient or thermal injection are present. Default aqueous
component properties set the density and compressibility for additional aqueous
volumetric components (surfactant and alcohols).

3.4.2 Rock Compressibilities

The compressibility of each rock type is entered on this screen. Compaction and
permeability reduction are also modelled on this screen.
At least one rock compressibility should be entered, generally at the reference
pressure (see control section).
If the compaction option is not selected, then one rock compressibility is entered,
then the porosity is calculated using the following equation.
c (P P )
= ref e r ref
If the compaction model is active then a table of porosity and permeability
multipliers should be entered as a function of pressure. The pressures and
multipliers should increase mononically down the table. More than one entry
should be made covering the entire range of predicted reservoir pressures to
ensure rock compressibility.

3.4.3 Miscellaneous
Several unrelated physical properties for various models are entered on this screen. The
options available will depend on models selected in the control section.
Wax dropout - this model is optional and no data entry is required. A rapid increase in oil
viscosity at the wax drop-out temperature may be modelled. The oil phase viscosity is
multiplied by the exponent of Bwax*(Twax-T)/(1+|Twax-T|), where T is the oil
temperature.
Interfacial tension - only required if the micro-emulsion surfactant model is selected in
the control section. The initial oil/water interfacial tension is entered. A value of about 25
dyne/cm is reasonable for many oils.
Wettability - only required if the wettability model is selected in the control section. This
parameter controls the desaturation (interpolation between high and low tension relative
permeability curves) as a function of adsorbed wetting agent.
The initial reservoir wettability is set in the initialisation section; where a value of 1
represents high tension and -1 represents low tension. As the wetting agent is adsorbed
from zero to its maximum value, the wettability is increased from the initial value (at zero
adsorbed wetting agent) to the initial value plus twice the wettability change parameter
entered on this page (at maximum adsorbed wetting agent).
Therefore if the reservoir is initially high tension (wettability = 1) and a value of -1 is
entered for the wettability change parameter, when the adsorbed wetting agent reaches
its maximum value, the wettability will be 1+2*-1 = -1 (low tension). Similarly if the
reservoir is initially low tension (-1) and a wettability change parameter of 1 is used, the
wettability at maximum adsorbed wetting agent will be -1+2*1 = 1 (high tension).

PETROLEUM EXPERTS LTD

CHAPTER 3 USING THE INPUT WIZARD

19 -84

Miscibility - only required if the miscibility model is selected in the control section. The
miscibility pressure is entered as a linear function of temperature. Pmisc = P1+P2*(TTref), where T is the temperature and Tref is the reference temperature entered in the
control section. If the block pressure rises above the miscibility pressure, then the oil and
gas phase viscosities are modified by a quarter power mixing rule or the Todd Longstaff
model [5] if this option is selected.
A miscible hydrocarbon phase viscosity is calculated, where g is is the non water blocked
gas fraction.
1
1
h = o g g o 4 + (1 g ) g 4

g=

Sg
So + S g Sowb

The Todd and Longstaff model modifies the oleic and gas phase viscosities using a user
defined mixing parameter .

o = o 1 h

g = g 1 h

The oil and gas phase viscosities approach each other as the mixing parameter () is
increased, with a value of 1 leading to the gas and oil phase viscosities being equal
(equivalent to the quarter power mixing rule).
See the relperm section for more information on the wettability and miscibility models.
Non-Newtonian fluid - entered within the PVT screens ('Visc Tables' button common to
all Petroleum Experts software). The viscosity of heavy oils may be dependent on shear
rate and this can be modelled by entering various rheological constants for the oil phase
as a function of temperature and pressure. If this model is present, then the oil viscosity
calculated from the standard PVT calculator will not be used.
For a non-Newtonian fluid, the shear stress () is related to shear rate () by an
exponential equation, where the rheological constants (0,k & n) are entered on this
screen as functions of temperature and pressure.

= 0 + k n
In contrast, a Newtonian fluid has the simplified form, where is the Newtonian viscosity.

=
Note that if 0=0 and n=1, then k is the viscosity. The internal units for shear stress are
Pa, shear rate s-1, therefore k has units of Pa.s^n, and in the Newtonian limit k has units
of Pa.s, where 1 Pa.s = 1000 cp.
The effective (Newtonian) viscosity () is obtained using the shear stress calculated for
each grid block, see the polymer and gel mobility section for more information about
shear. The block shear stress is obtained using the following equation.

= P

8k kr
9S

OCTOBER 2003

REVEAL MANUAL

20 - 84 CHAPTER 3 USING THE INPUT WIZARD

If a grid block contains an injection well, then the well shear stress may be modified to
reflect increased shear near the well. The well bore radius (rw) and Peaceman radius (r0)
are used to estimate the increased potential gradient near the wellbore.

r0

rw

= ln

The effective Newtonian viscosity to be used for oil mobility is calculated for an oil using
the formula below, which is derived by comparison with laminar flow pressure drop
calculations in tubing for non-Newtonian oils.

Note that when 0 is zero n is unity, it reduces to the Newtonian form where the viscosity
is equal to k.
A maximum cutoff viscosity is required (zero shear limit), since as the shear stress
approaches zero, the model predicts the apparent Newtonian viscosity approaches
infinity.
The following is a rheological plot of oil viscosity and shear stress as a function of shear
rate.

Corrected gas-phase viscosity for foam - The assumption is made that any gas
present forms a homogenous bubbly mixture (foam) with the oil, modifying (increasing)
the equivalent Newtonian viscosity (reducing the shear stress) for the foam mixture.
Additionally, the gas phase viscosity is set equal to the oil phase (foam) viscosity to

PETROLEUM EXPERTS LTD

CHAPTER 3 USING THE INPUT WIZARD

21 -84

reduce the mobility of the gas phase (bubbles within the oil phase).

1 n
k

where

= 0 + k

OCTOBER 2003

1 n

=
and

o
f

REVEAL MANUAL

22 - 84 CHAPTER 3 USING THE INPUT WIZARD

3.5 RelPerm Section

3.5.1 General Options

The first screen of the relperm section initialises various options that will affect the options
and data required for the remaining relperm screens.
Model - either Stone I or Stone II are available for three phase oil phase relative
permeability. If the miscibility option is set and the pressure is above the miscibility
pressure then the oil and gas phase relative permeabilities are calculated using the
miscibility model.

Stone I

S o krow krog

1 Srw Srow

kr *ow = krow
kr *ow (1 S w )(1 S g )
1 Srow
Sg
S + S m (Srw + Srm )
S o Sro
S g =
S w = w
S o =
1 Srw Sro Srm
1 Srw Sro Srm
1 Srw Sro Srm
Sg
= 1
Sro = Srow + (1 ) Srog
1 Srw Srog Srm

kro =

Stone II

kr

krog
+ krg krw krg
kro = kr *ow ow + krw
kr *ow

kr *ow

0 kro 1
Hysteresis - may be applied to the water, oil and gas phases, see residual saturations for
more information.
Endpoint scaling - may be apllied to water, oil and gas phases, see endpoint scaling for
more information.
Directional relperms - more than one relative permeability may be defined per grid
block. The relative permeability curve used will then depend on the direction of fluid flow.
The default state for this option is off. The All faces option allows upto seven separate
relative permeability curves for each grid block (one for each direction X-,X+,Y-,Y+,Z-,Z+
and one for connected wells). The Up,down,horizontal option allows upto four separate
relative permeability curves for each grid block (one for horizontal flow X-,X+,Y-,Y+, one
each for Z-,Z+ and one for connected wells). If the directional relperms option is set, then
separate curves may be defined for wells, or defined to be one of the directional curves.
Capillary desaturation - This option is set for models where surfactant (capillary
desaturation9) is present. The surfactant model is enabled in the control section. When
9

Capillary desaturation is parameterised by comparing the ratio of viscous and

capillary/interfacial tension forces between phases. The capillary number (Nc) is defined
for each phase.
PETROLEUM EXPERTS LTD

CHAPTER 3 USING THE INPUT WIZARD

23 -84

the capillary desaturation option is set, two relative permeability curves will be required for
low and high tension, with interpolation between the curves depending on the
desaturation model. If capillary desaturation is set, when the surfactant model is not
present then the capillary desaturation will vary only with the changing viscous forces
since the interfacial tension between phases is constant.
Capillary desaturation should be off when the wettability model is selected, since the
interpolation between high and low tension relative permeability curves will be dependent
on adsorbed concentration of wetting agent rather than by capillary desaturation.
Gas miscibility - this option is only available when the gas miscibility option is selected in
the control section. When the oil and gas phases are miscible (pressure above miscible
pressure) the oil and gas relative permeabilities are modified using the Todd & Longstaff
model [5] to calculate weighted means of the oil and gas phase relative permeabilities.
The residual interpolation scheme uses an interpolated (between oil and gas phases)
value for hydrocarbon phase residual saturation, while the simple model uses the oil and
gas phase residual saturations.
A residual saturation for the combined hydrocarbon phase is calculated, subject to water
blocking. The water blocked hydrocarbon residual saturation is a function of water
saturation and may be entered in tabular or parametric format on the capillary pressure
screen. The parametric model takes the following form.

S owb

kr (S )
= Srow 1 + a w ow w
krw (S w )

If capillary desaturation is present (see residual saturation page for definitions), then the
water blocked oil saturation is modified.

S owb = S owb (1 + S1o (log10 ( Nco ) + S 2 o ))

The hydrocarbon phase residual saturation is then calculated using the gas fraction of
non-blocked hydrocarbon phase (g).

g=

Sg
S o + S g + S owb

Srh = (1 g )Sro + g (Srg + S owb )

The model calculates the relative permeability of the entire miscible hydrocarbon phase,
using both the the oil/water and the gas data.

krh = (1 g )krow + gkrg

Mobile phase saturations (Sm) are calculated for the hydrocarbon phase using either a
simple or interpolated model.

Nc p =

ph

k P + Pc p
144

OCTOBER 2003

REVEAL MANUAL

24 - 84 CHAPTER 3 USING THE INPUT WIZARD

Simple model
permeabilities

Interpolated model

Oil and gas relative

Smow = S o + S g Sro

Smow = S o + S g Srh

krow (Smow )

Sm g = S o + S g S owb Srg

Sm g = S o + S g Srh

krg (Smg )

The oil and gas relative permeabilities are then calculated.

kro = (1 g )krh

krg = gkrh

3.5.2 Residual Saturations

The residual saturations are entered for each rock type permeability curve. Note that the
residual saturations are the same for all directional relative permeability curves for a given
rock type.
Separate residual saturations are required for high and low tension fluids if the
desaturation model is present. Data are required for two phase water/oil and oil/gas
relative permeability curves, and the micro-emulsion phase if the surfactant model is
present.
Gas hysteresis - the critical gas saturation (Srgc) is the saturation at which gas initially
becomes mobile, while the maximum residual gas saturation (Srgm) is used to model gas
hysteresis. The historical maximum gas saturation (Sgm) is used to estimate the current
gas residual (Srg) saturation. The residual gas saturation will always be within the range
defined by Srgc and Srgm, so if both values are identical no hysteresis will take place.

Srg = Srgc +

S gm Srgc
1 Srw Srgc

(Sr

gm

Srgc )

Srgc Srg Srgm

If the gas saturation is within the range [Srgc,Sgm], then the gas phase relative
permeability is calculated by using a normalised gas saturation (Sg*) and the original gas
relative permeability curve.

CHAPTER 3 USING THE INPUT WIZARD

25 -84

saturation. The residual oil saturation will always be within the range defined by Sroc and
Srom, so if both values are identical no hysteresis will take place.

S om Sroc
(Srom Sroc )
1 Srw Sroc

Sro = Sroc +

Sroc Sro Srom

If the oil saturation is within the range [Sroc,Som], then the oil phase relative permeability
is calculated by using a normalised oil saturation (So*) and the original oil/water relative
permeability curve. For the oil/water relative permeability curve, the oil saturation used is
that with no gas present, So=1-Sw.

Som (1 S w )
(Sro Sroc )
Som Sro

Sro* = Sroc +
So* = Sroc +

(1 S w ) Sro* (1 Sr
1 Srw Sro*

Sroc )

Water hysteresis - water phase hysteresis is controlled by altering the residual water
saturation for the water/oil relative permeability curves. A similar interpolation scheme to
the gas hysteresis option is employed for both the low (if desaturation is present) and high
tension water residual saturations. The high and low tension water residual saturations
are calculated first accounting for hysteresis, then interpolated to account for
desaturation.
The critical water saturation (Srwc) is the saturation at which water initially becomes
mobile, while the maximum residual water saturation (Srwm) is used to model water
hysteresis. The historical maximum water saturation (Swm) is used to estimate the
current water residual (Srw) saturation. The residual water saturation will always be
within the range defined by Srwc and Srwm, so if both values are identical no hysteresis
will take place.

Srw = Srwc +

S wm Srwc
(Srwm Srwc )
1 Srwc

Srwc Srw Srwm

If the water saturation is within the range [Srwc,Swm], then the water phase relative
permeability is calculated by using a normalised water saturation (Sw*) and the original
water relative permeability curve.

Srw* = Srwc +

S wm S w
(Srw Srwc )
S wm Srw

S w* = Srwc +

S w Srw*
(1 Srwc )
1 Srw*

The following plot demonstrates the hysteresis in relative permeability, for saturations
below a historical maximum (Smax).

OCTOBER 2003

REVEAL MANUAL

26 - 84 CHAPTER 3 USING THE INPUT WIZARD

Desaturation - if a surfactant model is present, then desaturation parameters are

required to control the interpolation between high and low tension values. The
interpolation is a function of the capillary number10 (Nc) and defined for water(1), oil(2)
and micro-emulsion(3) phases by two parameters each (S1 and S2).

S r = S rh [1 + S1 (Log10 ( Nc ) + S 2 )]
Srl<Sr<Srh, where Srl and Srh are the low and high tension residual saturations.
Wettability - if the wettability model is present, then the interpolation between high and
low tension residual saturations is calculated as a function of the wettability. The
wettability is initialised in the initialisation section and changes as the wetting agent is
adsorbed. A wettability of -1 corresponds to low tension, while a wettability of 1
represents high tension, with linear interpolation for intermediate values of wettability.
The change to the wettability as wetting agent is adsorbed is controlled by the wettability
change parameter defined in the physical section.
10

Capillary desaturation is parameterised by comparing the ratio of viscous and

capillary/interfacial tension forces between phases. The capillary number (Nc) is defined
for each phase.

Nc p =

ph

k P + Pc p
144

PETROLEUM EXPERTS LTD

CHAPTER 3 USING THE INPUT WIZARD

27 -84

3.5.3 Relative Permeabilities

Relative permeability curves must be entered for each region defined in the previous
residual saturation page. Select add or delete to create or delete relative permeability
curves for each region, and select entries in the relative permeability list to modify the
data entered.
Directional relperms - relative permeability curves must be defined for each direction if
directional relative permeabilities have been defined. However, directions with the same
curves may be grouped to prevent excessive repetitive data entry.
Tabular & parametric - for each relative permeability curve, either tabular or parametric
entry may be used. The water/oil curves are for two phase water and oil, while the gas/oil
curves are for two phase oil/gas in the presence of connate water. Therefore, for the
gas/oil curves, the gas saturation range should be from Sgc to 1-Srw, and the
corresponding oil saturation range should be from Srog to 1-Srw. Copy and paste may
be used to enter or edit relative permeability tables. Use the Validate button to check
coverage and the Plot button to view the relative permeability curves entered.
The parametric relative permeabilities take the following form for high and low tension
curves, where the end points (kr*) and exponents (n) are input.

S w Srw
krw = krw

1 Srw Srow

nw

S Srm
krm = krm m

1 Srm

krow = krow

1 S w Sro

1 Srw Srow

now

nm

krog = krog

1 Srw S g Srog

1 Srw Srg Srog

nog

S g Srg
krg = krg
1 Srw Srog Srg
*

ng

The numerators represent the mobile phase fraction, while the denominators represent
the maximum mobile fraction. Note that these correlations are based on the assumption
of an oil reservoir initially containing connate water only. If the mobile phase fraction is
exceeded (due to desaturation, aquifer or gas cap), then these correlations are not
extrapolated and therefore tabular input should usually be used.
For the oil in water relative permeability curves, So = 1-Sw, and for the oil in gas relative
permeability curves So = 1-Srw-Sg.
Surfactant - if the surfactant model is present, then in addition to the low and high
tension water/oil curves and the gas/oil curves, a middle (micro-emulsion) phase curve is
required as a function of mico-emulsion saturation.
Desaturation - for the capillary desaturation and wettability models the interpolation
between high and low tension relative permeability curves is parameterised by capillary
number and wettability in a similar manner to the interpolation used to calculate residual
saturations.

OCTOBER 2003

REVEAL MANUAL

28 - 84 CHAPTER 3 USING THE INPUT WIZARD

0.8

0.7

Relative permeability

0.6

0.5
high tension

0.4

low tension

interpolated
0.3

interpolated residual saturation

0.2

0.1

0
0

0.2

0.4

0.6

0.8

Satur ation

3.5.4 Capillary Pressures

Capillary pressure tables are required for oil/water and gas/oil.
Oil/water capillary pressure (Pco) - Po = Pw+Pco
At the residual (connate) water saturation the water/oil capillary pressure is a maximum,
and should fall monotonically as the water saturation rises. A minimum value of zero
should occur at a water/oil contact (Sw=1). If there is no water/oil contact, the maximum
water saturation will be 1-Sro.
If the J Leverett option is selected, then the dimensionless J function should be entered in
the water/oil table. The water/oil capillary pressure for any block is then calculated (in
oilfield units) by the following formula for each grid block.

Pco =

6.84945 k

where is the water/oil interfacial tension entered in the physical section, is the block
porosity and k is the harmonic average block permeability.
Gas/oil capillary pressure (Pcg) - Pg = Po+Pcg
At zero gas saturation (gas/oil contact) the gas/oil capillary pressure is zero, and should
rise monotonically to a maximum value at a gas saturation of 1-Srw.
Water blocking - if the miscibility option is active then a hydrocarbon phase waterblocked residual saturation is required. This is a function of water saturation and may be
constant, tabular or have a parametric form, parameterised by a wetting parameter. See

PETROLEUM EXPERTS LTD

CHAPTER 3 USING THE INPUT WIZARD

29 -84

the relperm options page for the equation of the parametric form.

3.5.5 Endpoint Scaling

Endpoint scaling is optional, and is only available if the option is selected on the relperm
options page.
Each relative permeability table has nominal endpoints defined as follows.
1] residual (connate) saturation is the minimum saturation entered
2] critical saturation is defined as the maximum saturation for which the relative
permeability is zero
3] maximum saturation is the maximum saturation entered in the table
If endpoint scaling is applied, then relative permeability and capillary pressure tables will
be rescaled between the entered critical and maximum saturation endpoints. Endpoints
are defined over the entire grid.
Water scaled saturations - data for the critical and maximum water saturations may be
entered.
Oil scaled saturations - data for the critical water saturations for oil/water and oil/gas
may be entered.
Gas scaled saturations - data for the critical and maximum gas saturations may be
entered.
Scaled endpoints - the maximum endpoint relative permeability may be rescaled for the
water, oil and gas phases.

OCTOBER 2003

REVEAL MANUAL

30 - 84 CHAPTER 3 USING THE INPUT WIZARD

3.6 Aquifer Section

3.6.1 Analytical Aquifer
This section enters data to define an analytical aquifer model if the aquifer model was
included in the control section. For more detailed information see Carter Tracy.
Region list - several aquifers may be connected to the grid. Use the Add button to add
an aquifer to the list and select an entry in the list to alter its properties. An aquifer may
be deleted by selecting it in the region list and then selecting the Delete button.
Boundary and connection area - for each aquifer the boundary region of the grid
connected to the aquifer must be defined. Note that an aquifer connected to inactive
blocks will not have any effect. Select the aquifer (grid) boundary (X-,X+,Y-,Y+,Z-,Z+)
that the aquifer is connected to and enter the 2D range of blocks, the All button may be
used to select an entire surface of the grid.
General properties - four aquifer models (linear and radial with either finite or infinite
acting) are available (Carter Tracy). The models are solutions to the constant terminate
rate diffusivity equation with different geometries and boundary conditions. For all four
models, some physical properties for the aquifer are required; porosity, permeability,
compressibility, temperature and pressure at a reference depth.
Geometry - depending on the aquifer model chosen, the physical extent of the aquifer is
entered. The contact area is defined by the aquifer thickness (height) and encroachment
angle, and either the aquifer width (linear) or inner radius (radial). The volume of the
aquifer for the finite acting models are defined by the length (linear) and outer radius
(radial).
Component concentrations - the concentration of components in the aquifer are
entered. This will usually be 1 for water and zero for other components.

PETROLEUM EXPERTS LTD

CHAPTER 3 USING THE INPUT WIZARD

31 -84

3.6.2 Cater Tracy model

The inflow to each grid block (IJK) connected to an aquifer is defined.

Qaq

ijk

= ijk ( Aaq Baq ( P ijk (t n +1 ) P ijk (t n )))

where aIJK is the fractional area of grid block IJK connected to the aquifer

ijk =

A ijk
Aijk

Aaq and Baq are aquifer constants calculated for the aquifer model defined by the user,
and are functions of the aquifer influence function (t) and the cumulative influx W(t).

C aq

Baq =

n +1

n +1 t n

t
t

W n (t )
Aaq = (P

C aq t
n

n +1

) Baq

where a dimensionless time t' and an aquifer constant Caq are defined. faq are the
encroachment angles.

