Cmost Manual 2009

User's Guide

CMOST

Version 2009

By Computer Modelling Group Ltd.

This publication and the application described in it are furnished under license exclusively to the licensee, for internal use only, and are subject to a confidentiality agreement. They may be used only in accordance with the terms and conditions of that agreement.

All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic, mechanical, or otherwise, including photocopying, recording, or by any information storage/retrieval system, to any party other than the licensee, without the written permission of Computer Modelling Group.

The information in this publication is believed to be accurate in all respects. However, Computer Modelling Group makes no warranty as to accuracy or suitability, and does not assume responsibility for any consequences resulting from the use thereof. The information contained herein is subject to change without notice.

Copyright 1987-2009 Computer Modelling Group Ltd. All rights reserved.

The license management portion of this program is based on:

Reprise License Manager (RLM) Copyright 2006-2009, Reprise Software, Inc. All rights reserved

CMOST, CMG, and Computer Modelling Group are registered trademarks of Computer Modelling Group Ltd. All other trademarks are the property of their respective owners.

Computer Modelling Group Ltd.

Office #150, 3553 - 31 Street N.W.

Calgary, Alberta Canada T2L 2K7

Tel: (403) 531-1300 Fax: (403) 289-8502 E-mail: [email protected]

CMOST Studio: Getting Started Contents i

Contents

Introduction 1 What is CMOST? .......................................................................................................... 1 CMOST Tasks .............................................................................................................. 1

Sensitivity Analysis (SA) ................................................................................. 2 History Matching (HM) ................................................................................... 2 Optimization (OP) ............................................................................................ 2 Uncertainty Assessment (UA) ......................................................................... 2

Work Flow .................................................................................................................... 3

Base Files 9 Base Dataset .................................................................................................................. 9 Base IRF ....................................................................................................................... 9 Base Session File .......................................................................................................... 9 Field History File .......................................................................................................... 9 Master Dataset ............................................................................................................ 10

CMOST Formulas.......................................................................................... 11 CMOST Formula Samples ............................................................................. 11 Using Include Files ........................................................................................ 12

CMOST Tasks 13 Creating a Task File .................................................................................................... 13 Saving a Task File ....................................................................................................... 14 General Properties ....................................................................................................... 15 Parameters ................................................................................................................... 18

Adding a New Parameter ............................................................................... 18 Task Readiness .............................................................................................. 20 Validate Task File Menu Item ....................................................................... 21 Deleting a Parameter ...................................................................................... 26 Moving Parameters in Table .......................................................................... 26 Editing a Master Dataset ................................................................................ 26 Importing Parameters from the Master Dataset ............................................. 27

ii Contents CMOST Studio: Getting Started

Objective Functions .................................................................................................... 27 Global Objective Function ............................................................................ 28 Local Objective Functions ............................................................................. 29 Task Readiness .............................................................................................. 31 Validate Task File Menu Item ....................................................................... 32 Treeview Icons .............................................................................................. 32 Experimental Design Errors .......................................................................... 33 Objective Function Terms ............................................................................. 34

Experimental Design .................................................................................................. 38 Experimental Design Job Patterns ................................................................. 39 Verification Job Patterns ............................................................................... 40 Designs .......................................................................................................... 40 Design Details ............................................................................................... 41

Influence Matrix ......................................................................................................... 42 Editing the Influence Matrix ......................................................................... 44

Result Observers ......................................................................................................... 44 Adding a New Result Observer ..................................................................... 45 Importing Result Observers ........................................................................... 47 Removing a Result Observer ......................................................................... 49

Constraints .................................................................................................................. 49 Hard Constraints ............................................................................................ 50 Soft Constraints ............................................................................................. 50

Restart Configurations ................................................................................................ 51 Simulation Auto Restart ................................................................................ 52 CMOST Result Files Reuse .......................................................................... 52 CMOST Restart Configuration ..................................................................... 53

Run Configurations .................................................................................................... 55 Scheduler Configuration................................................................................ 55 Simulator Configuration ................................................................................ 58 CMOST Engine Configuration ..................................................................... 59 Job Removal from Launcher Job List View .................................................. 62 Starting a Task ............................................................................................... 62

CMOST Results 65 Run Status Monitor .................................................................................................... 65

CMOST Engine ............................................................................................. 66 Run Display ................................................................................................... 67 Messages Tab ................................................................................................ 67 Schedulers Tab .............................................................................................. 68