Finite linear aquifer

k aq t
t =
aq aq c aq Laq 2

Infinite linear aquifer

k aq t
t =
aq aq c aq

Radial aquifer
k aq t

t =

C aq = f aq aq H aq LaqWaq c aq

C aq = f aq aq H aqWaq c aq

C aq = 2f aq aq H aq Riaq c aq

aq aq c aq Riaq 2
2

The aquifer influence functions are calculated as solutions to a constant terminal rate
diffusion equation.

= 2

t
The aquifer influence function has unit gradient at the inner boundary and zero initial
value. A zero gradient boundary condition is used for a finite aquifer at the outer
boundary, or a zero solution boundary condition for an infinite aquifer. Approximate
solutions to the influence function are obtained at the inner boundary and described
below. The radial solutions are due to Hurst and van Everdingen [1].
Infinite linear aquifer

(t ) = 2

Finite linear aquifer

t 0.15

OCTOBER 2003

t 0.15

REVEAL MANUAL

32 - 84 CHAPTER 3 USING THE INPUT WIZARD

(t ) = 2

1
(t ) = t + 0.20354e ( t )
2

Infinite radial aquifer

t 0.01

(t ) = 2
a = 3.15

0.01 t 1000

(t ) = c +

a
b = 1.15a 1 +
5
c = 0.112 b
s = ln (t )

b
a

t 1000

(s + 2)2 + a 2 (t ) = 0.80907 + ln(t )

Finite radial aquifer

t tc

t tc

(t ) = (t ) +
(t ) = (t )
1.93

Roaq
t c = 0.33
1

Ri

aq
(t ) is the infinite radial solution.

PETROLEUM EXPERTS LTD

2(t t c )
Roaq
1
Riaq

CHAPTER 3 USING THE INPUT WIZARD

33 -84

3.7 Mobility Section

3.7.1 Polymer and Gel
The mobility section is rather complex, since it contains numerous models requiring a
wide range of input data often consisting of correlation parameters with limited physical
significance. Its scope is primarily to describe the modifications to pure phase viscosities
(obtained from the PVT) arising from chemical additives including surfactants, polymers,
gels etc.
This screen initialises the polymer and gel model and sets the shear thinning parameters.
Permeability reduction is implicitly included in the viscosity correlations, since it is used in
the calculation of shear rate.
Polymer/gel viscosity - three models are available for viscosity resulting from polymer
and gel components (original, Carreau and Kuparuk). These models calculate a
thickening factor (TF) for the aqueous and micro-emulsion (if surfactant is present)
phases, subject to shear thinning near wells. The thickening factor is multiplied by the
pure phase viscosity to obtain the modified phase viscosity.
Details of the original and Kuparuk models (for polymer and gel) are given on subsequent
pages of the mobility section, when additional data input is required.
The Carreau model only applies to polymer (not available if gel is present) and requires
no additional data. It is a correlation derived from a xanthan polymer that includes
permeability reduction and shear thinning implicitly.
Shear rate - the shear rate () is defined for each rock type and is parameterised by a
critical shear rate (c) entered on this screen and the zero shear permeability reduction
(R0) (see adsorption section).

Original model
qw
=c
9
k S w krw
8

=c

Carreau and Kuparuk models

qw

9k S w krw
8R 0

The mean permeability in the direction of flow is defined.

2

1 qx w 1 qy w 1 qz w 1

=
+
+
k q w kx q w ky q w kz
If a grid block contains an injection well, then the well shear rate may be modified to
reflect increased shear near the well. The default is for this model to be inactive, see the
well section to activate the well shear model. The well bore radius (rw) and Peaceman
radius (r0) are used to estimate the increased Darcy velocity near the well-bore.

r0

rw

= ln

OCTOBER 2003

REVEAL MANUAL

34 - 84 CHAPTER 3 USING THE INPUT WIZARD

Viscosity calculation - for the original and Kuparuk models permeability reduction may
be subject to shear thinning or not (set in adsorption section). In either case the zero
shear thickening factor (TF0) is modified by the shear rate, using the user input
parameters gamma (h), alpha (n) and thickening parameter (s). The shear thickening
parameter is optional and will only have effect if it is greater than 1.
Shear thinning subject to permeability reduction

TF = 1 +

Shear thinning not subject to permeability reduction

s TF 0 R 0 1

1 +
h

n 1

s
TF
1
TF = R 0 1 +

n 1

1 +
h

3.7.2 Original Polymer Gel Model

This model calculates a zero shear rate (and zero permeability reduction) thickening
factor (TF0) that is modified by the shear rate (see shear rate) to obtain a total thickening
factor (TF). This model applies to both polymer and gel, although the input parameters
are different for the two components. The final thickening factor is multiplied by the pure
aqueous phase viscosity to obtain a modified water viscosity.
Effective polymer salinity - the water component is clearly present in the aqueous
phase, but may also be present in a micro-emulsion phase. Volumetric concentrations of
polymer (cp), gel (cg), salt (cs) and divalent cations (cd) within the water component of the
aqueous and micro-emulsion phases are calculated.
Effective polymer salinity (cse) is defined for the Original and Kuparuk models with

CHAPTER 3 USING THE INPUT WIZARD

3.7.3

35 -84

Carreau Polymer Model

This is a correlation model for polymer viscosity, including shear thinning and permeability
reduction. It is based on a xanthan polymer with data over ranges 50-1000 ppm and 2070 oC and shear 1-100 s-1. Note that the concentrations (cp) are in ppm and the
temperature (T) is in K. The thickening factor (TF) is multiplied by the pure phase water
viscosity to obtain the modified viscosity.

n 1

TF = 1 + 0 1 + 2 2
Zero shear specific viscosity

a4
1
T

0 = a1c p exp(a 2 c p ) a 3 exp

reciprocal of critical shear rate

a5c p

a8
=
exp a 7 c p +
T a6
T a 11

shear thinning constant

a9 c p a10
n = exp
T a6

The parameters a1 to a11 take the following values

a1 = 0.001295
a2 = 0.001784
a3 = 0.01046
a4 = 1910.5
a5 = 0.01577
a6 = 275.0
a7 = 0.0002939
a8 = 2.67
a9 = 2.189
a10 = 0.4334
a11 = 195.3

OCTOBER 2003

REVEAL MANUAL

36 - 84 CHAPTER 3 USING THE INPUT WIZARD

3.7.4 Kuparuk Polymer Gel Model

This model calculates a zero shear rate (and zero permeability reduction) thickening
factor (TF0) that is modified by the shear rate (see shear rate) to obtain a total thickening
factor (TF). This model applies to both polymer and gel. The final thickening factor is
multiplied by the pure aqueous phase viscosity to obtain a modified water viscosity.
Effective polymer salinity - the water component is clearly present in the aqueous
phase, but may also be present in a micro-emulsion phase. Volumetric concentrations of
polymer (cp), gel (cg), salt (cs) and divalent cations (cd) within the water component of the
aqueous and micro-emulsion phases are calculated.
Effective polymer salinity (cse) is defined for the Original and Kuparuk models with
reference to three user input variables, a minimum threshold salinity (cs*), and two further
coefficients, divalent cation dependence (1) and dependence S (2). The minimum threshold
salinity is optional (default = 0.1). The divalent cation dependence is also optional
(default = 1.0) and only applicable is the divalent cation component is present within the
model.

c se = [max(c s *, c s (1 1 )c d )]

Model interpolation temperatures - The Kuparuk model basically interpolates the

correlation between high (TH) and low (TL) temperature limits.
Polymer & gel viscosity parameters - the input parameters for both polymer and gel are
identical.
For the polymer the parameters are Ap1h (p1H), Ap1l (p1L), Ap2h (p2H) and Ap2l (p2L).
For the gel the parameters are Ag1h (g1H), Ag1l (g1L), Ag2h (g2H) and Ag2l (g2L).

) (

TF 0 = 1 + c se p1c p + p 2 c p + g1c g + g 2 c g
2

p1 = f p1L + (1 f ) p1H
g1 = f g1L + (1 f ) g1H
T T
f = H
T H TL

PETROLEUM EXPERTS LTD

p 2 = f p 2 L + (1 f ) p 2 H
g 2 = f g 2 L + (1 f ) g 2 H

CHAPTER 3 USING THE INPUT WIZARD

37 -84

3.7.5 Gelation and Degradation

Reactions between polymer and cross-linker to form gel, and between de-chelation agent
(chelated cross-linker) and cross linker is modelled by calculating mass reaction rates that
are applied to the component mass concentration equations.
Three models are available, depending on the components present. Note that in all of the
models the temperatures may be input in any allowed units, but the correlations are
expressed with absolute (R) temperatures. The concentrations are all volumetric fractions
within the water phase.
2 component model - this model applies when only polymer and gel are present. The
required data are rate coefficient (Rk), temperature coefficient (Bk), temperature threshold
(Tr) and polymer volumetric concentration threshold (Pc0).
If the temperature is greater than Tr and the polymer concentration is greater than Pc0
then a reaction between polymer and gel occurs with mass reaction rates per unit volume
of water for polymer (Rp) and gel (Rg).

1
1
R p = Rk p c p exp Bk

T Tref

1
1
R g = Rk g c p exp Bk

T Tref

3 component model - this model applies when polymer, gel and cross-linker are present.
The required data are rate coefficient (Rk), temperature coefficient (Bk), temperature
threshold (Tr), polymer volumetric concentration threshold (Pc0), cross-linker volumetric
concentration threshold (XLc0), polymer reaction rate multiplier (P0) and cross-linker
reaction rate multiplier (XL0).
If the temperature is greater than Tr, the polymer concentration is greater than Pc0 and
the cross-linker concentration is greater than XLc0, then a reaction between polymer and
cross-linker to produce gel occurs with mass reaction rates per unit volume of water for
polymer (Rp), cross-linker (Rxl) and gel (Rg).

1
1
R p = Rk P0 p c p c xl exp Bk
T Tref

1
1
R xl = R k XL0 xl c p c xl exp Bk
T Tref

1
1
R g = Rk (P0 + XL0 ) g c p c xl exp Bk
T Tref

4 component model - this model applies when polymer, gel, cross-linker and chelated
cross-linker are present. It is identical to the 3 component model, except that the
chelated cross-linker can react to form cross-linker, before the polymer and cross-linker
reaction occurs. The additional parameters for this model are de-chelation rate coefficient
(Rk1), temperature coefficient (Bk1), temperature threshold (Tr1), chelated cross-linker
concentration threshold (Cc0).
If the temperature is greater than Tr1, the chelated cross-linker concentration is greater
than Cc0, then a reaction between chelated cross-linker and cross-linker occurs with
OCTOBER 2003

REVEAL MANUAL

38 - 84 CHAPTER 3 USING THE INPUT WIZARD

mass reaction rates per unit volume of water for chelated cross-linker (Rc) and crosslinker (Rxl).

1
1
Rc = Rk1 c cc exp Bk1
T Tref

1
1
R xl = Rk1 xl cc exp Bk1
T Tref

The 3 component reaction then occurs as described above, except that the cross-linker
reaction rate is the sum of the two reactions involving cross-linker.
Polymer and gel degradation - both polymer and gel may degrade at rates defined by
rate (Rp0 and Rg0) and temperature coefficients (Bp0 and Bg0), reducing the reaction
rates for these components.

1
1
R p = R p R p 0 p c p exp B p 0
T T

ref

1
1
R g = R g R g 0 g c g exp B g 0
T T

ref

3.7.6 Inaccessible Pore Volume

The inaccessible pore volume is used to increase the apparent volumetric concentration
of polymer and gel components. This will affect the mobility correlations using polymer
and gel, and all other models that use polymer or gel concentrations. This is achieved by
dividing the calculated polymer and gel volumetric concentrations by 1-ipv (inaccessible
pore volume) entered on this screen. Inaccessible pore volumes may be entered for each
rock type.

3.7.7 Foam
This model is only available if the foamer component is present. It increases the gas
phase viscosity, thereby reducing its mobility and modelling the production of foam in a
heavy oil. The foamer model may be applied separately to each rock type, or all rock
types.
The concentration of foaming agent in the aqueous phase (cf) and five input parameters
describe the modification to the gas phase viscosity. The input parameters include CRFK
(f1), foam threshold (f2), ES (f3), oil saturation threshold (f4) and EO (f5).

g = g 0 f 1 min(

cf

f2

,1)

f3

max(1

So

f4

,0)

f5

For information on natural foams associated with heavy oils, see the physical section.

PETROLEUM EXPERTS LTD

CHAPTER 3 USING THE INPUT WIZARD

39 -84

3.8 Phase Section

3.8.1 Introduction
The purpose of the phase section is to model the effect of adding a surfactant to the
system. Surfactants emulsify oil and water into a new phase (an emulsion), causing
desaturation and potentially enhanced recovery.
Surfactants are generally used in conjunction with other chemicals, such as alcohols,
which affect the phase behaviour, and polymers, which may also affect the phase
behaviour and which are important for mobility control. Salinity and divalent ion
concentration have a strong influence on phase behaviour, as does temperature and the
equivalent alkane number (EACN) of the oil, which varies as the oil phase composition
(Rs or GOR) varies.
For surfactant flood simulations with REVEAL, up to six volumetric components may be
present (water, oil, gas, surfactant and up to two alcohols). Non-volumetric components
include salinity, and optionally polymer and divalent ions. Additionally tracer components
may also be present. Alcohols are always assumed to be active in the phase behaviour;
polymer may be active, or simply a mobility control agent. Adsorbed material is assumed
to play no part in the phase behaviour calculation.
Surfactant flood systems are described in many books on Petroleum Engineering; only an
outline of the theory is given below.
The phase calculations determine first the phases that are present in a given system, and
from there determine those phase properties. The steps that are taken are:
1. Formation of psuedo-components - the components present are formed into three
pseudo-components - aqueous/brine, oleic, and chemical. These then become the
coordinates on a ternary diagram. All the water is placed in the aqueous pseudocomponent, the oil in the oleic, and the surfactant in the chemical. Tracers, anions, and
gels are placed in the aqueous pseudo-component. Alcohols and, optionally, polymers,
are partitioned amongst all three pseudo-components as described in Alcohol and
Polymer partitioning.
2. Creation of a ternary diagram - a ternary diagram is constructed to determine the
phase behaviour of a system given a set of pseudo-component concentrations and a
calculated effective salinity. A more complete discussion of effective salinity and phase
behaviour is given in surfactant phase model. In brief, there are 4 possible systems (note
the comment below about gas):
a. Two phase oil/water (low/zero surfactant concentration)
b. Two phase oleic/microemulsion (Type II(-) system) (low effective salinity)
c. Two phase aqueous/microemulsion (Type II(+) system) (high effective salinity)
d. Three phase aqueous/microemulsion/oleic (Type III system) (intermediate
effective salinities)
The phase behaviour calculation is performed assuming all components throughout a
gridblock are thoroughly mixed and in equilibrium. This assumption is retained when gas
is present, even if the gas and oil are potentially immiscible; if the phase behaviour
calculation results in an excess oil phase being formed (i.e. the upper phase in type II(-),
type III, or no surfactant cases), then this phase is tested for immiscibility and split into

OCTOBER 2003

REVEAL MANUAL

40 - 84 CHAPTER 3 USING THE INPUT WIZARD

separate oil and gas phases if appropriate. Micro-emulsion phases containing both gas
and oil (upper phase in type II(+) cases, middle phase in type III cases, or single phase
cases) are always assumed to form a single miscible phase, regardless of whether the oil
and gas are miscible. If the Todd and Longstaff model [5] is used, it is applied to all
miscible upper phases (but not to middle phase or single phase micro-emulsions). The
following table summarises the number and type of phases present, and those phases to
which the Todd and Longstaff model is applied.
Phase
behaviour

Single phase

No surfactant

Type III

P>Pcr

1 Phase
Micro-emulsion

2 Phases
Oil/gas

Oil/gas

Oil/gas

Micro-emulsion

Micro-emulsion

Micro-emulsion

Water

Water

3 Phases

Type II(-)
2 Phases

Type II(+)
2 Phases

Water
P>Pcr

1 Phase

3 Phases

4 Phases

3 Phases

2 Phases

Micro-emulsion

Oil

Oil

Oil

Micro-emulsion

Gas

Gas

Gas

Water

Water

Micro-emulsion

Micro-emulsion

Water

Todd and Longstaff model is applied to these phases

The five phase regions identified are numbered from one to five, and may be plotted if
'Phase Region' is selected as a property to be stored during a calculation (Run
Simulation|Select Properties...).
Phase region
number

Description

1
2
3
4
5

Micro-emulsion phase only

Water, oil/gas phases present - no surfactant
Water, oil/gas and micro-emulsion phases present (type III system)
Oil/gas and micro-emulsion phases present (type II(-) system)
Water and micro-emulsion phases present (type II(+) system)

3. Calculation of interfacial tension - the interfacial tension between emulsion and

aqueous phases, and/or emulsion and oleic phases, is calculated using a correlation. See
the interfacial tension section for more information.
4. Calculation of viscosity - the viscosity of the emulsion phase is calculated using a
correlation. See viscosity for more information.

PETROLEUM EXPERTS LTD

CHAPTER 3 USING THE INPUT WIZARD

41 -84

3.8.2 Alcohol/Polymer Partitioning

See also: Introduction.
This screen contains the data for partitioning alcohol and polymer components among the
three pseudo-components. Press Plot to obtain a plot of the partitioning based on the
data that you have entered.
We introduce some symbols, definitions and identities.
Vcp
is the volume of component c in pseudo-component p
Ccp is the volumetric concentration of component c in pseudo-component p
cc
is the total volumetric concentration of component c
Vp
is the volume of pseudo-phase p
V
is the total pseudo-phase volume
vp
is the volumetric concentration of pseudo-phase p
Kcp
is the partitioning constant for component c between pseudo-phase p and the
aqueous pseudo-phase
Define pseudo-phase subscripts for the aqueous (a), oleic (o) and chemical (c) pseudocomponents, and component superscripts for the water (w), oil (o), surfactant (s), alcohol
(x), salinity (n), polymer (p) and divalent ion (d) components.
Component volumetric concentrations
identities

C ax =

Vax
Vaw

C ox =

Vox
Voo

C cx =

Total component concentration

cx =

Vx
V

vp =

c =C c +C c +C c
x
a

x
o

x
c

V = V p

= cc = 1
c

Vp
V
Polymer material balance

Volume

Pseudo-phase volumetric concentration

Alcohol material balance

x

Vcx
Vcs

Total phase volume

c = C ap c w + C cp c s = C ap c w + K p c s
p

Polymer partitioning - when the polymer is active in the phase behaviour, it is partitioned
between the aqueous and chemical pseudo-components according to a partition
coefficient Kp, which may vary linearly with temperature.

K p = K p ref + p (T Tref ) =

C cp
C ap

If a polymer is present but not active (i.e. it is simply a mobility control agent), polymer
partitioning may be switched on, in which case the polymer is put in the aqueous pseudocomponent, so that it partitions between phases in proportion to the amount of water in
each phase, or polymer partitioning may be switched off, in which case the polymer is
placed entirely in the aqueous phase at the end of the phase behaviour calculation, and
thus is prevented from partitioning into any middle or upper phase that forms.
Alcohol partitioning - alcohols are partitioned amongst all three pseudo-components,
according to one of two partitioning models. Data for one or two alcohols is required,
depending on the number of alcohol components in the model.
1. Hirasaki model - this is the default alcohol partitioning model. The alcohols are
partitioned between the pseudo-components with partition coefficients that are constant

OCTOBER 2003

REVEAL MANUAL

42 - 84 CHAPTER 3 USING THE INPUT WIZARD

(at a given temperature).

Partitioning constants relating partitioning of the alcohol components between the three
pseudo components are defined. The left column of data input correspond to the
aqueous/oleic partitioning and the right column corresponds to the aqueous/chemical
pseudo-component partitioning. K are the Alc coefficients and are the Alc Temp Dep
variables.

K ox =

C ox
C ax

K cx =

C cx
C ax

The partitioning constants vary linearly with temperature.

K ox = K ox ref + ox (T Tref

K cx = K cx ref + cx (T Tref

2. Prouvost model [15,16] - this alternative partitioning model, due to Prouvost, allows
for self-association of alcohols in the oleic pseudo-component and treats the chemical
pseudo-component as an interfacial pseudo-component. The alcohol partition
coefficients are calculated from more fundamental coefficients in each grid block at each
time step, according to the overall composition in the grid block. No provision is made for
these coefficients to be temperature dependent.
For each alcohol, the model requires the input parameters: oil/water partition (ko(1)),
chemical/water partition (ko(2)), oil/chemical partition (kc), equilibrium constant (k) and
partition parameter (a).

Salinity-dependent oil/water partition coefficient

C ax

x
c n 1 + C a
x
x
x
k o = k o (1) k o ( 2) w =
c
C ox

x
1 + Co
C ax
Vax
cn
=
x
w
Vaw + Vax is the volume fraction of alcohol in
where c is the salinity, and 1 + C a
aqueous phase.
ko is the ratio of the volume fraction of alcohol in the aqueous pseudo-component to the
volume fraction of monomeric alcohol in the oleic pseudo-component.

Oil/surfactant partition coefficient

C cx
x

a +Cx
c

x
kc =
C ox

x
+
1
C
o

kc is the ratio of the surface fraction of the alcohol in the chemical pseudo-component to
the volume fraction of monomeric alcohol in the oleic pseudo-component.
a is the ratio of the length of an alcohol molecule to the length of a surfactant molecule.

PETROLEUM EXPERTS LTD

CHAPTER 3 USING THE INPUT WIZARD

43 -84

i-meric alcohol concentration

kx
C ox ,i +1 =
C ox ,1C ox ,i
x
1 + Co

C x ,i

where o is the ratio of the volume of i-meric alcohol in the oleic pseudo-component to
the total volume of that pseudo-component.
k is a self-association equilibrium constant.
These equations, together with the equation for the overall material balance of the
alcohol, can be rearranged to express all quantities in terms of Co resulting in a cubic
equation for Co. Thus the solution can be found analytically.
When a second alcohol is present, similar input parameters are required for the second
alcohol, and analogous equations apply. Simultaneous solution of both sets of equations
then requires an iterative method. A quasi-Newton method is used; the user may set the
convergence criterion (Alcohol threshold), and a limit on the number of iterations, which if
exceeded produces an error message.

3.8.3 Surfactant Phase Model (Micro-Emulsion)

See also: Introduction.
This screen displays the data used to set up the ternary phase diagram.
Press Plot to view a plot of the data. You can choose between a plot of effective salinity
(against salinity) or a full ternary phase diagram.
Pseudo-component concentrations - since only water, oil, gas, surfactant and alcohols
are allowed to be volumetric in surfactant flood simulations, the volumetric concentrations
(volume fractions) of the pseudo-components are given by (see Polymer/Alcohol
partitioning for definitions):

(
= c (1 + C
= c (1 + C

v a = c w 1 + C ax =1 + C ax = 2
vo
vc

x =1
o

+ C ox = 2

x =1
c

+ C cx = 2

)
)+ c
)

These volumetric concentrations (which should sum to one) are used as the coordinates
in the ternary diagram calculations.
Effective salinity - the effective salinity is defined by:

C SE =

1
1
cn
(
1 + x =1 f cx =1 + x = 2 f cx = 2 ) (1 + p f c p ) (1 + T (T Tref
w
c

)) (1 + (E E ))
1

ref

Where cn/cw is salinity and fc are volume fractions of components in the chemical
pseudo-component. The coefficients are user supplied and entered in the effective
salinities table for alcohol-1, alcohol-2, temperature, polymer and EACN. Tref is the
reference temperature defined in the control section.

f cx =

C cx
1 + C cx =1 + C cx = 2

OCTOBER 2003

REVEAL MANUAL

44 - 84 CHAPTER 3 USING THE INPUT WIZARD

f cp

C cp

K p C ap =

K pc p
cw + K pcs

CHAPTER 3 USING THE INPUT WIZARD

45 -84

composition at CSEL or CSEU.

Threshold for phase behaviour - in all phase environments, if the total amount of
surfactant present in solution is less than a threshold value (Chem threshold) entered in
the three-phase boundary section of this screen, two phases only are assumed to be
present, one containing only the oleic pseudo-component, the other the aqueous and
chemical pseudo-components.
Binodal curve - in each diagram, only a single phase will be present if the overall
composition lies above the binodal curve, which is assumed to be symmetric and to be
given by the Hand equation:

vc2 = Ava vo
The Hand parameter A is related to the "height" of the curve, i.e. the maximum chemical
concentration. At maximum chemical concentration, the hand parameter may be
calculated since vc,max+va+vo=1 and va=vo at the maximum chemical composition.

vc2,max = Ava

2
2vc ,max
1 vc ,max

A =
= A

1
v

2
c , max

, therefore

It is found by linear interpolation in the effective salinity from values calculated at zero,
optimal, and twice optimal effective salinity (denoted by subscripts j = 0, 1, 2 respectively):

C SE < C SEOP

C SE > C SEOP

A = A0 + ( A1 A0 )

C SE
C SEOP

A = A1 + ( A2 A1 )

C SE C SEOP
C SEOP

where the values Aj are found from the binodal curve heights at the three effective
salinities:

2vc , j max
Aj =

1 v c , j max

which in turn are specified as (linear) functions of alcohol concentration, polymer

concentration, temperature, and EACN:

vc , j max = H BNC , j + H BNX 1, j f cx =1 + H BNX 2, j f cx = 2 + H BNP , j f c p + H BNT , j (T Tref ) + H BNE , j (E E ref

The H parameters are entered in the Binodal parameters section of this screen for zero,
optimal and twice optimal salinity, for a base value (HBNC), both alcohols (HBNX1 and
HBNX2), temperature (HBNT), polymer (HBNP) and equivalent alkane number (HBNE).
Three phase triangle - in the type III case, three phases exist if the overall composition
point lies in the three phase triangle below the binodal curve. The compositions of the
three phases are given by the vertices of the triangle, i.e. pure aqueous pseudocomponent, pure oleic pseudo-component, and a middle phase containing all
components, generally with a high chemical concentration but also with significant
amounts of both oil and water, implying high solubilisation parameters (ratios of oil or
water to surfactant). The position of the invariant or middle point (i.e. the apex of the

OCTOBER 2003

REVEAL MANUAL

46 - 84 CHAPTER 3 USING THE INPUT WIZARD

triangle) is assumed to move across the diagram from left to right as the effective salinity
increases from CSEL to CSEU; to be precise, the value of the expression

1 v a vo
+
2
2

increases linearly from 0 at CSEL to 1/2 at CSEOP to 1 at CSEU.

Two phase regions (tielines) - when two phases only are present, either in a II(+) or II(-)
phase environment, or if the composition point lies in one of the lobes between the threephase triangle and the binodal curve of a type III phase diagram, the compositions of the
two phases are given by the intersections with the binodal curve of the tieline passing
through the overall composition point. The locus of the tieline is determined by the
position of the appropriate plait point. This is the point on the binodal curve on which the
pseudo-component compositions of emulsion and excess phase (aqueous or oleic) are
identical - it is essentially a critical point. Two plait points are entered by the user in terms
of a pseudo-oil concentration - a left hand plait point and a right hand plait point. In the
case where the plait points are in the corners of the ternary diagram (0.0 and 1.0), the
excess phases will consist of entirely the oil or water pseudo-component.
In the type II(-) and type II(+) cases, the following Hand relation, known as the distribution
function, relates the (pseudo-component) compositions of the two phases:

C os C aw = EC as C oo
The Hand parameter E can be determined from the position of the plait point, the point on
the binodal curve to which the tielines converge. At the plait point, both phases have the
same composition (that of the plait point), and this composition lies on the binodal curve,
so:
o

E=

C
C

w
Pa
o
Po

1 CP o

1
o
AC P o +

(AC )

o 2
Po

o
+ 4 ACp oo 1 C P o

CP o

where C P o is the pseudo-oil concentration at the plait point. In type II(-) diagrams,
o
o
o
o
C P o = C PR o . Similarly, in type II(+) diagrams, C P o = C PL o . The plait points are input
in the binodal parameters section of the screen.
For the two phase lobes of a type III diagram, the distribution function is modified to
reflect the fact that the base of the lobe is a side of the three phase triangle rather than
the base of the ternary diagram, becoming:

C o s C a w = E C a s C o o
where for the type II(+) lobe:
C a o = C ao
C a s = C as

C mo + C mc
C mo

C ao C mc
C mo

C a w = 1 C a o C a s

PETROLEUM EXPERTS LTD

CHAPTER 3 USING THE INPUT WIZARD

47 -84

C P a

E =

C P o
o

where mk is the composition of the invariant point. Similar formulae apply for the type
II(-) lobe. The position of the plait points is taken to vary linearly with effective salinity
between the type II(-) and type II(+) values, so the pseudo-oil concentration of the plait
point in the II(+) lobe is given by:
o

CP o =
o

CP o

C SEL
1
o C
C PL o SE

2
C SEOP C SEL

for C SE < C SEOP

C C SEOP
1
o
= C PL o 1 + SE

2
C SEU C SEOP for C SE > C SEOP

and that in the II(-) lobe by:

o

C P o = C PR o +
C P o = C PR o +

) CC

1
o
1 C PR o
2
1
o
1 C PR o
2

C SEL

SEOP C SEL for C SE < C SEOP

C SE C SEOP

1 +
C SEU C SEOP for C SE > C SEOP
SE

In addition to satisfying the distribution function equation, the compositions of the two
phases must satisfy the binodal curve equation:
2

C as = AC aw C ao

and

C os = ACow C oo

and the relations:

C aw + C ao + C as = 1 and C ow + C oo + C os = 1
C aw S a + C ow S o = v a and C ao S a + C oo S o = vo and S a + S o = 1
which gives a total of 8 equations for the eight unknowns

C aw , C ao , C as , C ow , Coo , C os , S a

and o . In the general case, these equations are solved for the pseudo-oil concentration
in the micro-emulsion phase by an interval-halving iteration. In the special case where the
plait point lies in the corner of the phase diagram, the composition of the excess phase is
known (it is the pure pseudo-component), while the mico-emulsion phase has the same
ratio of chemical pseudo-component to oleic (type II(+)) or aqueous (type II(-)) pseudocomponent as the overall composition point; the equations can then be solved explicitly,
and no iteration is necessary.
A ternary diagram generated with the data entered on this screen can be viewed by
clicking the Plot button. Some examples that illustrate the above discussion are displayed
below.

OCTOBER 2003

REVEAL MANUAL

48 - 84 CHAPTER 3 USING THE INPUT WIZARD

PETROLEUM EXPERTS LTD

CHAPTER 3 USING THE INPUT WIZARD

OCTOBER 2003

49 -84

REVEAL MANUAL

50 - 84 CHAPTER 3 USING THE INPUT WIZARD

PETROLEUM EXPERTS LTD

CHAPTER 3 USING THE INPUT WIZARD

51 -84

3.8.4 Ternary Diagram

See also: Introduction.
This plot is invoked from the surfactant phase behaviour screen.
This screen can be used to view the data on the phase behaviour screen graphically.
Properties such as alcohol and polymer concentration, as well as component
concentrations of water and oil (for example) can be varied to see how the various
phases develop with these properties. The screen itself is divided into several sections:
Component Concentrations - these sliders can be used to set the relative
concentrations of oil, water, and surfactant to determine in turn the pseudo-component
concentrations. The values are normalised automatically - if you want to fix a value while
changing another, check the 'fix' box next to the slider.
Note that gas is not included in these basic calculations. The reason for this is purely to
simplify the input parameters: if gas were included, a PVT calculation would need to be
carried out at P,T, and the resulting gas concentration would depend on factors such as
miscibility and whether or not the system were repressurised. In the simulation, gas forms
part of the oleic pseudo-component as explained in the introduction to this section.
The absence of gas means that the binodal curve and effective salinity will not depend on
the EACN entered for the gas phase.
Variable Parameters - the following input parameters can be varied: alcohol
concentration, polymer concentration, salinity, and temperature. When the sliders are
adjusted, the phase diagram will change accordingly.
Calculation Results - the Hand parameter of the binodal curve and effective salinity for
the input set are output. A description of the system at the overall concentration point is
also given (e.g. 'two phase aqueous and oleic emulsion' for a type II(+) system).
Pseudo-component concentrations in phases - this gives the pseudo-component
concentrations of the resulting phases. In II(-) and II(+) systems, the phase concentrations
are given by the intersections of the tieline with the binodal curve, as plotted. For the type
III case, the concentrations are simply the vertices of the three phase triangle.
Interfacial tensions - Interfacial tensions for the emulsion/aqueous interface and the
emulsion/oleic interface are given, where applicable, if this input data has yet been
entered.
Plot Phase Composition - this presents a diagram with the relative phase saturations for
the system at the overall composition point.
Zoom - the chemical pseudo-component concentrations along the binodal curve are often
quite small and this makes the detail of the plot quite difficult to see. Click on the 'zoom'
arrows to effectively increase the chemical concentrations (note that this will distort the
plot slightly as the concentrations still have to be normalised).

OCTOBER 2003

REVEAL MANUAL

52 - 84 CHAPTER 3 USING THE INPUT WIZARD

3.8.5 Surfactant Interfacial Tension

The interfacial tension between aqueous, micro-emulsion and oleic phase are modified by
the surfactant model, resulting in capillary desaturation.
Once the phase compositions have bee calculated, the interfacial tension is calculated as
a function of solubilisation parameters (X). For an interface between water and microemulsion (type II(+) environment or the II(+) lobe of a type III environment) the
water/micro-emulsion interfacial tension (wm) is calculated. Similarly for an interface
between micro-emulsion and oil (type II(-) or the II(-) lobe of a type III environment) the
oil/micro-emulsion interfacial tension (om) is calculated.

Water/micro-emulsion
G11
log 10 ( wm ) = G12 +
1 + G13 X w

Oil/micro-emulsion
G 21
log 10 ( om ) = G 22 +
1 + G 23 X o

C cw
Xw = c
Cc

C co
Xo = c
Cc

In both cases, if the solubilisation parameter (Xw or Xo) is below a threshold Xmin
(minimum solubilisation parameter, default = 1), a linear interpolation between the
oil/water value (ow, entered in the physical properties section) and the the values given
by the formulae above is used.

log 10 ( wm ) = (1

Xw
X
) log 10 ( ow ) + w log 10 ( wm )
X min
X min

if

X w < X min

log 10 ( om ) = (1

Xo
X
) log 10 ( ow ) + o log 10 ( om )
X min
X min

if

X o < X min

Additionally, the interfacial tension is forced to approach zero at the plait point by
multiplying the calculated value of by

1 exp( )

1 exp( 2 )

2 = C pw C qw

) + (C
2

o
p

C qo

) + (C
2

s
p

C qs

where p and q denote the two phases between which the interfacial tension is being
calculated.
In the three-phase triangle of the type III phase diagram only, the interfacial tension may
be made a function of overall surfactant concentration at concentrations below a specified
threshold (), where the slope (S) is user defined.

cs

w
c

log 10 ( wm ) = log 10 ( wm ) + S 1

w
c

log 10 ( om ) = log 10 ( om ) + S 1

PETROLEUM EXPERTS LTD

CHAPTER 3 USING THE INPUT WIZARD

53 -84

3.8.6 Surfactant Viscosity

If the surfactant model is being used, then the aqueous, oleic and micro-emulsion phase
viscosities are calculated using one of two models, namely Pope and Barker (a variant of
the Barker model is also available). See the mobility section for more information on
viscosity models for polymer (thickening, shear thinning). At this stage, of a calculation,
the aqueous, oleic and gaseous phase viscosities have been calculated, accounting for
all additive except surfactant.
Each of the three phases (p), aqueous, oleic and micro-emulsion are considered in turn.
Within each phase, the volumetric component concentrations of water (Cwp), oil (Cop)
and surfactant (Csp) are known.
In addition to the model specific parameters required the user enters a thermal exponent
factor Alpha6 (mT). Note that this factor is used with absolute temperatures (R).
Pope model - the user is required to enter the parameters Alpha1 to Alpha5 (m1 to m5).
The volumetric concentration of hydrocarbon components (Chp) comprising oil+dissolved
gas is calculated.

p = C ps m1 exp mT

exp( m 2 C pw + m 3C op ) + C pw w exp( m 4 (C op + C ps )) + C ph o exp( m 5 (C pw + C

1
Tref

w and o are the water and oil viscosities calculated without surfactant being present.
When the micro-emulsion phase is being considered, o is modified to reflect any miscible
free gas within this phase. Note that no free gas will be present within the oleic phase.
The term free gas is used here to mean excess gas (GOR-Rs) beyond that associated
with the oil, that may be present within the micro-emulsion phase.

o =

o
g

g + (1 g )
o
g

1
4

1
4

g=

where

Rs

GOR
cg

The above sum is over all gaseous components (usually only one).
Barker model - the user is required to enter the parameters Alpha1 to Alpha5 (m1 to
m5). These are different from parameters entered for the Pope model.

Aqueous phase viscosity Oleic phase viscosity

w = TFw ( w0 + m1C

s
a

(Twax T )
o = ( o 0 R 0 R + m 2 C os )exp wax
(1 + Twax T )

The micro-emulsion phase viscosity depends on the phase present (i.e. region of ternary
diagram). The phases present are controlled in REVEAL by the effective salinity (cse),
which has values within a range extending from a minimum (cse,min) to a maximum
(cse,max), with an optimal value (cse,opt) between the minimum and maximum values.
Type II- (oleic and micro-emulsion phases only)

m = o

OCTOBER 2003

REVEAL MANUAL

54 - 84 CHAPTER 3 USING THE INPUT WIZARD

Type II+ (aqueous and micro-emulsion phases only)

m = w

Type III (aqueous, oleic and micro-emulsion phases)

m = m 3 + ( m 4 m 3 )

m = m 4 + ( m 5 m 4 )

(c

1
c se ,min )
1
exp mT
(cse,opt cse,min ) T Tref
se

(c

1
c se ,opt )
1
exp mT
(cse,max cse,opt ) T Tref
se

for

cse cse,opt

for

cse > cse,opt

For the type III micro-emulsion, the micro-emulsion viscosity is modified by the thickening
factor calculated for the water present within the micro-emulsion phase. This modification
is not applied if the variant Barker_x model is selected.

m = TFm m
The calculated micro-emulsion viscosity is then modified to reflect the presence of any
free gas.

m =

g m + (1 g ) g
g

1
4

PETROLEUM EXPERTS LTD

1
4

g=
where

Rs

GOR
cg

CHAPTER 3 USING THE INPUT WIZARD

3.9

55 -84

Adsorption Section

3.9.1 Adsorption Properties

This screen sets the components that will adsorb onto rocks, and defines the isotherm
describing the adsorption.
Any component apart from water, oil and gas may be adsorbed. Adsorbed components
are not transported with the fluid flow or subject to diffusion. Adsorbed minerals take no
part in water chemistry equilibrations. Adsorbed surfactant takes no part in desaturation
and micro-emulsion phase formation. Adsorbed components may reduce the
permeability (e.g. polymer, gel or water chemistry minerals). Adsorbed wetting agent is
used to model wettability desaturation.
Data for rock types - adsorption data may be added for any or all rock types, with
different data possible for different rock types. Select Add to add a new entry to the table,
select an entry in the table to modify the adsorption data, or Delete to delete all of the
adsorption data for the current rock type.
Fluid list - this lists the components in the model that may be adsorbed. Select a
component and use the Add or Delete buttons to add or delete adsorption data for that
component. Some components must have adsorption data (e.g. surfactant, polymer, H+)
if they are present, these components will have a red cross beside them and use the Add
button to add adsorption data (which may be null if no adsorption is required).
Adsorption isotherm - Four adsorption models are available, Langmuir, Linear, Stripping
and Bright Water. Use the Plot button to view the isotherm.
Permeability reduction - this is available for all adsorbed components apart from water
chemistry minerals, which have their own permeability reduction model. The zero shear
permeability reduction (R0) is a function of the adsorbed component concentration (Ca)
and the input permeability reduction coefficient (RF). See the mobility section for more
information on how the permeability reduction is used.

R 0 = 1 + ( R F 1)

Ca
Ca max

The total permeability reduction is the sum of the permeability reductions of all adsorbing
components.
Adsorption initialisation - the initial adsorbed concentration is entered in units of mass
per unit volume, where the volume is the total rock volume (including pore volume). This
data is optional, with a default of zero. Data may be entered as a constant for each rock
type, as a set of grid block ranges each with a constant value, or as a datablock11 with an
entry for each grid block. For more information on using the range and array options see
the reservoir section.

11

Potentially large arrays of data (e.g. grid coordinates) held in the REVEAL archive but not
displayed in the input script.

OCTOBER 2003

REVEAL MANUAL

56 - 84 CHAPTER 3 USING THE INPUT WIZARD

3.9.2 Adsorption Isotherms

Four isotherm models are included, describing the adsorption profile as a function of
concentration, temperature, permeability and salinity. Irreversible, reversible or partially
reversible models are available, and the adsorption may also be rate dependent.
All component concentrations have dimension of mass/volume. Mass fractions (c) in the
aqueous phase are defined by the ratio of a given component concentration divided by
the concentration of water (c=C/Cw).
The partitioning of the total component concentration (C) between that adsorbed (Ca) and
that remaining within the aqueous phase after adsorption (C*) is defined by the maximum
adsorbed concentration (Camax,A[k]) and the isotherm parameter (b), which describes
the shape of the isotherm.
Langmuir isotherm
The relation between adsorbed and aqueous component concentrations is characterised
by the following equations, where a is defined below.

Ca =

ac *
1 + bc *

c* =

C Ca
Cw

The maximum adsorbed concentration occurs when c* is large (i.e. Camax=a/b). If b is

small, then Ca rises slowly, and if b is large then Ca rises quickly.
In addition to Camax and b, the user enters the temperature (A[kt]), permeability (A[kx])
and salinity (A[ks]) coefficients used to calculate the parameter a (maximum adsorbed
concentration at current conditions of temperature, permeability and salinity).

a = a ref + a S (C S ) + aT (T Tref ) k

ax

a ref = bCa max

k =

kx 2 + ky 2
2

The effective salinity (CS) is calculated from the salts present, while the reference
temperature (Tref) is input by the user in the control section. The mean permeability (k) is
the mean horizontal permeability.
Linear isotherm
The linear isotherm is identical to the Langmuir isotherm, with the following modification.

If c* b then Ca = ac *
If c* > b then Ca = ab
Bright Water isotherm
The Bright Water isotherm is identical to the Linear isotherm, with the following
modification to the temperature dependence, requiring input parameters Tbw (A[kt]), abw
(C[bw]) and (epsilon).

a = a ref + a S (C S ) + tanh (abw (T Tbw )) k a x

PETROLEUM EXPERTS LTD

CHAPTER 3 USING THE INPUT WIZARD

57 -84

Stripping isotherm
The stripping isotherm adsorbs all available component to its maximum value, hence the
isotherm parameter (b) is not required for this model.