CMOST Studio: Getting Started Contents iii

Job Summary .............................................................................................................. 69 Job Summary Table ....................................................................................... 69 Table Configuration ....................................................................................... 71 Loading Jobs with other CMG Software ....................................................... 72 Viewing Log Files ......................................................................................... 73

Parameter Status ......................................................................................................... 73 Run Progress ............................................................................................................... 74

Graph ............................................................................................................. 74 Job Details ..................................................................................................... 75 Table .............................................................................................................. 75 Viewing Multiple Plots .................................................................................. 76

Results Observers ....................................................................................................... 77 Time Series Results Observers ...................................................................... 78 Fixed Date Results Observers ........................................................................ 79 Job Details ..................................................................................................... 80 Viewing Multiple Plots .................................................................................. 81

Parameter Effective Estimates .................................................................................... 81 Effective Estimate Table ................................................................................ 81 Tornado Plots ................................................................................................. 82

Response Surfaces ...................................................................................................... 86 Monte Carlo PDF ........................................................................................................ 87

Unconditional PDF ........................................................................................ 88 Conditional PDF ............................................................................................ 88 CDF Plots ....................................................................................................... 89 Tables ............................................................................................................. 90

Parameter Histograms ................................................................................................. 91 CMT in Results Section .............................................................................................. 92

General Operations 93 CMOST Host Manager ............................................................................................... 93

Starting the CMOST Host Service ................................................................. 94 Opening a Running CMR from Host Manager .............................................. 94

Opening a CMOST File .............................................................................................. 95 Operation Menu .......................................................................................................... 96

Task File Operations ...................................................................................... 96 Results File Operations .................................................................................. 96 Copying and Saving Images .......................................................................... 97 Copy ............................................................................................................... 97 Save Image ..................................................................................................... 98

iv Contents CMOST Studio: Getting Started

Validating Errors ...................................................................................................... 100 Task Readiness ............................................................................................ 100 Validate Task File Menu Item ..................................................................... 100 Treeview Icons ............................................................................................ 101 Experimental Design Errors ........................................................................ 101

CMOST Formula Editor ........................................................................................... 102 Functions ..................................................................................................... 102 Variables ...................................................................................................... 102 Text Size ...................................................................................................... 103

Printing Plots ............................................................................................................ 103 Page Setup ................................................................................................... 103 Print Preview ............................................................................................... 104 Print Setup ................................................................................................... 105

Exporting a Task File ............................................................................................... 106 Closing and Exiting .................................................................................................. 107

Close ............................................................................................................ 107 Exit .............................................................................................................. 108

Plot Properties .......................................................................................................... 108 Axis Tab ...................................................................................................... 108 Curve Tab .................................................................................................... 109 Legend Tab .................................................................................................. 110

Zooming for Plots ..................................................................................................... 111 How to Zoom .............................................................................................. 111 How to Remove Zooming ........................................................................... 112

Configuration Menu ................................................................................................. 114 Options ........................................................................................................ 114 Password ...................................................................................................... 116

Troubleshooting 117 Creating and Editing a CMOST Task ...................................................................... 117 Running a CMOST Task .......................................................................................... 118 Viewing CMOST Result .......................................................................................... 121

Appendices 123 Appendix A: CMOST Formulas .............................................................................. 123

Parts of a Formula ....................................................................................... 123 Constants in Formulas ................................................................................. 123 Functions in Formulas ................................................................................. 123 Variables in Formulas ................................................................................. 124 Operators in Formulas ................................................................................. 125 Formula Calculation Order .......................................................................... 125 List of Built-in Functions in CMOST ......................................................... 126

CMOST Studio: Getting Started Contents v

Appendix B: Names in CMOST ............................................................................... 133 Appendix C: Optimizers ........................................................................................... 133

CMG DECE Optimizer ................................................................................ 133 Particle Swarm Optimizer ............................................................................ 134 Brute Force Search ....................................................................................... 136 Random Search ............................................................................................ 136

Appendix D: Probability Density Functions ............................................................. 136 Uniform Distribution ................................................................................... 136 Triangle Distribution .................................................................................... 137 Normal Distribution ..................................................................................... 137 Log-normal Distribution .............................................................................. 137

Appendix E: Objective Functions ............................................................................. 137 History Match Error ..................................................................................... 137 Discounted Value ......................................................................................... 139 Raw Simulation Result ................................................................................ 140

Appendix F: Experimental Design ............................................................................ 140 Sensitivity Analysis Experimental Designs ................................................. 141 Uncertainty Assessment Experimental Designs .......................................... 142

Appendix G: Post Processing.................................................................................... 142 Sensitivity Analysis Post Processing ........................................................... 142 Uncertainty Assessment Post Processing ..................................................... 143

CMOST Studio: Getting Started Introduction 1

Introduction

This guide gives a quick overview of how to use CMOST and it is a useful reference for someone who has never worked with CMOST. Some basic knowledge of other CMGL products is assumed.