Ca = max(C , Ca max )

Reversibility
Adsorption is always completely reversible for changes in salinity.
Adsorption may be completely reversible, or completely irreversible for changes in
temperature. If the irreversible option is selected, then the maximum historical thermal
contribution to the adsorption parameter (a) is used.
Adsorption may be completely irreversible, completely reversible, or partially reversible for
changes in concentration for all isotherm models. Reversibility is controlled by entering
an irreversible concentration limit (Cairrev). If Cairrev is not entered (cell left blank), then
the model is assumed to be fully reversible. A Cairre value equal to Camax represents
the fully irreversible case, while intermediate values model partially reversible isotherms.
For a partially reversible isotherm, as the concentration of the adsorbing component
increases, the adsorption level rises in accordance with the isotherm. The Langmuir
curve is used as an example, where three points on the curve (Ca plotted as a function of
C*) are identified.
Point A is at the irreversible adsorption concentration (Ca = Cairrev).
Point B is the point on the isotherm curve where a straight line passing through (0,Cairrev)
has the same gradient as the isotherm curve (i.e. the tangent point).
Additionally, point C is defined as the point (Cairrev,0) used to define point B.
If the concentration falls before point A has been reached, then the adsorption does not
reverse (irreversible). If the concentration starts to fall between point A and point B, then
desorption occurs along a straight line from the maximum point reached on the isotherm
curve to point C. If the concentration begins to increase again, adsorption moves back up
the straight line until the isotherm is reached, and then continues moving up the isotherm.
If the concentration starts to fall after point B, then desorption proceeds down the
isotherm until point B is reached and then down the straight line connecting point B to
point C.

OCTOBER 2003

REVEAL MANUAL

58 - 84 CHAPTER 3 USING THE INPUT WIZARD

Adsorped concentratoin (Ca)

Reversible Langmuir Isotherm

B
C
A

Concentration in solution (C*)

Adsorption rate
Adsorption may be assumed to take place instantly, or the time taken for adsorption to
occur may be modelled using an input rate parameter (arate).

Ca n +1 = Ca n + a rate t (Ca Ca n )
where Ca is the newly calculated instantaneous adsorbed concentration, and Can and
Can+1 are previous and current time step values.

PETROLEUM EXPERTS LTD

CHAPTER 3 USING THE INPUT WIZARD

59 -84

3.9.3 Permeability Reduction

Water chemistry permeability reduction - for adsorbed water chemistry minerals, both
porosity () and zero shear permeability reduction (R0) are functions of the adsorbed
concentration, rock density (r), original porosity (0) and user defined permeability
reduction coefficient (PN).

Ca

R 0 = (1

) P

Permeability reduction coefficients may be defined for each rock type or for all rock types.
The total permeability reduction is the sum of the permeability reductions of all adsorbing
components. See the mobility section for more information on how the zero shear
permeability reduction is used.
The porosity reduction is used as an equivalent inaccessible pore volume (ipv = -/),
which increases the effective volumetric concentration of all trace components (i.e. water
chemistry ions).
Other permeability reduction factors - the shear dependence flag is available for
viscosity correlations when polymer or gel components are present. It allows permeability
reduction to be subject to shear thinning or not. See the mobility section for details of the
shear model.
By default permeability reduction only applies the aqueous phase, however, it may also
be applied to oleic phase by selecting the Oil+water phase option and entering a fraction
(PSPLIT) of the permeability reduction to be applied to the oleic phase. This option does
not affect the permeability reduction applied to the aqueous phase.

OCTOBER 2003

REVEAL MANUAL

60 - 84 CHAPTER 3 USING THE INPUT WIZARD

3.10 Water Chemistry Section

This section is only available for water chemistry calculations. The component model for
water chemistry calculations is set in the control section. The water chemistry model
predicts precipitation and dissolution of minerals within the aqueous phase.
H+ (pH), e- (pe) are required species and should be included for all water chemistry
models. Solid phase mineral concentrations are similarly input and tracked within the
aqueous phase.
PHREEQC
The public domain water chemistry program PHREEQC is used to solve the chemical
equilibrium within the water phase. It takes as input the aqueous master and mineral
species molalities (mol/kgw), and returns their molalities following equilibration and
mineral reaction.
Redox reactions are possible, altering the electron activity, e.g. HS- <> SO42- + 9H+ + 8e- 4H2O. Note that the redox reactions may have significant effect on the solution pH and
pe.
The reaction species and thermodynamic reactions are defined within a PHREEQC
database.
A default PHREEQC database is included with the REVEAL installation, and is written to
the current data directory with the name 'phreeqc.dat'. The current data directory is set
using the menu option File|Data Directory. To use a different chemical database, simply
replace the 'phreeqc.dat' file in the current data directory, ensuring that the replacement
database has the correct format for version 2.4.2 of PHREEQC.
Partitioning - the PHREEQC model also returns the molality of dissolved CO2 and H2S.
The CO2 and H2S may be partitioned between the aqueous and oil/gas phases.
CO2 (oil/gas) <> CO32- + 2H+ - H2O (aq)
H2S (oil/gas) <> HS- + H+ (aq)
The change in pH can dramatically affect the precipitation/dissolution of mineral species.
Adsorption - precipitated minerals may be adsorbed. Adsorbed minerals are not
transported and take no part in the water chemistry equilibration.
Output region - this defines are region of the grid that will have the results of the water
chemistry equilibration output to the water chemistry results file. This file may be viewed
by selecting Results|Debug|Debug water chemistry from the main REVEAL window.
Calculation/output frequencies - the water chemistry equilibration may be performed
every timestep or with a defined frequency in timesteps. Similarly, the water chemistry
output may be output every timestep or with a defined frequency.

PETROLEUM EXPERTS LTD

CHAPTER 3 USING THE INPUT WIZARD

61 -84

Electrical neutralisation - one species pair (anion and cation) is required to be internally
adjustable to ensure precise electrical neutrality. This is usually chosen to be Cl- and Na+.
Electrical neutrality of the mixture of ionic species (excluding free electrons, e-) should be
ensured for both the reservoir and injected waters. This may be checked using the input
wizard calculator in the initialisation and injection sections.
New partition coefficient - if this option is set, then the water/oil partition coefficients for
CO2 and H2S will be initialised using the initial components present, rather than using the
value entered in the mobility section. See the mobility section for more information on the
partitioning model. This will only affect water chemistry models where partitioning can
take place (i.e. the components CO2oil/CO2gas or H2Soil/H2Sgas are present).
Simplified equilibration - this flag is only available if there are no mineral precipitate
components within the model. If this option is selected, then the PHREEQC chemical
equilibrium program is not used, and therefore aqueous speciation of the tracked master
species does not take place (i.e. the ionic master species do not associate to form
aqueous dissolved species). However, H2S and CO2 partitioning and the H2S souring
models are still available. These models calculate the aqueous dissolved CO2 and H2S
using the PHREEQC database (for equilibrium constants and activity coefficients),
assuming that all of the ionic master species (CO3-2 and HS-) are in ionic form (fully
dissociated) and available to form aqueous dissolved CO2 and H2S. This approximation
results in large cpu savings.
Master species components - chemical components (ions) corresponding to a master
set of reacting species are input and tracked. The water chemistry chemical components
are defined to have the same density as the water present, therefore volumetric and mass
concentrations are equivalent. Usually their concentrations are input as mass
concentrations (e.g. ppm).
Note that the H+ and e- concentrations are related to pH and pe with the following
relations.

[ H + ] ppm = 10 (3 pH )

and

[e ] ppm = 10 (3 pe )

Seawater - a typical seawater composition is given below. Both sulphur valence states
are included; this is a general requirement and all ions with multiple valence states should
be included. Additional reservoir species (e.g. Ba+2) and possible minerals (e.g. Barite)
should be included in the model.
Ion
H+
eCa+2
Mg+2
Na+
K+
ClCO3-2
SO4-2
HS-

Concentration (ppm)
6E-6 (pH=8.22)
3.5E-12 (pe=8.45)
412
1292
10768
399
19353
142
2712
0

OCTOBER 2003

REVEAL MANUAL

62 - 84 CHAPTER 3 USING THE INPUT WIZARD

3.10.1

CO2 and H2S Partition Coefficients

This screen is present if the water chemistry model is active and partitioning of CO2 or
H2S between aqueous and hydrocarbon phases is possible. For CO2 partitioning, the
components CO2oil and CO2gas must be included in the model, and for H2S partitioning
H2Soil and H2Sgas must be included (see control section for component selection).
CO2 partitioning - Partition coefficients (PCwo and PCog) are defined for CO2
partitioning between aqueous and oleic phases, and between oleic and gas phases. The
activity of CO2 in water is approximately equal to the molality (mole/kg water) of dissolved
CO2. The partitioning is achieved by considering the CO2 dissolved in the hydrocarbon
phases to be represented by an ideal gas in equilibrium with the aqueous phase. The
ideal gas CO2 partial pressure is related to the mass concentration of CO2 within the
hydrocarbon phase.
PCwo = (Mole/litre of CO2 in oil) / (Activity of CO2 in water)
PCog = (Mole/litre of CO2 in gas) / (CO2 concentration in oil)
The partition coefficients are pressure dependant and take different forms depending on
whether the pressure is above or below the partition reference pressure.
If the pressure is above the partition reference pressure.
PCwo = Partition coefficient + (Pressure-Partition reference pressure)*Kow above
reference pressure
PCog = 0
If the pressure is below the partition reference pressure.
PCwo = Partition coefficient + (Pressure-Partition reference pressure)*Kow below
reference pressure
PCog = Gas/oil partition coefficient + (Pressure-Partition reference pressure)*Kgo below
reference pressure
H2S partitioning - this model has the same data input and form as the CO2 partitioning
model above.
Initial partition coefficients - if the New Partition Coefficient flag is checked in the water
chemistry section, then the water/oil partition coefficients (Part Coefficient for both CO2
and H2S) are calculated from the initial conditions (i.e. initial value of CO2oil/H2Soil and
initial equilibrated CO2/H2S in water).

PETROLEUM EXPERTS LTD

CHAPTER 3 USING THE INPUT WIZARD

3.10.2

63 -84

H2S Souring

The H2S souring model is only available if the water chemistry and souring options were
selected in the control section.
The souring model requires the components SO4-2, HS- and nutrient to be present. In
addition, the H2S partitioning components H2S_oil and H2S_gas should usually be
present within a souring model. HS- may also be an adsorbed component, modelling its
retardation. The simplified equilibrium model may be used for souring calculations that do
not contain precipitating minerals.
The basis of the souring model is that SO4-2 reduces to HS- under the catalytic presence
of sulphate reducing bacteria (SRB).
The growth of SRB is controlled by the presence of the nutrient component within a
suitable temperature range. As the nutrient is used up, the rate of SRB growth reduces,
but the SRB remain present and available to reduce SO4-2 ions if they are present i.e. it
is assumed that the SRB may metabolise at a constant concentration with no nutrient
present due to remineralisation and resting cell metabolism. The SRB are assumed to
not flow with the injected water, but remain and grow within grid blocks. The net result is
that as the SRB grow, they are available to reduce SO4-2 ions within a cooled region
near an injector, and the reduced HS- ions may then form aqueous and hydrocarbon
phase (assuming partitioning) concentrations of dissolved H2S, which may then be
transported in the aqueous and hydrocarbon phases.
Temperature dependence - minimum and maximum temperature at which the SRB may
grow.
Concentration dependence - minimum and maximum SRB concentration. The
minimum concentration is assumed to be present initially everywhere within the reservoir
as a starting condition, in the form of SRB spores. The maximum concentration is the
maximum concentration that the SRB can achieve, even it the conditions (temperature
and nutrient availability) are otherwise favourable for SRB growth.
SRB growth rate - half life for SRB concentration to double under favourable conditions
(temperature within set bounds and nutrient available).
Maximum SO4(-2) reduction rate - the sulphate reduction rate (molality/day) at
maximum SRB concentration. The sulphate reduction rate at reduced SRB
concentrations is a linear interpolation between zero and the maximum value entered
here as a function of SRB concentration between its minimum and maximum values. The
sulphate reduction rate is also limited by the sulphate ions present.
Nutrient/SRB mass ratio - the mass ratio of nutrient to SRB, required to reduce the
nutrient concentration as the SRB concentration increases (SRB growth).

OCTOBER 2003

REVEAL MANUAL

64 - 84 CHAPTER 3 USING THE INPUT WIZARD

3.10.3

Scale Inhibition

The scale inhibition model is only available if the water chemistry option was selected in
the control section, and there are precipitating minerals and the component Scale
Inhibitor is present (see component selection).
The scale inhibition model increases the solubility of minerals as a function of inhibitor
concentration within the aqueous phase. The scale inhibitor may be adsorbed. Scale
inhibition is based on the assumption that the scale inhibitor acts to provide nucleation
centres for the precipitating mineral that prevents large particulate precipitation.
The solubility of the minerals is controlled by a saturation index (SI). The default value is
zero, and as the SI increases the soluble concentration of the mineral rises
logarithmically, so that an SI of 1.0 allows the soluble concentration of the mineral to be
ten times larger than the default value (defined by the PHREEQC database), before
precipitation occurs.
Scale inhibition can be applied to none, any or all precipitating minerals, but they all use
the same scale inhibitor concentration to calculate the reduced solubility.
Maximum effective inhibitor concentration - the maximum effective scale inhibitor
concentration. Scale inhibitor concentrations above this maximum will not further
increase solubility.
Maximum saturation index - the maximum mineral saturation index (SI). The SI used
will rise linearly from zero to the maximum SI entered as the scale inhibitor concentration
varies between zero and the maximum effective value.

PETROLEUM EXPERTS LTD

CHAPTER 3 USING THE INPUT WIZARD

65 -84

3.11 Wells Section

3.11.1

Location and Properties

This section initialises the location and properties of wells. Use the Plot button to view the
wells (green) and completions (red) over a wire-frame representation of the grid.
Data for all wells - these switches are applied to all wells.
The block or average well models define how the concentration of components within
phases from producing wells are calculated. The default option (block) sets the
component concentrations equal to the concentrations present within the producing block,
and is generally recommended. The average option sets the component concentrations
equal to a horizontal spatial average, which makes the assumption that component
concentrations have a cubic form near the well.

c=

28c i , j , k (c i 1, j , k + c i +1, j , k + c i , j 1, k + c i , j +1, k )

24

Etrapolate lift curves is generally not recommended, lift curevs should be generated over
sufficient data ranges. The extrapolation may produce very erroneous results if the
extrapolation is large.
The shear model applies a shear thinning correction for water injection wells containing
polymer and gel components. See the mobility section for more details.
Well list - wells may be added using the Add button. Once a well has been added, its
properties may be modified by selecting the well in the list. A well may be deleted by
selecting it in the well list and using the Delete button.
Well label - a name for the well that can be edited to identify the well.
Enter well location - the method of defining the well completions can take three forms.
Either a completion table containing the I,J,K coordinates of the completed blocks, or a
vertical column of grid blocks identified by I,J block coordinates, or a general multi-lateral
well may be defined.
If a well is to be located within a refined region, then a fourth option (separate refinement)
should be selected, and the refined region name entered. The extent of the refined
region and its name should be entered in the reservoir section. The well location and
completions within the refined region are defined in a separate refinement script. See the
grid refinement section for more information.
For radial and curvilinear geometries the first two wells are defined by the geometry and
their location cannot be altered. Additional wells for these geometries are not
recommended.
Layer data & completion table - for the completion table or vertical column method of
defining the well location, the data for each block associated with the well are entered.
For the completion table, the I,J,K coordinates of the well must be entered for each block.
A block may be set to be perforated by using the left mouse button within the table
perforation column. For the completion table, the PI (Productivity Index) may be input

OCTOBER 2003

REVEAL MANUAL

66 - 84 CHAPTER 3 USING THE INPUT WIZARD

directly, or calculated using the Peaceman [13] model. For the vertical column method
the perforated layers should be set and the data required by the Peaceman model
entered.
The well radius and skin are used to calculate a PI for anisotropic non square grid blocks
using the Peaceman model. The fractional completion angle is a multiplier to the
Peaceman calculated PI, modelling partially completed or deviated wells. It is up to the
user to supply this data.
For the well-bore heating model, the earth and electrode locations are identified in the
final column of the completion or layer data table.
Multilateral well [6] - if the well is a multilateral or defined by a deviation survey, then it
must be given an identifier name. Select the Edit Multilateral button to enter a deviation
survey. A reference location for the top of the well must be entered in either grid or map
axes. A hierarchy of connected tubing sections (possibly consisting of several branches)
can be added or deleted from the tubing list. For each tubing section a deviation survey
is entered using either grid coordinates or deviation survey coordinates. Any interval on
the tubing section may be completed and given a radius and skin. The well may be
viewed using the View button at any time. Note that deviation surveys can be imported
from previously exported files.
REVEAL initially calculates the intersection of the completions with the matrix grid blocks.
It then calculates the standard Peaceman connection [7,13] for a vertical well connected
to each intersected grid block and correction factors accounting for the actual production
to the intersected matrix blocks.
The correction factor for each block intersected by a multilateral well is calculated using
point-source solutions to the flow around the entire well, therefore including the interaction
of well completions in different grid blocks. The point-source inflow to the block being
considered is then compared to the block inflow calculated with a similar point-source
solution obtained from a reference vertical well passing through the grid block centre.
This approach effectively calculates a correction factor multiplier (skin) to each intersected
block taking account of the non-radial flow, which is assumed by the Peaceman model.
For example, the inflow at the ends of a completion will be greater than the inflow at the
centre of the completion, and the interaction of multi-lateral completions will influence the
production of neighbouring completions.
The Peaceman well connection (PI) accounts for the logarithmic pressure distribution
around a vertical well, and for a given rate (Q) and bottom hole pressure (Pwf) sets the
appropriate mean block pressure (P) that will give the correct inflow (Q=PI(P-Pwf)) and
total rate to surrounding grid blocks using the linear Darcy formula. The multilateral
correction, essentially modifies the Peaceman connection factor to account for deviated,
partially penetrating or multilateral completions.
The reference Peaceman well is chosen to be vertical (even for horizontal wells), since
the prevailing reservoir flow will most often be vertical pseudo-radial, following initial
periods of horizontal radial and linear flow. This assumption is pessimistic with respect to
PI calculation.
Lift curve and well type - a PROSPER lift curve (*.tpd) may be imported. Generally it is
recommended that the lift curve is imported and associated with a well in the schedule
section rather than in the well section. Care should be taken to ensure that the imported
lift curve is of the correct type (e.g. water injector or producer) for the well.

PETROLEUM EXPERTS LTD

CHAPTER 3 USING THE INPUT WIZARD

67 -84

Pwf reference - the default (top of well tubing) location of the Pwf reference pressure
depends on whether the well is defined by completion table, vertical column or a
multilateral. For a completion table or vertical column, the Pwf is located at the top centre
of the first block in the completion list, whether this block is completed or not. For a
multilateral well the Pwf is located at the first (reference) node.
Alternatively, a reference depth may be entered for each well.
Non Darcy factor - a non-Darcy factor may be applied to the gas phase of any well. This
factor is scaled for each completed block according to the Darcy connection factor
(assumes equal drawdown for completed blocks) and applied as a rate dependent skin.
The block geometrical connection factor (CF) prior to the non-Darcy effect is defined as

CFb =

2kh
r
ln e + S
rw

and is modified to include the non-Darcy skin as

CFbnonDarcy =

2kh

r
ln e
rw

+ S + Db Qb
Db =

, where

Dwell CFb
b

CFb

Note that if the well completion blocks are entered as a table with only the total
connection factor, the nonDarcy skin will be applied with the denominator of the
connection factor assumed to be unity.
Wellbore friction - if this option is selected then an internal diameter and absolute
roughness should be entered for every well segment, whether it is completed or not. This
option is not available for curvilinear or radial grids, or for wells defined by a completion
table. The wellbore friction model uses a simple no-slip frictional pressure drop model,
taking into account a mean density and viscosity at each node within the wellbore.
Cross Flow - a cross flow model is available for any well. It is recommended that
crossflow is used for wells for which a trajectory and wellbore friction has been set,
otherwise the crossflow fluid will be the combined produced or injected fluid not taking
into account the local fluid within a wellbore. This model is not recommended to be used
in conjunction with fractures, since cross flow within the fracture is not modelled.

OCTOBER 2003

REVEAL MANUAL

68 - 84 CHAPTER 3 USING THE INPUT WIZARD

3.11.2

Fracture Properties

This screen is only active if fracture models is selected in the control section. It initialises
the fracture type (producing or thermal injection), location and initial dimensions.
All fractures have an origin at the centre of a grid block (with coordinates IJK), associated
with a completed well. The fracture then has a two-dimensional shape running down the
centre of adjacent blocks with constant (X, Y or Z) coordinate.
Fracture list - use the Add button to add fractures to the model. Select a fracture in the
fracture list to alter its properties or use the Delete button to delete the fracture.
Fracture position - give the fracture a label name or use the default and select a well to
be associated with the fracture. Every fracture must be associated with a well. For both
injection and producing fractures more than one fracture (multi-layer fractures) may be
associated with a well.
The location of the fracture origin is then defined. For wells with a completion list or
vertical wells, the completed block number for the fracture is entered. For a multi-lateral
well the coordinates of the fracture centre are input; use the Calculator button to select a
completion and interpolate a position along the completion (using either an X,Y or Z
coordinate).
Note that for multi-lateral wells, the fracture origin will be moved from the input position to
the centre of the block containing the input position when the calculation is run. This is
because the fracture origin must be defined at a block centre, since the connection of the
well to the block is defined by reference to the block centre.
Fracture model - three fracture models are available, a 1D analytical thermal fracture
model (Geertsma deKlerk), a finite-element thermal fracture model (3D model) and
producing fracture model. The first two models are thermal injection fractures, where the
shape of the fracture evolves as the fracture propagates, while the producing fracture
model has a static fracture shape, assumed to be sustained by propant or acid treatment.
For all three fracture models the initial dimensions of the fracture are required; a half
height above and below the fracture origin and a half length. For the injection (thermal)
fractures, this data is used to describe the initial fracture dimensions arising from
formation damage or naturally occurring fractures. The propagation of a fracture starts
when the well Pwf is large enough to generate a stress intensity at the fracture tip greater
than the critical stress intensity. Prior to propagation initiation, no inflow from the fracture
is assumed. A smaller initial fracture size will require a larger initiating Pwf and hence
longer Darcy (well completion only) inflow time before the fracture is initiated.
For the 3D fracture and producing fracture, a two dimensional finite-element grid is used
and explicitly models the fluid flow and pressure drops within the fracture. For these
fracture models, the permeability (k) within the fracture must be defined. Three models
are available, including constant permeability, Fcd (dimensionless fracture constant) and
parallel plate. The latter two models depend on the fracture width (w), fracture half length
(L1/2) and X-direction reservoir permeability (kx) for the Fcd model, and are defined
below.

k=

Fcd k x L1 / 2
w

PETROLEUM EXPERTS LTD

CHAPTER 3 USING THE INPUT WIZARD

69 -84

w2
k=
12
An Fcd value of 500 to 1000 represents the infinite conductivity approximation.
Geertsma deKlerk - this model is basically a volume balance model based on a leakoff
correlation. The leakoff and spurt coefficients are the only additional data required for this
model. This model is a single wing model and the fracture origin should be positioned in
the first row of the grid (I=1 or J=1) so that the wing may propagate in the X+ or Y+
direction. The propagation direction depends on the minimum principal stress direction.
Minimum principal stress direction
X
Y
Z

Fracture propagation direction (length)

Y+
X+
X+

Fracture height direction

Z (vertical)
Z (vertical)
Y (horizontal)

3D fracture - this model calculates the fracture shape including effects of stress changes
due to injected fluids, elastic fracture opening, propagation depending on rock strength,
flow within the fracture, leakoff rate balancing the well injection rate and Pwf.
The propagation direction is dependent on the minimum principal stress direction.
Minimum principal stress direction
X
Y

Fracture propagation plane

YZ plane
XZ p

70 - 84 CHAPTER 3 USING THE INPUT WIZARD

The shape of the fracture plane may be rectangular or elliptical. The width at the fracture
origin is entered, and the width profile defined to be either constant (=central value) or
decrease with an elliptical cross-section towards the fracture boundaries.
A fracture skin may be included for this model, reducing productivity with a positive value.
Thermal fractures - the stress near the fracture is reduced if injection water at a
temperature below the reservoir temperature is used. Similarly the stress rises if the
pressure rises above the initial reservoir pressure (pressure at which the in-situ stress is
defined). The 1D Geertsma de Klerk model only uses the stress at the fracture tip, and
therefore this model is not greatly influenced by the stress reduction parameters.
All fracture models may use constant pressure, constant rate, or injection curves for the
well from which the fracture propagates. The resulting fracture will have the following
properties.
Pressure and rate consistent with the injection well constraints.
Leakoff rate from the fracture to the rock matrix consistent with the fracture size and
leakoff model.
Stress intensity at the fracture tip equal to the rock critical stress intensity.
Various iterations are performed to ensure that the fracture shape (lateral and width),
pressure and leakoff rate are consistent with the well performance, current stress and
reservoir conditions.

PETROLEUM EXPERTS LTD

CHAPTER 3 USING THE INPUT WIZARD

3.11.3

71 -84

Geertsma deKlerk 1D Thermal Fracture

This is a simple model that only allows a single-wing fracture to be modelled. It requires a
leakoff and spurt coefficient to be supplied, defining the leakoff rate as a function of
fracture area. It is most applicable in cases where the rock is fairly elastic and relatively
large fracture volumes may be present.
Fracture opening and propagation criteria - this model assumes a rectangular vertical
fracture with half-length (Lf) and a constant pressure (Pf) within the fracture. If this
pressure is greater than the rock stress (), then the fracture can open. The criterion for
fracture propagation is that Pf generates a stress intensity at the tip greater than the
critical stress intensity, KIc. Propagation stops when the stress intensity at the tip is equal
to the critical stress intensity.

Open condition

Pf >

Propagation condition
K Ic
Pf > +
L f

Width and length equations - the width of the fracture at its widest point (ww) (injection
site) and the fracture half-length are related to the total fracture injection rate (Qf), leakoff
rate (QL), leakoff (C) and spurt (SP) coefficients, time (t), half-height (h), and history of the
evolving fracture area (A(t)). Basically, this model is a material balance model, where the
changing volume of the fracture is the difference between the injected and leakoff rates.
The volume of the fracture is a function of its length and central width, which in turn are
related to the elasticity and tensile strength of the rock for a given internal pressure.

Elastic strain relation describing fracture shape Material balance equation

V
A
2(1 )
ww =
L f ( Pf )
= Q f QL S p
G
t
t
Fracture volume and area

V=

hL f ww

and

A = 4L f h

Leakoff equation
t
C dA
d
QL =
d

G is the shear modulus and defined by Young's modulus (E) and Poisson's ratio ().

G=

E
2(1 + )

These equations are solved for the fracture pressure and leakoff rate for an input total
fracture injection rate. The result depends on whether the fracture is propagating or not.
The fracture pressure is then coupled to the injection lift curve, finally after iterations
returning a consistent total injection rate Q=Qf+Qw.

OCTOBER 2003

REVEAL MANUAL

72 - 84 CHAPTER 3 USING THE INPUT WIZARD

3.11.4

3D Thermal Fracture

Both single-wing and complete (double-wing) fractures are possible using this model. A
finite-element grid is introduced, with triangular internal elements and quadrilateral
boundary elements. The fracture is therefore approximated by a 2D plane with fracture
widths defined over this plane.

A skin (S) may be defined for the fracture. The mobility connection factor (M) for a grid
block intersecting the fracture with an area of intersection (A) and pressure drop (P)
between the fracture and grid block is calculated assuming linear flow in the a normal
direction (Y direction in equations below) away from both sides of the fracture.

M
Q = A
P
1+ S

M =8

Ky

krp

Internal pressure width equation the pressure and width within the fracture are
related by the following equation.

( P )( x, z ) =

G
1
( ) wdA

4 (1 )
R

R = ( x x ) 2 + ( z z ) 2

G is the shear modulus and defined by Young's modulus (E) and Poisson's ratio ().

G=

E
2(1 + )

Singularities are dealt with analytical expressions on the internal triangular elements. On
the outer quadrilateral elements, it is assumed that w
the tip and w=0 on the outermost boundary.

a , where a is the distance from

The critical width (wc) at which the fracture will just propagate is defined at a small fixed
distance (a) from the fracture tip such that the stress intensity at the tip is equal to the
critical stress intensity for the rock (KIc).

wc =

4 K Ic (1 ) a
G
2

PETROLEUM EXPERTS LTD

CHAPTER 3 USING THE INPUT WIZARD

73 -84

Incompressible fluid fracture flow equation - the flow and pressure within the fracture
are related to the fracture leakoff rate.

w2 2
h
wdA V0 Q = 0
(P
)dV + M ( P P mat )dA +
f
12
144
t f

Qf is the total fracture injection rate, added at the central node of the finite-element grid.
Volumetric storage rate within the fracture is included, reflecting the volume increase

wdA V

of the fracture since the previous fracture update

t f

Finite-element equation - the fracture width and leakoff equations are written in finiteelement form.

Finite-element width equation

Finite-element leakoff equation

Lij w j V0i
h j
Fij ( Pj
) + Lij M j ( Pj P mat j ) +
Qi = 0
144
t f

Wij w j = Pi i

Solution method - the finite-element equations are combined and solved iteratively for
the fracture widths using the Newton Raphson method.
1

Fi ( w j ) = 0 wi = wi J ij F j
J ij =
where
algorithm.

Fi
w j

is the Jacobian and is a scale factor found using a line-search

The finite-element fracture width solver is supplied with a total rate (Qf), and the shape of
the fracture is iterated on until a consistent shape is found with the stress intensity at the
fracture tip equal to the critical stress intensity (KIc) for the fracturing rock. The pressure
at the centre of the fracture (Pf) is returned and a top level of iteration is performed to
ensure that Qf and Pf are consistent with the total well rate (including flow directly from the
well into the rock matrix) and the bottom hole flowing pressure (Pwf).
Note that for a single wing fracture only the flow from the well completions and one wing
of the fracture are injected into the grid. However, the total well rate and Pwf must be
defined for both wings of the fracture. The result is that the reported injection rate (well
and one fracture wing) may be as little as 50% of the total well injection rate entered in
the schedule.
Stress calculation - the stress on the fracture surface is calculated from the in-situ stress
field (insitu) and poro/thermo-elastic stress reductions (P & T). The in-situ stress is
defined by the user as a function of depth or by horizontal layer. The poro and thermo
stress reductions take a similar form and the equations used were obtained from Koning
[18].

= insitu P T = insitu yy

The Goodier () displacement potential is first calculated over the entire FD (finite
difference) grid by solving Laplaces equation.

2 =

(1 + ) (A
E

OCTOBER 2003

P + AT T )

REVEAL MANUAL

74 - 84 CHAPTER 3 USING THE INPUT WIZARD

where

P = Pinitial P and T = Treference T

AP and AT are poro-elastic and thermo-elastic constants.

The displacement potential could be used to calculate any component of the stress
reducing tensor. The Y direction stress reduction (perpendicular to fracture surface) is
then calculated. Similar forms for the stress reduction are performed if the principal stress
direction is in the X or Z direction.

yy

E 2
+ A p P + AT T
=
(1 + ) y 2

This method is effective for thermal fracturing only once the cooled region is at least the
same size as the grid blocks being used. An analytical model for fracture initiation and
propagation within the fracturing block is used. The simplified model was derived from
references [9, 12, 14, 22]. An estimate of the cooled and flooded region radii are
calculated from the total downhole volume of injected water (W).

Cooled region radius

Cp w w
3V
Vcooled = W
rcooled = 3 cooled
Cp ,
4

Flooded region radius

W
3V flood
=
r flood = 3
(1 S r )
4
,

V flood

The pressure and temperature distribution within the cooled and flooded regions are
estimated as a function of radial distance (r) from the fracture centre. These simple
approximations assume that the temperature and pressure are at a mean between the
injection and reference/initial conditions at their respective cooled/flooded radii.

Cooled region temperature

Tcooled = Tinj (Tinj Tref

r2

Flooded region pressure

Pflood = Pinj (Pinj Pinit )

2 ln(

2r 2 cooled

r
)
rw
r flood

ln(

rw

The thermo-elastic and poro-elastic stress reductions on the fracture surface within the
fracturing block are then calculated, assuming a circular shape function of 0.5.

yy = 0.5 AT (Treference Tcooled ) + AP (Pinitial Pflood )

As the fracture extends beyond its initial block, the stress reductions calculated using the
Goodier displacement potentials are interpolated onto the 2D fracture surface. For this
interpolation, the stress reduction for the fracturing block is calculated using the simple
model above with the cooled temperature set equal to the injection temperature
(Tcooled=Tinj), and the flooded pressure set equal to the block average pressure
(Pflood=Pblock). Note that this stress reduction is only applicable once the fracture has
grown beyond its initial block.
Implementation - the two-dimensional finite-element (FE) fracture grid described above
is coupled to the three-dimensional finite-difference (FD) main grid. The FE grid is
located down the centre-line (constant Y) of the FD grid. The origin of the fracture is
located at the centre of a FD grid block.
The FE grid has internal triangular and quadrilateral elements on its boundary. The size
of the elements may vary as the fracture propagates, although their topology remains
constant. Variables are defined at the nodes of the grid. These variables include the
fracture width, internal fracture temperature and pressure, fracture surface stress, and
derived quantities such as injection water density and viscosity.
PETROLEUM EXPERTS LTD

CHAPTER 3 USING THE INPUT WIZARD

75 -84

The FD grid has hexahedral elements, which have fixed size. Variables are defined at
the centre of the grid blocks and represent volume average properties, and therefore are
dependent on the grid block dimensions (especially near an injection source, where the
spatial variation of physical properties such as pressure and temperature are far from
linear). Relevant variables include pressure, temperature, permeability and phase
mobilities.
Some variables are required on both the FD and FE grids and linear interpolation is used
to transfer properties from one grid to the other. The FE grid basis functions are used to
interpolate to the FD grid, while interpolation from the FD to the FE grid is performed
using bilinear interpolation of four values at FD grid block centres.
The temperature within the FE fracture grid is set using the following formula (Meyer).

T = Tinj + (1 )Tmatrix

(
1 x) 1 + 2
=

2
1 + x + 2
2 1

)(

where x is the fractional distance from the centre of the fracture to its tip, and is a user
specified heat flux coefficient. Note that the temperature at the FE grid nodes is not a
function of the grid shape or area, since x remains constant as the grid grows.
The displacement mobility ((kr/)) of fluids within the FD grid is calculated using the FD
grid temperature, pressure and phase saturations. These are merely the mobilities used
for the flow within the FD grid. The previously injected water displacement mobility
(calculated as a water blocking fracture skin) uses the internal fracture temperature (FE
grid) and FD grid pressure to calculate the water viscosity with the water phase relative
permeability end-point. The thickness of the water blocking layer is calculated from the
total volume of water injected.
Finally, once the fracture shape, pressure and total rate have been calculated, the
resulting fracture is connected to the FD grid as a set of connection factors and a
constant rate injection. The connection factors are calculated by considering the total
flow from the FE fracture grid to each FD grid block intersected by the fracture. Once the
volumetric rate into each FD grid block is calculated, the connection factor for that block is
simply the rate divided by (Pf-Pi), where Pf is the internal fracture pressure and Pi is the
FD grid block pressure. Several FD grid time-steps may be taken before the fracture is
updated again. The connection factors (and rate) remain constant during these timesteps, resulting in the Pwf increasing until the fracture shape is updated again.

OCTOBER 2003

REVEAL MANUAL

76 - 84 CHAPTER 3 USING THE INPUT WIZARD

3.11.5

Rock Stress and Elasticity

This screen is only available if thermal fractures are present and defines the mechanical
properties of the rock to be fractured and the initial in-situ stress.
Stress - an initial reference stress and reference temperature are input. All thermal
fractures have the same minimum principal stress direction and these fractures will
propagate with a plane normal to this direction. The inactive block stress, is the stress
that will be applied if the fracture propagates into inactive blocks or outside of the gridded
region - this value should be chosen to be large enough to confine the fracture within the
grid.
Stress properties per layer - the elastic coefficients (young's modulus and Poisson's
ratio), elastic stress reduction parameters (poro and thermo-elastic stress reduction) and
the critical rock stress intensity (rock strength) are entered layer by layer.
The poro-elastic stress coefficient is defined as:

AP = P

E
(1 ) and

1 cg

c
P = b

(1 2 )

E is Young's modulus
is Poisson's ratio
P is the linear poro-elastic expansion coefficient
cg

is the matrix grain compressibility

cb is the bulk matrix compressibility

The thermo-elastic stress coefficient is defined as:

AT = T

E
(1 )

T is the linear thermal expansion coefficient

Select the variable to entered, then select Add to enter data. Different data may be
entered for different vertical layers (Z increasing with depth), or the All button may be
used to enter the same data for all layers. Select a layer from the layer summary table to
modify the data or Delete the layer. The variable will turn from red to green when all of
the data is entered, or use the Validate button to check all required data has been input.
Stress layering - a term may be added to the reference stress as a function of layer or
depth. Enter a top and bottom stress increment at the top and bottom limit of each layer
range entered. This may be used to define stress barriers or a vertical stress gradient.
The increments need not be continuous (i.e. the bottom of one layer need not be the
same at the top of the layer immediately below). Note that if layer stress variations are
used, then data for each layer should be included if there are inactive blocks within the
grid.

PETROLEUM EXPERTS LTD

CHAPTER 3 USING THE INPUT WIZARD

77 -84

A fracture will tend to propagate from its origin into low stress regions. To minimise large
distortions to the finite-element grid the in-situ stress field should be reasonably smooth
(and not a high stress barrier) at the fracture origin.

OCTOBER 2003

REVEAL MANUAL

78 - 84 CHAPTER 3 USING THE INPUT WIZARD

3.12 Well-bore Heating Section

This section is only active if the well heating model is selected in the control section. The
location of the electrodes (electrode and earth) are defined in the wells section and the
voltage to be applied is set in the schedule section.
The heating model applies a microwave frequency voltage across two points on a well
and calculates the heating effect within the reservoir, which in turn will reduce the
viscosity of heavy oils. The voltage generated (V) within the reservoir is calculated on the
finite difference grid used for the flow equations by solving Laplace's equation with
boundary conditions V=V0 on the electrode, V=0 on the earth electrode and no voltage
flux on the boundary of the grid.

( V ) = 0
The heat supplied to the grid (W) is then calculated from the voltage field.

W = (V )

Resistivity - Rw is the water resistivity and is calculated using Arp's law, or input as a
table of values as a function of temperature.

R w = R ref

(T

ref

+ C)

(T + C )

where C is Arp's constant 6.77F

Lithology, saturation and tortuosity - the conductivity of the reservoir () is calculated
from Archie's law.

mSwn
aR w

where is the porosity, a is the lithology coefficient, m is the tortuosity (cementation)

coefficient and n is the saturation exponent. The lithology coefficient, tortuosity coefficient
and saturation exponent may be entered for multiple regions of the grid, defined by
ranges of blocks. See the reservoir section for information on how to add range data.
Use the Validate button to check coverage is complete.
Archie's law is applied up to the boiling point of water defined by T=0.08329742*P+458.76
(F); once this temperature is reached, the conductance is set to zero and the lithology
coefficient set to 1E6 to make the loss of conductance virtually irreversible.

PETROLEUM EXPERTS LTD

CHAPTER 3 USING THE INPUT WIZARD

79 -84

3.13 Initialisation Section

3.13.1

PVT Initialisation

This section initialises the PVT for each region defined within the reservoir section. Each
PVT region must be non communicating, so that different PVT fluids do not mix within the
reservoir.
Region List - selects the region to be initialised. All regions must be initialised.
PVT Files - select the PVT file to be associated with the current region. The PVT files
may be added to or viewed. Generally, the PVT files are entered within the physical
section.
RS - Gradient - This section is optional and can only be entered if the miscibility option
has been set for the PVT. The miscibility option provides the possible for multiple
saturated oil tables and a variable initial RS. The initial RS versus depth should be
entered within this table. The table is not extrapolated so sufficient depth range should be
entered.

3.13.2

Equilibration Initialisation

This section initialises the reservoir for each equilibration region specificed within the
reservoir section. Each equilibration region must have only one PVT, i.e. equilibration
regions must be a subset of the PVT regions.
The equilibration includes pressure and contact (oil/gas water/oil or water/gas) depths.
The pressure and contact depths are use during the initial reservoir equilibration to assign
initial pressures and saturations to all grid blocks.
Region List - selects the region to be initialised. All regions must be initialised.
Reference depths - one reference pressure is required at a reference depth for each
equilibration region.
If a gas/oil contact is to be defined then the reference pressure will be the bubble point
pressure and does not require to be entered.
If an oil/water or gas/water contact depth is defined, then the reference pressure is only
required if there is no gas/oil contact.
If no contacts are present, then a reference pressure and depth should be entered.
Although the reference depth can take any value, it should generally lie within the
reservoir. If a GOC is then calculated during equilibration then a gas cap will be present.
Thermal gradient - the thermal gradient must be entered for each equilibration region.
The reference temperature is the temperature at the temperature reference depth, and
the gradient is the initial thermal gradient (increasing with depth).
Miscellaneous Thermal Initialisation - the overburden and underburden temperature
should also be input, they are not required to be input for each equilibration region. For

OCTOBER 2003

REVEAL MANUAL

80 - 84 CHAPTER 3 USING THE INPUT WIZARD

an isothermal model the temperatures input should all be the same and the thermal
gradient should be zero.
Initial wettability - the initial wettability is only required for models with relative
permeability desaturation controlled by adsorbed wetting agent. This model is enabled in
the control section. A value of -1 corresponds to initial low tension relative permeability,
while a value of +1 corresponds to initial high tension relative permeability. Intermediate
values are possible and correspond to intermediate (interpolated) desaturation.

3.13.3

Trace Component Concentrations

This screen is only available if more than three components (water, oil and gas) are
present. The initial concentrations of the trace components in their initial phase (aqueous
phase for most components) in mass fraction (component mass per unit mass of host
phase).
The default is zero concentration.
Select a component and enter its initial concentration. The concentration may be defined
over a number of range regions or for each grid block. See the reservoir section to get
information on how to use the range and array selection methods. Use the right mouse
button within the concentration input cell to toggle units between kg/kg and ppm.
Electrical neutrality - only used for water chemistry models, where electrical neutrality is
required (all ions except e-). Neutrality will be maintained internally by adding or
removing the ion pair defined in the water chemistry section, but the electrical neutrality of
the initial concentrations of ions entered may be tested and modified (usually by adding or
removing Na+ or Cl- ions) using the Calculate button. Alter the tolerance parameter to
view the grids with charge imbalances greater than a defined limit. Generally the electric
neutrality should be better than 0.05 charge/mole for all grid blocks.