What is CMOST? CMOST is CMGs history matching, optimization, sensitivity analysis, and uncertainty assessment tool. It can launch multiple simulation jobs to multiple computers using different input parameter values. It may be used in any situation where a user runs multiple simulation jobs with the intention of either converging on a better solution to some problem or seeing the effect of input parameter changes on output properties.

CMOST can make full use of all available computers and licenses. Once a job has been created by CMOST, it will automatically submit simulation jobs and check their status periodically. Once simulations have completed, CMOST will automatically process the results. It will then visualize the results in ways that will provide insight to the problem.

To make the best use of CMOST, the user should have a good understanding of the reservoir model they are working with. A rough understanding of which parameters need to be adjusted and the possible outcomes of changing each parameter should be known. The user should also clearly know the specific goal of the project.

CMOST Tasks CMOST Studio can be used for completing four different task types:

Sensitivity Analysis History Matching Optimization Uncertainty Assessment

2 Introduction CMOST Studio: Getting Started

Sensitivity Analysis (SA) A Sensitivity Analysis Task is used for determining the overall variation of simulation results under different parameter values and/or which parameters have the greatest effect on simulation results. A sensitivity analysis normally uses a small number of simulation runs to determine how sensitive results are to different adjustable parameters. This will help the user to achieve a better understanding of how different parameters affect results. The information can later be used in other tasks such as History Matching, Optimization, or Uncertainty Assessment which make use of greater numbers of simulation runs, as it will help determine which parameters should be varied and their approximate ranges.

History Matching (HM) History matching with CMOST provides an effective and efficient way to match simulation results to production history. CMOST can automatically create multiple derived simulation datasets from the Master Dataset by varying selected dataset parameters and then run the simulation jobs. As jobs complete, CMOST will analyze the results to determine how well they match the production history. An optimizer will then be used to determine parameter values for new simulation jobs. As more simulations complete, the results will converge to multiple optimal solutions which provide satisfactory history match if user specified parameters and their ranges are appropriate.

Optimization (OP) An optimization task is used to identify optimal field development plan and operating conditions that will produce either a maximum or minimum value for objective functions the user specifies. These objective functions may be physical quantities, such as cumulative oil produced, recovery factor, and cumulative steam oil ratio. More importantly, CMOST allows that monetary values can be assigned to different physical quantities. Thus optimization can be conducted using simplified net present value.

Uncertainty Assessment (UA) Uncertainty Assessment is normally used to determine the likely variation in simulation results due to uncertainty usually uncertainty in reservoir variables. It makes use of some simulation to develop a Response Surface (RS) for each objective function (such as NPV, CSOR, cumulative oil production, etc.) of interest with respect to each of the uncertain variables (e.g. porosity, permeability, endpoint saturations, oil viscosity, etc.), and then conducts a Monte Carlo simulation by selecting tens of thousands of variable value combinations and determining the value of the objective function for each combination. The results are probability density function (PDF) and cumulative density function (CDF) plots for each objective function.


Uncertainty Assessment can be used to obtain a sensitivity analysis and an optimization in certain circumstances, and has certain advantages over SA and Optimization as well as certain limitations.

The advantages are that: it uses three values for each variable instead of the two required by SA; It results in Response Surfaces that map the change in objective functions with respect to changes in the variables, which is a considerable improvement over SA; It provides a conditional PDF for each objective function with respect to each variable which gives good indication on the sensitivity of each variable; It allows some optimization of variables to be carried out, which cannot be done by SA; And it requires many fewer simulation cases to be run than does Optimization.

The limitations mainly apply to the Optimization component in that the RSs are a less accurate way to obtain an optimum solution than full Optimization, and are potentially limited to interpolation (rather than extrapolation) among the variable values. Also the number of variables, and variable values, which can be assessed, is more limited than would be the case with full Optimization and Sensitivity Analysis.

Before any of these tasks are started, some basic information is required. This is discussed in the Base Files section.

Work Flow CMOST tasks can be performed independently in CMOST Studio. It is considered as a good practice to conduct a Sensitivity Analysis prior to History Matching, Optimization and Uncertainty Assessment if the reservoir model is complex and you are not sure which parameters need to be adjusted. Since the tasks can be performed independently, a user may also wish to redo a task such as a refined history matching after an initial history matching is complete. Before starting to work with CMOST, it is worthwhile to know the main components of CMOST as shown in the figure below.


The general steps for each task type are as follows:


Create a base dataset file & corresponding IRF fileBase.DAT and Base.IRF

Create a CMOST Master Dataset (CMM file) for indicating where and how CMOST should modify the dataset. Original dataset can be copied and manually edited.

Base.CMM

Create a new Uncertainty Assessment Task in CMOST Studio. Enter parameters into the parameters section. Parameters that are marked in the CMM file can be imported directly.

Normally, the middle value should be the mean and the High and Low values should be the highest/lowest practically possible values. Enter objective function information. Choose an

Experimental Design and Results Observers. Save the new task file. Uncertainty.CMT

Use Unconditional PDF plots to assess uncertainties of production forecast results. Use conditional PDF plots to understand the effect (sensitivity) of each uncertain parameter on

forecast results.

Uncertainty Assessment

(Estimate reliability of the predictions)

Present your findings to management

Run the task. CMOST Result file:Uncertainty.CMR

Is generated along with a number of SR2 files specified in the Experimental Design step. In the Results, look at the Response Surface plots. The Response Surfaces are trained using SR2

files of verification runs. A few additional runs are then performed and plotted against the Response Surface plots to see if they match the predicted objective function values using the

Response Surfaces. If these additional values are along the line of the plot, then the predicted values are very close to the actual simulation values. This is an indicator of the viability of a

Response Surface and subsequent Monte Carlo simulation results.

CMOST Studio: Getting Started Base Files 9

Base Files

Each CMOST Task is based on a previously completed simulation. CMOST needs access to a number of files from this base simulation, and can make use of other files if they are available. The types of files that can be referenced are mentioned below.

Base Dataset Before beginning to work with CMOST Studio (CMOST), a Base Dataset should be created. The Base Dataset can be any valid dataset for any CMG simulator. This file is not directly used by CMOST but will give a base state for which different variations can be made. See the Master Dataset and Base IRF sections.

Base IRF The Base Dataset should be used to create the Base IRF (simulation results file). This file gives CMOST basic information about the dataset, such as the type of simulator that was used, well lists, simulator start and end dates, etc. If the corresponding MRF file is located with the base IRF to complete the SR2 pair, then the base information from these files will be shown in the CMOST Results plots. The base IRF is a required component.

Base Session File A base session file is a useful file to have but is not required by CMOST. The base session file is a file created with Results Graph using the base IRF. CMOST can later use the session file to quickly display different plots in Results Graph with simulation runs created by CMOST.

Field History File A field history file contains production data for the reservoir. This file can be created using Builder or by following the steps outlined in the Results Graph Manual. Field History Files are only required for History Matching Tasks.

10 Base Files CMOST Studio: Getting Started

CMOST treats all the data points in a field history file equally. Therefore, the temporal distribution of data in the field history file will have certain effect on the final history match results. If users do not pay attention to this, CMOST could converge to results that are not desired. For example, the goal of a history match problem is to match the overall trend of Cumulative Oil SC of 5-years history (2000-01-01 to 2005-01-01). Ideally, the Cumulative Oil SC (or Oil Rate SC) data in the field history file should be equally distributed in time (daily, weekly, or monthly). In this way, CMOST will try to match the trend of Cumulative Oil SC during the entire five years. If the field history file has daily Cumulative Oil SC data for the year of 2001 and monthly data for the remaining four years, the simulation results of 2001 (365 data points) will contribute more to the history match error than those of 2002-2004 (48 data points). CMOST may converge to a match that is perfect for the year of 2001 but very bad for the years of 2002-2004.

NOTE: Current version of CMOST only supports FHF using the following time units: days, day, ISO_DATE_FORMAT, YYYY/MM/DD, YYYY MM.

Master Dataset The Master Dataset is a version of the Base Dataset that has been modified to tell CMOST where to enter different values into the dataset. Parameters in the dataset that are of interest to the user have their values substituted with CMOST formulas allowing CMOST to substitute different values at runtime. For example, the following line is used to set water-oil contact depth in a STARS base dataset:

*DWOC 80.0

In a Master Dataset, water-oil contact depth can be specified by:

*DWOC this[80.0]=WaterOilContact OR *DWOC WaterOilContact

WaterOilContact is a parameter (independent variable) which will be defined in the corresponding CMOST Task File created in CMOST Studio.