3.13.4

Residual Saturations

This section is optional and residual (connate) saturations may be entered. If data is not
entered, then the connate saturations will be the minimum saturation entered in the
relative permeability section. This data is used during equilibration to set connate
saturations and as end points for the capillary pressure tables.

PETROLEUM EXPERTS LTD

CHAPTER 3 USING THE INPUT WIZARD

81 -84

3.14 Schedule Section

3.14.1

Well Schedule

This section defines the schedule of well control (production and injection) and the
timestep control.
It is characterised by a sequence of schedules, that run sequentially until a termination
(exit condition) is met, when the next schedule is started.
Schedule list - this lists the schedules that will be applied in chronological order.
Schedules my be Added, Deleted or Inserted. The Clear button deletes all of the
schedules. All of the remaining data entered on this screen is applied separately for each
schedule entered.
Restart file - a restart file may be generated at regular intervals or not generated at all. If
a restart file is generated it may overwrite the previous restart file or be added to a list of
restart files.
Two types of restart files are possible. The default restart file stores only the data
required to perform a restart and should be used under most circumstances. A debug
restart file stores all current data structures (including workspace) and may be very large
and slow to write; the debug restart files should only be used for the fracture IPR
calculations or for diagnostic reasons if requested by Petroleum Experts.
It is recommended that for most models the restart files are suppressed. A restart file
may generated interactively at any time during a calculation by pausing the simulation
and using the menu option Run Simulation|Generate restart step. If the simulation
pauses before the current timestep is completed and the restart option is unavailable, use
the Run Simulation|Do One step command to complete the timestep. The list of restart
files may viewed or deleted using the menu option Project|Edit view project.
Timestep size - this section sets the initial timestep control and data entered is
dependant on the solver options (IMPEC or implicit) used.
The first schedule must have an initial timestep size, this should generally be set to a
small value, that will increase as the simulation progresses, this prevents a potentially
large number of retrys at the start of the simulation where transient behaviour may be
severe. Subsequent schedules may also have initial timesteps set, but this is not
necessary.
The volumetric error should be set. This is the principal timestep control parameter for
the IMPEC solver. It is recommended that a value 1E-4 with 3 iterations or 1E-5 with 5
iterations is used. The material balance volumetric error is defined as the fractional
volumetric error at the end of the timestep (i.e. difference between porosity and total
volume fraction of fluids present within a gridblock calculated using the pressure and
temperature at the end of the timestep). A smaller tolerance for the volumetric error may
result in smaller timesteps and require more solver iterations. Material balance iterations
are not used with the implicit solver, which only generates material balance errors as a
result of explicit thermal effects and incomplete convergence. A a value of 1E-4 is
recommended for the implicit solver volumetric error.
The saturation overflow limit is only required for the IMPEC solver, and limits saturation
OCTOBER 2003

REVEAL MANUAL

82 - 84 CHAPTER 3 USING THE INPUT WIZARD

changes near the residual saturations of fluids, preventing undershoot in saturation

(saturations below the residual saturation). The default value is 0.01 and may be reduced
to 0.001 in cases where very mobile phase saturations are oscillating near the residual
saturation (e.g. production below the bubble point). A reduction of this limit will result in
smaller timesteps being used and slow the simulation to improve stability.
The limit dC option only applies to the IMPEC solver and sets a limit on the fractional
change in trace component concentrations within a timestep. This option should
generally not be used. It may be applied to reduce timesteps near large concentration
gradients to limit numerical dispersion in some models.
The maxdt and mindt are minimum and maximum timestep sizes and should not
generally be set. Sometimes the maximum timestep size may be set to increase temporal
resolution.
The limdT and limdP options are limits on temperature and pressure change during a
timestep. Again these options should not generally be set, except in cases where rapid
changes in temperature or pressure are generating instability in the solutions obtained
(e.g. large oscillations in thermal fracture behaviour).
Well schedule - data is entered for each well by selecting the well in the list and selecting
the well to be shut-in, producing or injecting.
For producing or injecting wells, three control mechanisms are available, constant Pwf,
constant rate (oil+water, oil, water or gas for producer) or constant Pws (well head
pressure). For injection wells, a voidage replacement option is available, where a fraction
of the total production voidage for a group of producing wells is specified.
For any control mechanism (required for Pws control) a lift or injection curve may be
associated with the well. Use the Import button to import the lift curve. Ensure that the
curve type (e.g. Oil - naturally flowing for an oil producer or Water injector for a thermal
fracture well) matches the well. A label name may also be given to the lift curve. The lift
curve should cover the full range of possible condition (GOR, water cut, rate etc.) for the
model being investigated.
For an injection well, the bottom hole temperature and composition must be entered. If
an injection curve generated by PROSPER has bottom hole temperature data, then this
may be used by selecting the Use TPD Temp option. Use the Injection button to set the
injector as water or gas and set the concentration of any trace components in the injection
fluid. Electrical neutrality for ions present in the injection water for water chemistry
calculations can be checked. See the initialisation and water chemistry sections for more
information.
In addition, the relative permeability model for the injection must also be set. The options
are:
1. Standard (the default model). In this case, the fluid mobility for a connected grid block
is given by the sum over phases of the rel perm for that phase divided by its viscosity.
2. End point. The relative permeability used will be the end point of the rel perm curve for
the injected phase.
3. Fixed. Enter a fixed value for the relative permeability that will be used in injection.
Note that options 2 and 3 are not available if wellbore friction or well cross-flow are being
modeled due to the possibility of multi-component injection.

PETROLEUM EXPERTS LTD

CHAPTER 3 USING THE INPUT WIZARD

83 -84

Use the Well control button to enter constraints on the well.

For wells that have completions defined by completion table or vertical well (not
multilateral) use the Perforations button to enable/disable or modify the perforation
properties.
If a wellbore heating model is present, then the heating can be turned on and an
electrode voltage applied. If the heating is turned off, then no heating will take place.
Exit conditions - the schedule will run until an exit condition is met. If several exit
conditions are entered, then the first to be met will end the schedule and the next
schedule (if there is one) will be started. Use the left mouse button to select options in
the exit condition table. Usually an exit condition will simply be until a given time in days.
Cycling of Schedules / Loops - it is possible to set the simulation to cycle over existing
schedule entries in the list for a given amount of time or until a constraint is breached.
This is potentially useful for applications such as WAG injection or polymer treatments.
To create a cycle, select the Add or Insert Sched./Cycle button. If there is more than one
schedule prior to this in the list, you will be given the option to create the new schedule as
a cycle.
Select from the list boxes the start and end points of the cycle. REVEAL does not allow
nested cycles, or cycles over schedule entries that have conditional exit conditions (e.g.
exit when water cut exceeds a certain value). This is reflected in the validation of the list
box entries. You may, however, enter a conditional exit condition on the cycle itself.

3.14.2

Thermal Fracture Schedule

This screen sets the thermal fracture update options, and is only displayed if a thermal
fracture is present. The update options are applied separately to each schedule and each
thermal fracture. The 3D thermal fracture calculation is cpu intensive and the fracture
shape should not generally be updated every timestep.
If the fracture update is set to off, then no fracture calculation will take place for the
current schedule. If it is set to on, then three update criteria are provided. The first
criteria to be met will cause the fracture shape to be updated. If an update criteria is left
blank then that criteria will not be used.
The fracture is automatically updated every time a new schedule is started.
If more than one fracture is associated with a well, then a fracture update will be
performed including all fractures associated with the well every time any of the fracture
update criteria are met.
Days - the fracture will be updated at a constant frequency.
Timesteps - the fracture will be updated every given number of timesteps. Often
recommended to ensure the fracture is updated with some regularity (say every 10
timesteps).
Pressure - the fracture will be updated when the pressure at the centre of the fracture
changes by a fixed amount. The fracture model calculates the fracture size, grid
connection factors, central pressure and injection rate every time it is updated. The well

OCTOBER 2003

REVEAL MANUAL

84 - 84 CHAPTER 3 USING THE INPUT WIZARD

constraint was used during the fracture update to calculate the fracture pressure, injection
rate and connection factors. Therefore, after subsequent normal iterations (without
fracture update using the same connection factors), a change in the fracture pressure
reflects changes in the reservoir arising from the injection and is a good measure of when
the fracture should be undated. This option is recommended with an update pressure of
about 50 psi. If the well connected to the fracture is controlled by a constant Pwf, then
this update parameter will not result in fracture updates.

PETROLEUM EXPERTS LTD

4 Grid Refinement
This section describes the grid refinement available within REVEAL. The following
image is of a grid with five refined regions.

Grid refinement is not available for radial or curvilinear grids, or using the original
IMPES solver. Grid refinement is incorporated directly into the solvers with no
additional iterations required.
Every REVEAL simulation has a master grid, which consists of a topologically regular
hexahedral grid, with NX, NY and NZ blocks in the X, Y and Z directions. Therefore
a total of NX.NY.NZ grid blocks are defined and may be identified by three integer
indices (I,J,K). Some of the master grid blocks may be inactive (reference porosity
set to zero). Grid refinement is achieved by defining a set of contiguous (touching)
master grid blocks, and then defining refinement parameters for the refined regions.
More than one refined region is possible.
A maximal range is defined for each refinement, it includes all of the master grid
blocks contained within a region (Imin,Jmin,Kmin) to (Imax,Jmax,Kmax), where
subscripts min and max are the minimum and maximum master grid indices in each
principal direction for the refined region.
The following graphic provides an example. Refinement (yellow blocks) is defined for

2-6

CHAPTER 4 GRID REFINEMENT

(1,3,4) to (4,4,4) and (3,5,4) to (7,5,4). The maximal range is (1,3,4) to (7,5,4),
(outlined in red).

The following is a list of limitations applied to the selection of refined regions.

1

Refined blocks must consist of a connected (contiguous) selection of master grid

blocks.

The boundary of each refined region must either be coincident with the master
grid boundary or be surrounded by master grid blocks, not other refined regions
i.e. refined regions must not touch each other. This limitation is currently
extended to preclude any two refined regions that have any overlap in their
maximal range.

Any well or fracture (thermal or producing) present within a refined region must
be completed entirely within the refined region. Wells or fractures may not cross
the boundary between refined and unrefined regions.

It is best (minimises memory requirements) to specify the refined region in as few

contiguous regions as possible. For example, the above example could have
been inefficiently defined block by block as (1,3,4) to (1,3,4) and (2,3,4) to (2,3,4)
and (3,3,4) to (3,3,4) and (4,3,4) to (4,3,4) etc.

The extent of the refined region is defined within the reservoir section of the main
input script. Any wells or fractures present within a refined region are identified in the
wells section of the main input script, and their properties assigned in a separate
refinement script. The well control (fixed Pwf, Pws or rate and constraints) for all
wells is controlled within the schedule section of the main input script.
Once the extent of the refined region and any wells within the refined region are
identified within the main input script, the refinement script may be initialised. This is
achieved by selecting Project|Create a refinement file... from the main dropdown
menu and selecting the refinement to be edited. The refinement input script has the
same form as the main input script, using a script generating wizard, except that
fewer sections are required. The following pages detail the input options required to
be entered in the refinement script.
refinement section - defined the degree of refinement
reservoir section - initialise physical properties for the refined region
aquifer section - add aquifer to refined region

PETROLEUM EXPERTS LTD

CHAPTER 4 GRID REFINEMENT

3-6

wells section - initialise well location and completion properties within the refined
region
voltage section - physical properties for well-bore heating model
initialisation section - initial component concentrations within refined region

4.1 Refinement Section

This section contains two screens and initialises the degree of refinement for a
refined region previously defined in the reservoir section of main input script.
The first screen is used to enter a label for the refinement. Generally it is best to use
a separate label for each refinement datablock. Enter a unique label name and
select Edit Refinement. Once the refinement has been entered use the Plot and
Validate buttons to check the refinement coverage.
The second screen defines the grid refinement. The coverage section at the top of
the screen identifies the maximal range for the refinement. The maximal range is
listed for each principal direction (X,Y and Z) and covers blocks on the master grid.
Only some of the blocks implied by the maximal range may actually be refined (e.g. if
the refined region is not a complete hexahedral).
The refinement is defined separately in the X, Y and Z directions. Select the
direction for data to be entered. X, Y and Z direction data must all be entered. For
each direction, data must be entered for each master grid block listed in the
coverage, regardless of whether refinement is actually required. Use the Add and
Delete buttons to add data for each master grid block in the maximal range.
The refinement data is therefore entered for each master grid block in each of the
three principal directions. Refinement data takes the form of normalised ratios to
subdivide the master grid block. A value of 0.5 will subdivide the master grid block
into two equal blocks, values of 0.2, 0.4, 0.6 and 08 will divide the master grid block
into five equal blocks, and if no data is entered, then no refinement is recorded. In
this last case a warning will be applied, but the simulation may be run.
In the following example, the master grid is 15 by 15 by 1 (NX by NY by NZ) and the
refined region is identified as including master grid blocks (9,10,1) to (10,10,1) and
(9,9,1) to (9,9,1) representing a simple 'L' shape. Therefore, the maximal range in
the X, Y and Z directions are (9 to 10), (9 to 10) and (1 to 1). The following
refinement data was entered.
direction
X
X
Y
Y
Z

master block
9
10
9
10
1

refinement
0.5
0.3 and 0.7
0.5
0.3 and 0.7
0.2, 0.4, 0.6 and 0.8

The table entry for X direction, master grid block 10 is shown below.

OCTOBER 2003

REVEAL MANUAL

4-6

CHAPTER 4 GRID REFINEMENT

This results in a refined grid with NX = 5, NY = 5 and NZ = 5, in an 'L' shape.

PETROLEUM EXPERTS LTD

CHAPTER 4 GRID REFINEMENT

5-6

4.2 Reservoir Section

This section contains five screens and initialises the required reservoir properties for
the refined region (porosity, permeability, transmissibility multipliers, net to gross ratio
and rock type). Use the Plot and Validate buttons where available to check
coverage. See the reservoir input script for the master grid input script for more
information entering data on these screens.
Porosity - either enter porosities for the refined grid using the same input method as
was used for the master grid, or select the Use parent grid values button, which will
copy the values from the master grid into the refined grid variables.
Permeability - either enter permeabilities for the refined grid using the same input
method as was used for the master grid, or select the Use parent grid values button,
which will copy the values from the master grid into the refined grid variables. Note
that if parent grid properties are required, then these should be set for each of the X,
Y and Z directions separately using the tab at the bottom of the screen.
Transmissibility - this is optional, the default being unit transmissibility multipliers.
Any transmissibility multiplies set for the master grid will be used automatically for
refined grid blocks on the boundary of the refined region, but transmissibility
multipliers required internally within the refined region should be entered explicitly on
this screen.
Net to gross - this is optional, the default being unit net to gross ratios. Either enter
net to gross ratios for the refined grid using the same input method as was used for
the master grid, or select the Use parent grid values button, which will copy the
values from the master grid into the refined grid variables.
Rock type - enter the coverage of each rock type within the refined grid. Rock types
are defined within the main grid script and will be listed in the rock list. Ensure full
coverage of the refined region.

4.3 Aquifer Section

This section contains one screen and is only available if an aquifer model is specified
in the control section of the main input script. See the aquifer input screen in the
main input script for more information on the data entry required.

4.4 Wells Section

This section contains one screen for well description, and optionally two screens for
fracture and mechanical rock properties. The well(s) and fracture(s) (if present) will
have already been identified within the wells section for the main grid. The position
and properties of the wells/fractures are entered in this section in a similar manner to
wells/fractures within the main grid.
See the wells section in the main input script for more information on the data entry
required.

OCTOBER 2003

REVEAL MANUAL

6-6

CHAPTER 4 GRID REFINEMENT

4.5 Well-Bore Heating Section

This section contains one screen and is only available if a well-bore heating model is
specified in the control section of the main input script. See the voltage screen in the
main input script for more information on the data entry required.

4.6 Initialisation Section

This section contains one screen and is only available if more additional components
(other than water, oil and gas) are present within the model. The initial component
concentrations are entered in a similar way to those within the main grid. See the
initialisation screen in the main input script for more information on the data entry
required.

PETROLEUM EXPERTS LTD

5 Menu Commands
This section describes the dropdown menu items on the main REVEAL bar. These
options are available when a REVEAL archive has been loaded. If the current
window is the 3D graphics window, then the Options, Edit and Results menus are
different. The additional dropdown menus Playback and Multi views are also
available if the current window is the 3D graphics window (see the graphics help for
more information on the 2D and 3D displays).
File
Options
Edit
Input
Project
Run Simulation
Results
View
Window
Help
Playback (3D only)

5.1

File

New
Open...
Close
Save Project
Save Project As...
Save Scripts
Import
Print...
Print Preview
Print setup...
Data Directory
(files)
Exit

open a new REVEAL archive.

open an existing REVEAL archive.
close the current REVEAL archive.
save the current REVEAL archive.
save the current REVEAL archive with a new name.
save changes made to the input script to the current REVEAL
archive, does not save result data.
import a script file (*.bpi), PVT or lift curve. It is recommended
that the PVT or lift curve are imported into the required
sections (physical and injection schedule) as data is entered.
print the input script or 3D display, depending on which window
is active.
preview and print the image to printed.
initialise the printer and associated options.
set the default directory tree that will be used for Open and
Save operations.
list of the previous 10 REVEAL archives, select one to open it.
end the REVEAL session.

2-5

CHAPTER 5 MENU COMMANDS

5.2

Options

Units
Memory Allocation

set or change the units system to be used, all data in the input
script and wizard will be changed to reflect changes to the
units.
maximum memory before REVEAL swaps refinement grid data
to disk. This should generally be left at the default value of
4GB corresponding to the maximum memory allocatable on
the windows 32 bit architecture. The operating system will
automatically swap data to the disk as required, if the model is
larger than the RAM available.

If the current REVEAL window is the 3D graphics then the following menu options are
available.
General
Viewing Properties
Scale
Playback
Preset Options
Reset View

5.3

select variable to be displayed, use cell (recommended) or

point data (only for regular grids) interpolation.
various options to select part of a grid to view, alter
transparency, shrink grid blocks, view wireframe.
alter the colour table end points manually, exact end points or
rounded end points, also set linear or logarithmic range.
movie playback settings, frequency, range and speed.
save or load previously saved 3D settings.
rescale the view to show the entire grid.

Edit

Undo
Cut
Copy
Paste
Clear
Select All
Find...
Find Next
Replace...
Goto Section

undo last change made while editing the input script.

delete selected text in the input script.
copy selection to clipboard, from any text (input script, debug
etc.).
paste clipboard (input script only).
delete selected text in the input script.
select entire text (input script, debug etc.)
find text in an text file (input script, debug etc.).
repeat previous find.
find and replace text in the input script.
scroll the input script to the selected section.

If the current REVEAL window is the 3D graphics then the following menu options are
available.
Select Multiple Blocks select some blocks using IJK coordinates to view 2D plot of
variables (against time). Use Results to view the selected
blocks.
Unselect All Blocks deselect all blocks selected.

PETROLEUM EXPERTS LTD

CHAPTER 5 MENU COMMANDS

5.4

Input

Script Wizard
Control
Reservoir
Physical
Relperm
Aquifer
Mobility
Phase
Adsorption
Water Chemistry
Wells
Well-bore Heating
Initialisation
Schedule
Import from simulator or ASCII file ...
Water Chemistry Equilibration Calculation
Thermal Fracture IPR Calculation
View System Geometry

5.5

3-5

start the input wizard, for a single section

or multiple sections.
start the wizard at a selected screen in
the control section.
start the wizard at a selected screen in
the reservoir section.
start the wizard at a selected screen in
the physical section.
start the wizard at a selected screen in
the relperm section.
start the wizard at a selected screen in
the aquifer section.
start the wizard at a selected screen in
the mobility section.
start the wizard at a selected screen in
the phase section.
start the wizard at a selected screen in
the adsorption section.
start the wizard at a selected screen in
the water chemistry section.
start the wizard at a selected screen in
the wells section.
start the wizard at a selected screen in
the well-bore heating section.
start the wizard at a selected screen in
the initialisation section.
start the wizard at a selected screen in
the schedule section.
import basic geometry and physical
properties. See importing for more
details.
water chemistry calculator - requires
water chemistry data to be present in
input script
thermal fracture IPR calculator - requires
thermal fracture calculation to have
saved (debug) restart files
view the entire model topology, including
wells and refined regions.

Project

Edit / View Project... view, delete or reimport various files contained in the REVEAL

4-5

CHAPTER 5 MENU COMMANDS

5.6

Run Simulation

Select Properties...
Start/Resume
Do One step
Run until
Pause
Stop
Restart Simulation...
Generate a Restart Step

Generate a Debug Restart Step

Show Output
Show Input

5.7

select the properties and frequency of variables to

be stored for visualisation during and after the
calculation.
start a calculation, or resume after a calculation
has been paused.
do one timestep and pause the calculation.
run the calculation until a specified runtime is
reached.
pause a calculation.
stop a calculation, a stopped calculation cannot
be restarted.
restart a simulation from a restart file.
manually generate a restart file, pause (do not
stop) a calculation before using this option. When
a simulation is paused, it may not pause at the
end of a timestep, to complete the current
timestep and allow the restart file to be generated,
use the 'Do One step' command. The restart file
contains just the data required to restart a
simulation.
manually generate a debug restart file, pause (do
not stop) a calculation before using this option.
The debug restart file contains every data
structure (including workspace) used by REVEAL
and may be very large: in general the first restart
option (above) should be used.
show/hide the calculation summary window.
show/hide the input script window.