Anywhere CMOST is required to substitute a value or text into the Master Dataset, a CMOST formula should be entered. CMOST formulas can appear anywhere in the Master Dataset. However, each CMOST formula must be completed in a single line.

The Master Dataset is a required component.

NOTE 1: The first and last date/time keywords for a Master Dataset must be *DATE. Errors will be encountered if *TIME keywords are used.

NOTE 2: Current version of CMOST requires that simulator output time unit (OUTUNIT) must be day(s). Otherwise, Errors will be encountered.

CMOST Studio: Getting Started Base Files 11

NOTE 3: It is recommended that master datasets DO NOT contain the action STOP in well operating and monitoring constraints to force simulation to terminate. CMOST will ignore simulation results that are terminated before the expected stop time is reached.

CMOST Formulas All CMOST formulas that are entered into the Master Dataset must be nested within a start tag and an end tag . It is optional but recommended to use this[OriginalValue]= to start a formula. The original value indicates the value that was used by the base dataset. Also, if the syntax this[OriginalValue]= is used, the variable this can be used in the formula to reference the original value that was entered in the dataset.

CMOST formula syntax and description of built in functions is provided in Appendix A: CMOST Formulas.

CMOST Formula Samples this[1.0]=varAvarA

Substitutes the value of parameter varA directly into the dataset this[1]= this + varB

Substitutes the result of adding parameter varB to the original value this[26]=varA-varB+5varA-varB+5

Substitutes the result of varA minus varB plus 5 this[203.9]=179.79*POWER(varA/varB, 0.248)

Substitutes the result of 248.0

BvarAvar79.179

this[1800.0]=MAX(varA*varB, 1200)If the result of varA multiplied by varB is greater than 1200, the multiplied parameters will be substituted. Otherwise, 1200 will be substituted.

this[OPEN]=IF(varA>=600, OPEN, CLOSED)

If varA is greater than or equal to 600, OPEN will be substituted. Otherwise, CLOSED will be substituted.

this[402.57]=LOOKUP(varB, {3.0, 5.0, 7.3}, {524.62, 402.57, 188.75})

12 Base Files CMOST Studio: Getting Started

If varB matches any of the values in the first set of values, the corresponding value in the second set of values will be substituted. For example, if varB was set to 7.3, the value 188.75 would be substituted.

this[permMid.inc]=LOOKUP(varA, {porLow.inc", "porMid.inc", "porHigh.inc"}, {"permLow.inc", "permMid.inc", "permHigh.inc"})

If varA matches any of the entries in the first set, the corresponding entry in the second set will be substituted. For example, if varA was set to porMid.inc, permMid.inc would be substituted into the dataset.

this[0.68]=MAX(MIN(varA, 1), 0)A range of acceptable values is set for varA. If varA is less than 0, the value 0 will be substituted. If varA is greater than 1, the value 1 will be substituted. Otherwise, if varA is between 0 and 1, the value of varA would be substituted.

Using Include Files If large arrays of data need to be substituted, it may be easier to use include files. For example, porosity may have a value for each grid block in a reservoir. It would be unrealistic to create a parameter for each grid blocks porosity. Multiple include files can be used where each one can contain a different geostatistical realization for porosity.

Include files can be used anywhere in a dataset. The syntax that is used for entering an include file into a master dataset is:

*INCLUDE ArrayIncFile The parameter ArrayIncFile would then be defined as a Text on the Parameters page with the include files listed as candidate values.

The include files would contain the block of text that is to be substituted into the dataset. For example, porosity could be used in an include file as follows for a reservoir with dimensions: ni = 10, nj = 3, nk = 2

*POR *ALL .08 .08 .081 .09 .12 .15 .09 .097 .087 .011 .15 .134 .08 .087 .157 .145 .12 .135 .18 .092 .074 .12 .12 .154 .167 .187 .121 .122 .08 .08 .095 .13 .12 .157 .17 .18 .184 .122 .084 .09 .11 .12 .134 .157 .157 .18 .18 .098 .09 .09 .08 .09 .144 .143 .123 .16 .165 .102 .10 .10

Other include files would be created with similar syntax but with different values entered.