Results

Average Reservoir Results

Well Results
Reservoir Graphics
Debug

display and plot average and total reservoir properties.

display and plot well properties.
start the 3D graphics display.
view the debug or water chemistry debug files. These
text files contain diagnostic information not displayed
elsewhere.

If the current REVEAL window is the 3D graphics then the following menu options are
available.
Results of Selected Cells

PETROLEUM EXPERTS LTD

display and plot the results (2D variable v time) of

selected data cells in the 3D display.

CHAPTER 5 MENU COMMANDS

5.8

View

Toolbar
Statusbar
Navigator

5.9

5-5

enable/disable the run/graphics toolbar towards the top of the

main REVEAL window.
enable/disable the status bar at the bottom of the REVEAL
window.
enable/disable the HelpViewer and Navigator window on the
left of the REVEAL window.

Window

Cascade
Tile
Arrange Icons
(windows)

cascade the REVEAL windows.

tile the REVEAL windows, best viewed with four windows.
arrange any iconified windows.
bring a selected window to the foreground.

5.10 Help
Help Index
Help Search
Script On-Context
About REVEAL

Start the online help at the first page (Contents).

Start the online help.
Brief description of keyword in input script, move the mouse
over the script to view the on-context help.
REVEAL version and Petroleum Experts contacts.

5.11 Playback
Forward
Backward
Stop
Pause
Resyncronise
Goto Timestep

OCTOBER 2003

move forward (in time) through previously run case.

move backward (in time) through previously run case.
stop the 3D movie or stop the 3D update during a calculation.
pause the 3D movie or pause the 3D update during a
calculation.
after pausing the 3D update during a calculation, reconnect to
the calculation.
goto the 3D results for an input timestep.

REVEAL MANUAL

6 External Data Import

6.1

Importing Overview

This section describes how data can be imported into REVEAL from other simulators
or ASCII files. In general, there is no one-to-one correspondence between REVEAL
data and that of other simulators, and so a certain amount of post-import
manipulation may be required. See the individual sections below for more
information.
Enter the import facility by invoking Input|Import from simulator or ASCII file.
The import procedure consists of an Import Wizard of three screens to take you
through the import process. The three steps are:
1. Load the simulation or ASCII case into REVEAL
2. Specify the properties that you wish to import
3. View the import results to check for errors.
1. Loading the case to import
Select from the top the format that the data is in. At the moment the following formats
are supported: ASCII, Eclipse binary, Eclipse ASCII, VIP MAP, and VIP VDB (split
or non-split). The details of this screen obviously depend on the format selected - for
more information go to the specific sections in the help.
2. Specifying the import properties
Regardless of the import format, the second screen presents the following options:
Geometry/Reservoir property data - if you import any of these, you must also
import the grid nx, ny, nz dimensions. The sections are as follows:
Corner point geometry - this data represents the (xyz) corner points of each vertex
of each grid hexahedron, and so specifies the topography of the reservoir system.
Data sets are always imported in corner point geometry, regardless of how they were
originally generated (for example, as a Cartesian or Radial grid).
Porosity - this imports the porosity reference value for each grid block.
X-/Y-/Z-Permeability - this imports the directional permeability data for each grid
block.
Net To Gross
This imports the net-to-gross value for each grid block.
X-/Y-/Z-Transmissibility multiplier - this imports the directional transmissibility
multiplier for each grid block.
X-/Y-/Z-Transmissibility - this imports the absolute inter-block transmissibilities for
each grid block (in the positive direction, i.e. X+, Y+, Z+)

2-9

CHAPTER 6 EXTERNAL DATA IMPORT

Non-Neighbour Connections - this imports non-neighbour connection data.

PV Multiplier - this imports the multiplier applied to the pore-volume.
Tabular data - this includes the relative permeability and PVT data tables. See the
individual sections on Eclipse or VIP for more information.
Rock data - this imports the rock type array and, where possible, rock
compressibilities. See the individual sections on Eclipse or VIP for more information.
PVT/Equilibration region data - this imports regional data.
Restart Data - solution data. It is possible to restart REVEAL from an Eclipse or VIP
restart file. Select the required date from the drop down lists.
3. Completing the import
A status screen will be displayed: this tells you whether the import of each section
has been successful or not. If any section has failed to import, highlight that section
in the list and press Details. A message box will be displayed with the reason for the
failure.
On the right hand side of the status screen are some notes on the data that has been
imported. See the individual sections on the data source in question for more
information.

6.2

Importing from ASCII

A single text file is required for the ASCII data import option. The text file to import
should be entered on the first screen in the import wizard. The units for the data
should be those currently selected for the REVEAL case.
The following keywords and data are required:
DIMENSION - followed by the grid dimensions NX, NY and NZ.
Either NODES or NODES_8PT keywords are required.
NODES - followed by the node coordinates X,Y,Z for each node. There will be a total
of (NX+1).(NY+1).(NZ+1) nodes, and they will be ordered cyclically in the X, Y then Z
directions with the X direction cycling fastest.
NODES_8PT - followed by the node coordinates X,Y,Z for each node. There will be
a total of (2*NX).(2*NY).(2*NZ) nodes, a total of 8 corner nodes per block. The 8
nodes for each block are entered sequentially, and the blocks are ordered cyclically
in the X, Y then Z directions with the X direction cycling fastest. Therefore there will
be 24 coordinates entered for each block. The 8 corner nodes should be ordered as
shown below.

PETROLEUM EXPERTS LTD

CHAPTER 6 EXTERNAL DATA IMPORT

3-9

3
Y

2
8

END - marks the end of the file.

The following keywords and data are optional for grid block data. For each case
there will be NX.NY.NZ entries, and they will be ordered cyclically in the X, Y then Z
directions with the X direction cycling fastest.
POROSITY - followed by the grid block porosities (inactive blocks have a zero
porosity).
PERMX - followed by the grid block permeabilities in the X direction.
PERMY - followed by the grid block permeabilities in the Y direction.
PERMZ - followed by the grid block permeabilities in the Z direction.
NETTOGROSS - followed by the grid block net to gross ratios.
TRANSX - followed by the grid block transmissibility multipliers in the X direction.
The transmissibility multipliers are applied to the block face in positive direction.
TRANSY - followed by the grid block transmissibility multipliers in the Y direction.
TRANSZ - followed by the grid block transmissibility multipliers in the Z direction.
ABSTRANSX - followed by the absolute grid block transmissibility values in the X
direction
ABSTRANSY - followed by the absolute grid block transmissibility values in the Y
direction
ABSTRANSZ - followed by the absolute grid block transmissibility values in the Z
direction
ROCKTYPE - followed by the grid block rock types (integers > 0).
PVTREG - followed by the grid block PVT region identifier (integers > 0).
EQLREG - followed by the grid block equilibration region identifier (integers > 0).
PVMULTIPLIER - followed by the grid block pore volume multipliers.

OCTOBER 2003

REVEAL MANUAL

4-9

CHAPTER 6 EXTERNAL DATA IMPORT

NNC - followed by the number of non-neighbour connections, followed by the

connections themselves in the format: (x1,y1,z1)-(x2,y2,z2), T. See below for an
example.
A trivial example file is listed below.
DIMENSION
222
NODES
000
100 0 0
200 0 0
0 100 0
100 100 0
200 100 0
0 200 0
100 200 0
200 200 0
0 0 100
100 0 100
200 0 100
0 100 100
100 100 100
200 100 100
0 200 100
100 200 100
200 200 100
0 0 200
100 0 200
200 0 200
0 100 200
100 100 200
200 100 200
0 200 200
100 200 200
200 200 200
POROSITY
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
PERMX
100
100
100
100
100
100
100
100
NETTOGROSS
11111111

PETROLEUM EXPERTS LTD

CHAPTER 6 EXTERNAL DATA IMPORT

5-9

TRANSZ
100 100 100 100 100 100 100 100
NNC
1
1 1 1 2 2 2 0.3
END

6.3

Importing from Eclipse (binary files)

There is no one-to-one correspondence between an Eclipse data set and a REVEAL

data set. This section describes the special processing carried out by REVEAL to
import Eclipse cases.

6.3.1

Data files Required

To import the geometrical and physical data (including tabular data), the Eclipse files
required are the binary output grid geometry (*.grid or *.grd or *.egrid) file and
initialisation file (*.init or *.ini). Formatted version of these files can also be read (.fgr
and .fin). See the Eclipse manual if these files have not been generated.
To import the restart data (solution and well data) restart file(s) will be required (*.rst
or *.unrst). These may be in individual or unified format (this must be specified on the
load screen).

6.3.2

Rock Type Arrays and Compressibilities

There is no direct analog to the REVEAL rock type within Eclipse. Rock type arrays
can be imported optionally from the SATNUM, PVTNUM, or EQLNUM arrays.
Rock compressibilities can only be imported if a rock type array is imported from the
PVTNUM data in Eclipse. (This is because there should be a rock compressibility
value defined for each individual value in the PVTNUM table). The rock
compressibility is only available if the rock compaction facility is not being used in the
Eclipse data set.
Note also that the rock compressibilities are set at a given reference pressure. This
reference pressure is not imported into the REVEAL data set and should be set
manually if required. This is because the REVEAL reference pressure is used in
calculations other than that for rock compression.

6.3.3

Relative Permeabilities

In REVEAL, residual saturations, relative permeabilities, and capillary pressures are

all dependent on a defined rock type. Each rock type has a set of residual saturations
for all phases, and for each set of residual saturations there will be a set of capillary
pressure tables and relative permeability tables. There may be more than one set of
rel perm tables for each set of residual saturations - indeed, this would be required if
you were using directional rel perms.
In Eclipse, there is no direct analog to the REVEAL rock type. Arrays such as
SATNUM and PVTNUM have similar meanings but are used in different ways within
Eclipse (for example, SATNUM is used to specify rel perm tables, PVTNUM is used

OCTOBER 2003

REVEAL MANUAL

6-9

CHAPTER 6 EXTERNAL DATA IMPORT

to specify separate PVT areas and, optionally, rock compressibility values).

For this reason, no attempt is made to assign the imported rel perm data to any
particular rock type regardless of whether the rock regions have been defined or not.
Each rel perm table (of which there may be several) is assigned to All Rock Types,
and it is up to the user after the import has been carried out to assign the rel perms to
the required rock region. See the Relperm section on the input wizard for more
information on how to do this.
For Eclipse two phase systems, REVEAL will generate dummy rel perm tables for the
phases that are not present in the simulation. These, of course, will not affect the
calculation if the phase in question remains absent.

6.3.4

PVT

REVEAL imports oil or gas PVT data from Eclipse. It is not currently possible to
import an Eclipse condensate PVT.
The Petroleum Experts black oil PVT module consists of two sections: a correlation
data set and a tabular data set. If both sets are present, the tabular data is used in
preference: the correlations are used to fill in 'GAPs' in the tabular data. When a PVT
table is imported from Eclipse, it is expected that all the required tabular data will be
present and so there will be no need for the correlation data. Dummy correlation data
is generated by the import process: this should not affect the PVT calculations.
The REVEAL PVT is always three phase (water, oil, and gas). It is nevertheless
possible to import Eclipse 'dead oil' (i.e. no gas phase or no dissolved gas) cases: in
these situations a nominal low value for Rs is provided for the saturated tables and
no undersaturated tables are generated.
When undersaturated tables are used by REVEAL, a controlled miscibility setting
must be provided by the user. This determines the rate at which gas will dissolve in
the oil if the reservoir is pressurised. The import process sets this value to zero (i.e.
gas does not redissolve once liberated): the user is free to change this value after the
import.
Eclipse PVT tables are isothermal - REVEAL tables are not. It is possible in REVEAL
to generate several saturated tables (each with up to 5 undersaturated tables
'hanging' from each) for various temperatures. When the import is carried out, only
one saturated table will be generated. The temperature for this table will be the
reference temperature for the simulation if specified, or 100 degrees F if not.
Eclipse may contain several PVT tables (corresponding to the PVT NUM array).
REVEAL only uses one black oil PVT. After the import the status screen will display
the PVT files that have been generated by REVEAL. Select the PVT file that you
would like to use in subsequent REVEAL simulations. The other PVT files will remain
within the archive and can be swapped into the simulation at a later date.
When a 'live' oil PVT is imported, you must tell REVEAL the initial bubble point
pressure for the reservoir. The is used to calculate the initial Rs from the saturated
curves that have been imported.

PETROLEUM EXPERTS LTD

CHAPTER 6 EXTERNAL DATA IMPORT

6.3.5

7-9

EndPoint Scaling

The enpoint scaling saturation arrays corresponding to the Eclipse keywords SWL,
SWCR, SWU, SOWCR, SOGCR, SGL, SGCR and SGU may be imported.
Additionally, the relative permeability endpoint scale factors KRW, KRO and KRG
may be imported.

6.3.6

Well Data

This is imported as a table over completed blocks (ijk) with connection factors as
calculated by Eclipse.

6.3.7

Solution Data

A REVEAL format restart file will be generated from which simulation restarts can be
run. See the Run Simulation menu item for more information.

6.4

Importing from Eclipse (ASCII files)

It is currently possible to bring the corner points of an Eclipse grid into REVEAL
directly from the Eclipse ASCII deck. Specifically, REVEAL can read the COORD and
ZCORN cards and translate the data into a REVEAL datablock.
The units for the data should be those currently selected for the REVEAL case.
On the first import wizard screen select the Eclipse (ASCII) option from the drop
down list. On the resulting screen you must specify the dimensions of the grid that
you are about to import, as this data may not be present in the grid file that you
specify later on. The table on the lower part of the screen allows you to enter several
ASCII files that contain the data for different areas of the grid. The columns of the
table are as follows:
Box definition - x1, y1, z1, x2, y2, z2, 'all grid': These define the sub-region of the
total grid that is to be imported from the ASCII files that follow. Alternatively, click 'all
grid' if the ASCII files contain coordinates over the entire grid.
ACTNUM file (optional): If you wish, you may specify a file which contains the
ACTNUM array for the grid so that only coordinates of active blocks are imported
COORD file: This file contains the COORD card of the Eclipse deck.
ZCORN file: This file contains the ZCORN card of the Eclipse deck.

OCTOBER 2003

REVEAL MANUAL

8-9

CHAPTER 6 EXTERNAL DATA IMPORT

6.5

Importing from VIP

There is no one-to-one correspondence between a VIP data set and a REVEAL data
set. The logic applied during the import process is the same as that applied for
Eclipse - see the Eclipse import page for more information.
Note that the VIP analogues of PVTNUM, SATNUM, and EQLNUM are IPVT, ISAT,
and IEQL.

6.5.1

Data Files Required

Two different VIP formats are supported. For any format the import process starts be
selecting 'Input | Import from simulator or ASCII file' from the main menu.
1. MAP data
In this case the files required are:
1. Core data - .map file (binary) and i.dat file (text)
2. Exec data - .map file and r.dat file - only required for the import of restarts
The core data contains the grid properties (topography, petro-physical properties)
and tabular data (PVT and relative permeabilities). The exec data contains the
solution (restart) data.
To load data of this format, select 'VIP MAP format' from the drop down list at the top
of the screen. Enter the root directory for the case under 'Directory'. Enter the files to
read: the 'core file name' is the ASCII file ending 'i.dat' (the 'i.dat' will be appended
automatically). The 'core file MAP name' is the binary file ending '.map'.
If you choose to import restart/solution data then check the box marked 'Load
Solution/Restart data'. You will then have to fill in equivalent fields to those above for
the 'exec' data.
2. VDB data
This format can appear in two different but equivalent forms - VDB files from VIP are
either 'flat' (the data is archived into a single .vdb file) or 'split' (the vdb file is actually
a separate directory under Windows).
First determine whether you have a flat or split format VDB file and check the 'Split
VDB file' box if it is necessary.
In this case the files required are:
1. Study data - .vdb file (binary)
2. Core file - ending i.dat (as above)
3. Exec file - ending r.dat (as above) - only required for the import of restarts
To load data of this format, select 'VIP VDB format' from the drop down list at the top
of the screen. Enter the root directory for the case under 'Directory'. Enter the vdb file
name (the extension .vdb will be appended automatically) and click 'Load'. The cases
in the archive will be listed in the box. You must now select the studies from the list
for the core data (grid properties) and exec data (solution data).
Enter the i.dat (and optionally the r.dat) file name in the fields at the bottom of the

PETROLEUM EXPERTS LTD

CHAPTER 6 EXTERNAL DATA IMPORT

9-9

screen. The 'i.dat' and 'r.dat' are appended automatically.

Important Note
Much of the array data (such as ISAT, IPVT etc) is not placed in the map file by
default. To ensure that all the required data is present, include the instruction 'MAP
ALL' in the VIP input deck. See the VIP manual for more information.

OCTOBER 2003

REVEAL MANUAL

7 Graphics
7.1

Visualisation Overview

Use the P,WR,3D, and Xflow buttons to view the results. The results may also be
viewed interactively during a calculation.
The P button provides average reservoir properties such as mean phase pressures,
saturations and volumes. These data may be viewed, plotted and printed, or copied
to the clipboard for import into another application.
The WR button provides well results data, including instantaneous and cumulative
rates, GOR, water cut and phase pressure and saturation data at well blocks.
The 3D button provides a three dimensional display of the fluid properties that were
selected for display (Run Simulation|Select Properties...). Note that the variables
and frequency with which they are to be stored must be set before a calculation is
run. It includes graphical representation of wells, completions and fractures.
The Xflow button provides a three dimensional schematic of the wellbore with the
intersected grid blocks. If a wellbore model has been set up in the input deck this
view will display the results of the wellbore calculations, i.e. the calculated properties
at the nodes and in the tubing. It also provides information on any crossflow that may
be occurring.
The 3D graphics requires the display driver to be operating with a screen area of at
least 1280*1024 pixels and 16 bit (high colour) resolution.
If the current window is the 3D graphics window, then the dropdown menus Options,
Edit and Results are specific to 3D graphics. The additional dropdown menu
Playback is also available if the current window is a 3D graphics window .

7.1.1

Interaction with the 3D Visualisation Windows

Rotation
- to rotate the view, click the left button of the mouse and drag the mouse in the
direction that you would like to rotate. The rotation axis is perpendicular to the
direction of mouse movement and in the current focal plane of the view.
- to rotate the view about an axis normal to the screen and passing through the
current focal point, press the ctrl key with the right mouse button. The speed of
rotation is proportional to the distance from the centre of the screen to the mouse
position.
Zooming
- area ('rubber band') zooming can be achieved by holding down the ctrl key while
dragging the mouse with the left mouse button depressed.
- alternatively, click on the magnifying glass icon (+) in the toolbar and then click and
drag across the area to be zoomed into.
- a single click with the magnifying glass icon (+) selected will perform a zoom in of a
fixed ratio, centred on the position at which the mouse was clicked.

2-4

CHAPTER 7 - GRAPHICS

- similarly, a single click the the magnifying glass icon (-) selected will perform a
zoom out of a fixed ratio, centred on the position at which the mouse was clicked.
Panning
- panning can be achieved by holding the shift key while dragging the mouse with the
left button depressed. Essentially, the viewing position is shifted parallel to the focal
plane of the view, although the effect is to 'drag' the objects from one position to
another.
Other controls
- F5 will reset the display to the initial size, such that the entire system is visible.
- W acts as a shortcut to present a wireframe mode view of the entire system.
- S acts as a shortcut to present a surface mode view of the entire system.
3D visualisation preferences
To invoke the preferences screen from any 3D view, select Options | Preferences...
from the main menu. This is also accessible from a right click.
The screen is divided into two tabs:
Lookup tables - the colour tables for different view properties can be changed. The
default is for a basic RGB graduation for pressure and default properties, with phase
saturations represented by a single colour with a graduation in colour saturation. To
change a lookup table, adjust the palette at the top of the screen to the starting (low
value) colour and click on the button at the left hand side of the lookup table that you
are changing. The table will change to reflect the selection. Then perform the same
action for the ending (high value) colour. The lookup table is calculated by
interpolating HSV values between the top and bottom of the table.
View properties:
- the perspective mode determines whether the view is generated with parallel rays
or by applying a false (but perhaps more visually pleasing) perspective based on a
view position.
- the background colour can be changed by adjusting the sliders on the colour
palette.
- the mouse rotation sensitivity governs how quickly the view is rotated for a unit
movement of the mouse position.
Main 3D View Functions
The graphics toolbar (below the main toolbar at the top of the REVEAL screen)
contains most of the functions that are available from the visualisation window. Some
functions are different depending on whether the current view is a 3D view or a
wellbore (Xflow) view - these differences are detaield below.
Going from left to right across the toolbar:
Animation buttons:
Advance the view by one step
Animate the view results from the current step
Advance to the last step
Stop an animation
... with similar buttons performing the reverse procedures...

PETROLEUM EXPERTS LTD

CHAPTER 7 GRAPHICS

3-4

If a simulation is currently being run, this pauses the graphics (i.e. disengages
the graphics from the current run)
If a simulation is currently being run, this re-synchronises the graphics with
the run (if the graphics had previously been paused)
Selection buttons:
(3D) Allows cells to be selected. As the mouse is moved over the view, the
grid cells below the mouse pointer are highlighted. Click on the required cell
to select it - 2D results for this cell can then be viewed by invoking Results |
Results of selected cells. Many cells can be selected at once.
(Xflow) Allows results for an item in the view to be displayed at an instant in
time. Move the mouse over the item in question and a window will appear
giving the current timestep results for the item. If the view contains grid blocks
and wellbore items (tubing and nodes) it can be difficult to select a tubing
node rather than a grid block. To get round this, invoke the 'Object Properties'
screen (either from the toolbar or Options | Viewing properties) and change
the 'pick mode' selections.
(3D only) Unselects all cells.
(3D only) Invokes a screen to allow cells to be selected from IJK lists.
Zoom buttons:
These allow the view to be zoomed in or out, as described above.
Results:
(3D) Invokes a screen with the time dependent results for the currently
selected set of cells.
(Xflow) Invokes a screen with time dependent results for all the items in the
system (grid blocks, calculation nodes, and tubing)
Property buttons:
(3D) Invokes a screen allowing you to change the variable that is currently
displayed. It also allows you to view point scalar data rather than cell data.
The point data is interpolated from the grid block corners and weighted by the
inverse square of the distance to the corners.
(Xflow) Invokes a screen allowing you to change the variable that is currently
displayed for the different object categories. The tubing variable can be set to
the calculated fluid rate, or alternatively can be set to display simple
interpolated values between the end nodes.
(3D) Invokes a screen that allows the properties of the objects in the view to
be altered, e.g. the transparancies or visibilities of the grid objects, and
whether to display surface or wireframe data. It is also possible, from this
screen, to display subsets of grid cells (normally a plane of blocks in X, Y, or
Z) and to change the z scale factor.
(Xflow) In addition to setting general object properties, the 'pick mode' can be
changed. The pick mode determines those objects that are candidates for

OCTOBER 2003

REVEAL MANUAL

4-4

CHAPTER 7 - GRAPHICS

picking when 'selection mode' is entered. It may be desirable, for example, to

remove grid blocks from the pick mode to allow easier picking of nodes and
inflow objects.
Scale. Sets the ranges of values that are mapped to the colour tables.
Playback. Sets various playback / animation options.
The next field is the Xflow view well selection list box. Select from the drop down list
the well data that you wish to view.
This is equivalent to pressing F5, i.e. the view is reset to the maximum zoom.
Other functions
Once a view has been created, its properties may be saved and reopened later. Use
Options|Preset Options|Save File and Options|Preset Options|Load File to achieve
this. Enter a name for the view, select the properties to be saved (usually All) and
select OK. Note that the REVEAL archive1 must be saved for the view to be saved
permanently. Many of the REVEAL example cases have default views saved.

The REVEAL archive file (*.rvl) contains the script and additional project files such as PVT,
lift curve, output graphics etc. The archive files present may be viewed, added to or deleted
using the option Project|Edit / View Project....

PETROLEUM EXPERTS LTD

8 DeBug Output
The debug output is contained within the REVEAL archive1 as a text file. It may be
viewed during or once a simulation has been completed using the menu command
Results|Debug|Debug Simulation. Note that the menu command Results changes its
function when the active REVEAL window is the 3D display.
Additional debug output is created during a water chemistry calculation and can be
view using the menu command Results|Debug|Debug Water Chemistry.
The finite element fracture variables are also output in detail to the debug output file.
The debug output file contains much general model information not displayed
elsewhere. The most useful information it contains falls into three categories.
Initialisation data - including the analytical well PI calculation results and post
equilibration saturations and pressures for each block.
Runtime data - includes timestep data which gives information on the number of
iterations (IMPES and implicit formulations), what variable is controlling the timestep,
maximum volumetric error and maximum variable changes. See the table at the end
this page for information on the timestep control summary (8 digit summary at the
end of each completed timestep). Additionally, fracture data for the finite element
grid is output in detail for each thermal fracture calculation.
Completion data - includes error and warnings at the end of a calculation, total
runtime statistics and some information on the final state of the calculation.

Digit
1

Meaning
Retake & on-hold

Timestep limited

Cause of retake

Character
#
*
h
r
v
t
c
p
i
f
d
e
n
v
m
p
t

Description
OK
Timestep to be reduced
Timestep to be retaken
Timestep to be unchanged
None
Max growth rate
Volumetric flow limit
Temperature constraint
Concentration constraint
Pressure constraint
Injection schedule
Fracture update
Implicit solver (diverging Newton steps)
Implicit solver (estimated next timestep)
No retake
Negative concentrations
Volumetric flow limit
Material balance constraint
Pressure constraint
Temperature constraint

The REVEAL archive file (*.rvl) contains the script and additional project files such as PVT, lift
curve, output graphics etc. The archive files present may be viewed, added to or deleted
using the option Project|Edit / View Project....

2-2

CHAPTER 8 DEBUG OUTPUT

Cause of on-hold

Limit reached

Volumetric flow

Implicit solver

PETROLEUM EXPERTS LTD

c
d
m
p
t
c
d
l
u
c
s
w
i
l
d
c
p
s

Concentration constraint
Implicit solver (diverging)
Not on-hold
Material balance constraint
Pressure constraint
Temperature constraint
Concentration constraint
Implicit solver (diverging)
Not limited
Minimum timestep limit
Maximum timestep limit
Total volume inflow (Courant)
Mobile phase saturation undershoot
Total well inflow/outflow
Pre-conditioner failure
Linear solver failure
Newton steps diverging
Newton steps not converged
Pressure change too large
Saturation change too large

9 Engineering
9.1 Introduction
This section provides descriptions of the reservoir engineering models and solution
methods provided by REVEAL that are not described elsewhere in detail. It
concentrates on the underlying physical processes and the numerical approximations
used to model them.
Phases and components
Transport equations
Dispersion and diffusion
Thermal equations

definitions of phases and components within REVEAL

flow equations applied for IMPES and implicit solver
options
trace component flow
heat transport applied for IMPES and implicit solver
options

9.2 Phases and Components

Phases - a phase is a distinct (immiscible with other phases) fluid. The three phases
aqueous, oleic and gaseous are always possible, but may not be present. Phases
are characterised by their saturations (volume fraction of pore volume), which sum to
one. If the surfactant component is present, then a middle (micro-emulsion) phase
may be present between the aqueous and oleic phases.
The transport equations are formulated on the basis of phase properties (saturation,
density, relative permeability, viscosity and capillary pressure).
Components - a component is a chemical species (or mixture of chemicals) that
forms the basis of the fluids present. The components water, oil and gas are always
present within all models.
Components are partitioned within phases. For an oil reservoir, the water component
is present within the aqueous phase, the oil component present within the oleic
phase and the gas component present within the oleic and gaseous phases. For a
condensate reservoir, the gas component is only present within the gaseous phase,
while the oil component is partitioned between the oleic (dropout) and gaseous
phases. For a dry gas reservoir, both oil and gas components are partitioned within
the gaseous phase (no oleic phase). If a 4th micro-emulsion phase is present, then
the water, oil and gas components will be partitioned between their original phase
and the micro-emulsion phase.
Only components that affect the phase densities are considered volumetric. These
components include water, oil, gas, non-reservoir injected gas, surfactant, and the
two possible alcohol components. All other components are non-volumetric, they
only contribute to the transport equations by modifications to phase mobilities, or not
at all for tracer components.

2-5

CHAPTER 9 - ENGINEERING

9.3

Transport Equations

Transport equations - laminar flow through a porous medium is assumed, and

quantified by Darcy's equation in oilfield units for each phase (p).

q p = 6.3266

k .krp

ph

Pw + Pc p
144

where qp is the Darcy velocity (ft/d)

k is rock permeability (D)
krp is phase relative permeability
p is the phase viscosity (cp)
p is the phase density (lb/ft3)
h is the depth (ft)
Pw is the water phase pressure (psi)
Pcp is the phase capillary pressure (psi)
The accumulation of mass (mp) of a phase into a control volume (V) is then defined
by a summation of fluxes over the the surface area (A).

m p = ( p q p A)
The accumulation of mass into volume Vp is then subject to phase compressibility
(cp) and rock compressibility.

cp =

1 V p
1 p
=
V p P
p P

mp =

(V p p )
t

V p = VS p

Equating these accumulation terms leads to a material balance equation.

( p S p )
t

( p q p A) = 0

=1

This differential equation is non-linear and also requires to be discretised both

spatially and temporally before it can be solved numerically.
For all solution methods, a harmonic average1 permeability (k) is used at grid block
faces, while different upstream weightings for face mobilities (Mp) are available for
the implicit/explicit solver options. A face volumetric transmissibility (Tp) is also
defined, resulting in a final version of the transport equation.
1

The harmonic average is the 'resisters in parallel' average. The following averages are
defined for a variable x, with weightings w.
arithmetic average

x=

xi wi

harmonic average

x=

PETROLEUM EXPERTS LTD

w
w
x

i
i

CHAPTER 9 ENGINEERING

Mp =
V

3-5

krp

( p S p )
t

T p = 6.3266 AkM p

p h

= 0
+ p T p Pw + Pc p

144

Implicit solution - for this solver option, the transport equation is solved with respect
to the oil phase pressure (Po) and the three phase saturations (Sp). A single point
upstream weighting is applied for mobility and density.
IMPES solution - for this solver option, the transport equation is solved in two steps
using a total volumetric implicit pressure (explicit transmissibility and capillary
pressure) solution to calculate the water phase pressure (Pw) and an explicit
concentration (saturation) update using the calculated phase Darcy velocities. The
implicit pressure solution uses a total compressibility term (c) including rock and fluid
phase compressibilities.

ph

Pw
= 0
+ T p 2 Pw + Pc p
t
144
p

ph
Tp

q p = Pw + Pc p
144
A

Vc

The explicit nature of this formulation allows more sophisticated upstream weightings
for phase transmissibility, trace component concentrations and density, reducing
numerical dispersion, but more severe limitations on the time-step size are required
to ensure stability.
Trace component concentrations are updated using the phase Darcy velocities and
upstream weightings for component volume fractions.
For more information on the solver options see the control section of the input wizard
help.

9.4 Dispersion and Diffusion

The dispersion and diffusion of trace components within phases is optionally
available in the physical section, and is only possible using the IMPES solver option.
The IMPES solver, first solves an implicit pressure model, resulting in phase Darcy
velocities (qp) between neighbouring grid blocks. The transport of trace components
in the explicit phase of the solve, is calculated by calculating mass fluxes of the trace
components (Fc). If dispersion and diffusion are not present, then the trace
component mass fluxes are calculated using upstream weighting of component
concentrations within associated phases (Ccp) and the phase Darcy velocities.

Fc = c q p C pc
p

If dispersion or diffusion are present, then the trace component mass fluxes are
modified by calculating a dispersive flux (Gcp), dependent on the component

OCTOBER 2003

REVEAL MANUAL

4-5

CHAPTER 9 - ENGINEERING

concentration gradient. Note that trace components are only partitioned within one
phase (unless a middle phase micro-emulsion is present) and the component density
is equal to its host phase density (i.e. volumetric and mass fractions of component
within phase are equivalent).

Fc = c q p C pc G pc
p

G pc = S p D pc + q p Tp C pc +

qp
qp

Lp

Tp ) q p C pc

where Dcp are component diffusion coefficients, and Lp and Tp are longitudinal
and transverse phase dispersion lengths.
Numerical dispersion may be minimised by selecting the fct/2-point upstream
weighting for component concentration in the control section of the input script.

9.5 Thermal Equations

For both the IMPES and implicit formulations, the thermal heat transport equations
within REVEAL are applied after the pressure and component concentrations have
been updated. Therefore, the phase Darcy velocities between grid blocks is known
and the heat transport is calculated explicitly with respect to the flow and pressure.
Convective heat transport (heat flowing with the fluid) and optionally conductive heat
transport (heat flowing with thermal gradient) are included.
The accumulation of heat (J) into a control volume (V) is defined by a summation of
conductive and convective fluxes over the surface area (A). The thermal conductivity
is a harmonic average2 between adjacent blocks, and includes contributions from the
rock and all phases present. The phase heat capacities (Cpp) are those calculated
at the end of the timestep. In oilfield units, J has dimensions of (BTU/d).

J = ( AkT ) (q p p Cp p AT )
p

The accumulation of heat into volume V then results in a temperature change (heat
transport equation).

(HT )
J =0
t

The harmonic average is the 'resisters in parallel' average. The following averages are
defined for a variable x, with weightings w.
arithmetic average

x=

xi wi

harmonic average

x=

PETROLEUM EXPERTS LTD

w
w
x

i
i

CHAPTER 9 ENGINEERING

5-5

where H is the total heat capacity (BHU/F) of the control volume V (grid block),
calculated before and after the flow is updated (beginning and end of timestep).
For the IMPES formulation, or the implicit formulation when the CLF3 limit is selected,
the heat transport equation is solved explicitly for temperature (T) i.e. J is first
calculated, then T is updated. This has the advantage that upstream weighting
models (fct/2-point recommended) may be used for convective thermal fluxes,
reducing numerical dispersion. The upstream weighting is the method used to
estimate the heat capacity (p.Cpp) at the interface between adjacent blocks.
For the implicit formulation, the CLF limit is not observed, and the explicit method
used for the IMPES formulation is unstable. Therefore, the temperature is calculated
by implicitly solving the heat transport equation in one step, a 1 point upstream
weighting is used for calculating convective thermal fluxes.
Heat source terms are also added for well-bore heating, well inflow/outflow and
overburden/underburden heat flow.

Courant Flux Limit (CFL) ensures that no more than one pore-volume throughput through
any grid block occurs within a time-step.

OCTOBER 2003

REVEAL MANUAL

Appendix A - References
A.1

References

Van Everdingen A.F., Hurst W., The Application of the Laplace

Transformation to Flow Problems, Trans. AIME 156, 1949.

Carter R.D., Tracy G.W., An Improved Method for Calculating Water Influx,
SPE 1626-G, 1960,

Geertsma J., deKlerk F., A Rapid Method of Predicting Width and Extent of
Hydraulically Induced Fractures, SPE 2458, 1969.

Wiliams B.B., Fluid Loss from Hydraulically Induced Fractures, SPE 2769,
1970.

Todd M.R., Longstaff W.J., The Development, Testing and Application of a

Numerical Simulator for Predicting Miscible Flood Performance, SPE 3484,
1972.

Cinco-Ley H., Ramey H.J., Miller F.G., Pseudo-skin Factors for PartiallyPenetrating Directional-Drilled Wells, SPE 5589, 1975.

Peaceman D.W., Interpretation of Well-Block Pressures in Numerical

Reservoir Simulation, SPE 6893, 1978.

Clifton R.J., Abou-Sayed A.S., On the Computation of the Three-Dimensional

Geometry of Hydraulic Fractures, SPE 7943, 1979.

Zolotukhin A.B., Analytical Definition of the Overall Heat Transfer Coefficient,

SPE 7964, 1979.

10

Vinsome P.K., Westerveld J., A Simple Method for Predicting Cap and Base
Rock Heat Losses in Thermal Reservoir Simulators, J. Can. Pet. Tech., 1980.

11

Clifton R.J., Abou-Sayed A.S., A Variational Approach to the Prediction of the

Three-Dimensional Geometry of Hydraulic Fractures, SPE 9879, 1981.

12

Perkins T.K., Gonzalez J.A., Changes in Earth Stress Around a Wellbore

Caused by Radially Symmetrical Pressure and Temperature Gradients, SPE
10080, 1984.

13

Peaceman D.W., Interpretation of Well-Block Pressures in Numerical

Reservoir Simulation with Nonsquare Grid Blocks and Anisotropic
Permeability, SPE 10528, 1983.

14

Perkins T.K., Gonzalez J.A., The Effect of Thermoelastic Stresses on

Injection Well Fracturing, SPE 11332, 1985.

15

Prouvost L.P., Pope G.A., Rouse B., Microemulsion Phase Behaviour: A

Thermodynamic Modeling of the Phase Partitioning of Amphiphilic Species,

2-2

APPENDIX A - REFERENCES

SPE 12586, 1985.

16

Prouvost L.P., Satoh T., Sepehrnoori T., Pope G.A., A new Micellar PhaseBehaviour Model for Simulating Systems with up to Three Amphiphilic
Species, SPE 13031, 1984.

17

Christie M.A., Bond D.J., Multidimensional Flux-Corrected Transport for

Reservoir Simulation, SPE 13505, 1985.

18

Koning E.J.L., Fractured Water Injection Wells - Analytical Modelling of

Fracture Propagation, SPE 14684, 1985.

19

Vandamme L., Jeffrey R.G., Curran J.H., Pressure Distribution in ThreeDimensional Hydraulic Fractures, SPE 15265, 1986.

20

Meyer B.R., Heat Transfer in Hydraulic Fracturing, SPE 17041, 1989.

21

Clifton R.J., Wang J-J., Multiple Fluids, Propant Transport, and Thermal
Effects in Three-Dimensional Simulation of Hydraulic Fracturing, SPE 18198,
1988.

22

Detienne, J-L., Creusot M., Kessler N., Sahuquet B.,Bergerot J-L., Thermally
Induced Fractures : A Field Proven Analytical Model, SPE 30777, 1995.

23

Parkhurst D.L., User's Guide to PHREEQC-a Computer Program for

Speciation, Reaction-Path, Advective-Transport, and Inverse Geochemical
Calculations, US Geological Survey, 1995.

PETROLEUM EXPERTS LTD

Appendix B Sample Files

B.1

Sample Files

The following is a brief description of the sample files included with the installation.
Depending on where REVEAL was installed, they may be found in C:\Program
Files\Petroleum Experts\IPM #\Samples\reveal, where # is the IPM (Integrated
Production Management) release version number.
Fracture examples
thermal_fracture.rvl
producing_fracture.rvl
Water chemistry examples
water_chemistry.rvl
watchem_curvilinear.rvl - simple Barite production example using curviliear grid.
souring.rvl - simple souring example with Barite precipitation.
souring_simple.rvl - simple souring example without minerals.
scale_inhibitor.rvl - simple souring example with Barite scale inhibition.
Chemical additive examples
surfactant.rvl - surfactant micro-emulsion phase created.
foam.rvl - simple foamer treatment to reduce gas coning.
gel.rvl - simple isothermal gel/polymer injection affecting mobilities.
Grid geometry examples
radial.rvl - radial grid.
curvilinear.rvl - repeated 5 point well curvilinear grid.
corner.rvl - simple corner point example with analytical multi-lateral well.
Aquifer examples
aquifer_radial.rvl - simple grid with a radial aquifer.
Gas-Condensate examples
condensate.rvl - example with a condensate PVT description.
dry_gas.rvl - example with a dry gas PVT description.
Thermal examples
wb_heating.rvl - example heating a heavy oil to reduce viscosity.
Openserver examples
Well_control.xls - example of the use of the OpenServer to automate well control
(THP and injection temperature).
Batch_matching.xls - example of the use of the OpenServer to automate batch
calculations for sensitivity/matching analysis.
Well_production.xls - example of the use of the OpenServer to report block by block
well production rates.
Step-by-step examples
example1.rvl - step by step example creating a simple model from scratch.

2-2

APPENDIX B SAMPLE FILES

Implicit solver examples

producing_fracture.rvl
Grid refinement examples
producing_fracture_refine.rvl

PETROLEUM EXPERTS LTD

Appendix C FQAs
C.1

Frequently Asked Questions

Q1

Data lost when wizard ended.

Q2

Oscillations in results.

Q3

Copy and paste between wizard screens and other applications.

Q4

Which files do I require to import an Eclipse data set.

Q1:

Data lost when wizard ended.

A1

Data is held in temporary files until the archive is saved using the Save menu
option. The temporary files in the archive may be viewed using the menu
option Project|Edit / View project, or by opening the REVEAL archive (*.rvl)
using winzip. The principal temporary file is the input script, which contains the
changes made using the wizard.
Changes made using the wizard will be updated to the input script if the wizard
is ended using Finish, or if the wizard ends because Next was selected and no
more wizard screens are available (either the end of the schedule section or the
end of the current section is only one section is being edited with the wizard).
In this case, the data in the input script is over-written and previous data will be
lost.
If the wizard is ended using the Cancel button, then the changes made using
the wizard are not saved to the input script (these changes are lost).

Q2:

Oscillations in results.

A2

Oscillations are generally caused by incomplete convergence of solutions

(material balance errors) or time-steps that are too long.
To improve convergence, use the implicit solver option where possible
(especially for cases with only 3 components, water, oil and gas). Using the
implicit solver, the convergence criteria may be tightened by reducing the
pressure, saturation and implicit well limits from 0.01 (default) to 0.001.
Using, the IMPES solver, material balance may be improved by reducing the
volumetric error from 0.0001 to 0.00001. More iterations may be required to
achieve the reduced material balance error so also increase the number of
iteration from 3 to 5. Also, the saturation underflow limit may be set to 0.001
to minimise oscillation in gas flow near its residual saturation.
Also using the IMPES solver, under some circumstances, the timestep may
also be controlled by setting a maximum limit on pressure, temperature or
component concentration change.

2-2

APPENDIX C FREQUENTLY ASKED QUESTIONS

If tabular PVT data is being used, check that it covers the full range of
pressures and temperatures, and has no discontinuities. If a matched (or
unmatched) correlation PVT data is being used, it may be useful to convert it
to tabular format using the option Calc Tables found on the PVT input screen
(see physical properties section).
Check relative permeability and capillary pressure data are reasonable
(smoothly monotonic).

Q3:

Copy and paste between wizard screens and other applications.

A3:

Data may be copied and pasted between the REVEAL wizard screens and
other applications such as EXCEL. To copy from REVEAL, select the region
to be copied and use Control+C to copy and Control+V to paste the selection.
Alternatively, entire sections may be copied using options obtained using a
right mouse click. To paste data into REVEAL, the size (number or columns)
of the region to be pasted must match that required by REVEAL. The
simplest way to check the column format required is to select and copy a
blank region from REVEAL into EXCEL (for example).

Q4

Which files do I require to import an Eclipse data set.

A4

See the external data import section.

PETROLEUM EXPERTS LTD