CMOST Studio: Getting Started CMOST Tasks 13

CMOST Tasks

Creating a Task File Once all the base files have been created, a new task file can be started.

When CMOST Studio is opened, a blank window will appear. To create a new task file, select New from the File menu. The following dialog will appear:

Task Type According to the type of task that is wished to be performed, select the appropriate Task Type. See CMOST Tasks for more details.

Base IRF File Name The Base IRF contains basic information about the datasets that CMOST will create, such as the type of simulator that was used, well lists, simulator start and end dates, etc. See Base IRF for more details.

14 CMOST Tasks CMOST Studio: Getting Started

The file path can be entered manually or the Browse button can be used to search for the file.

Task File Folder The Task File Folder is the directory where all files related to the CMOST Task will be located. If the Task File Folder is going to be the same directory as directory where the base IRF is currently located, check the The same as the Base IRF File checkbox. If not, uncheck the box and browse for the directory using the Folder button. If this option is chosen, the base IRF file will be copied to the Task File Folder.

NOTE: If CMOST will be scheduling jobs remotely, the Task File Folder must be in a UNC directory.

When all the information is filled in, click OK.

Saving a Task File At any point during the progress of a task, the CMOST task file can be saved. To save, click the File menu and select Save or Save As. The following dialog will appear:


The first time a file is saved, the default file name is newtask.cmt. However, the file can be named anything. The file should be saved to the Task File Folder chosen when creating the new task file. If it is not, all related files will be copied to the new file location. When the correct location and a file name have been chosen, click Save. The new task file will be saved.

If the file needs to be saved later after changes have been made, select either Save As through the File menu to save the file with the same or new name, or select Save to overwrite the old file with any changes that have been made.

General Properties Once a new task is created, the studio will bring up the General Properties page of the task. The General Properties page is where the base files can be entered for the task. This page is also used for error checking using the Task Readiness display.

Master Dataset The Master Dataset is a version of the Base Dataset that has been modified to tell CMOST where to enter different values into the dataset. The file path can either be entered manually or found using the Browse button. If the file is not in the Task File Folder, CMOST will copy the file to the folder automatically. The Master Dataset is a required component.


The Edit button can be used to open up the master dataset in a text editor. If there is no text editor associated with the master dataset file type (CMM), a dialog will appear to set the file type association.

See the Master Dataset section for more information about the file.

Base Dataset The Base Dataset is entered only for future reference. This is not a required component.

The Builder button can be used to open the dataset in Builder for editing or to visualize the dataset.

See the Base Dataset section for more information about the file.

NOTE: Changes made in the Base Dataset will not be reflected in the Master Dataset or vice versa. Each must be edited separately.

Base Session File The Base Session File is not required but is very useful for quickly analyzing simulation results. See the Base Session File section for details.

IRF file from which origin information was imported: The location of the Base IRF is listed in this section. The Base IRF is only required when creating the Task File. See the Base IRF section for more information about the Base IRF File.

If changes in the Master Dataset are made that were not included in the Base IRF when the task file was created, a new IRF file may need to be imported to update different sections of the Task File.

For example, if a new well is added to the Master Dataset after the task file is created, a new IRF needs to be imported if the user wishes for CMOST to use results from the new well. To do this, the Base Dataset will need to be updated to contain the new well and then be run with a simulator. The new IRF file can then be imported by using the Import button in the Optional Files section on the General Properties page.

Task Readiness The Task Readiness display shows if there are any errors on the pages listed. As tasks are completed, the Xs will turn to checkmarks if there are no errors. If errors are encountered, the cell can be expanded to give a description of the error. A task cannot be started unless all errors have been fixed.

If there are any warnings, they will appear as exclamation marks. The description of the warning can also be viewed by expanding the cell with the warning. A task can be started if there are warnings, but it is recommended that all warnings are examined before starting.

See Validating Errors for more information on checking errors.


Task Name The Task Name contains the full path name of the task file that is currently open. This information may not appear until the file is saved and reopened. The file name can also be found in the title bar of the studio.

Task Type The Task Type that is used for the current task file is listed here. See CMOST Tasks for more information on the different task types.

Unit System The unit system is the set of units that will be displayed in the plots in the CMOST Results file. If the unit system is different from the units in the Master Dataset then the units will be converted for the plots.

Description The description box is not used by CMOST but can be used to note any details that may be important for future reference.


Parameters The Parameters page can be found by selecting Parameters in the tree view. The Parameters page is where different values are entered for each parameter that will later be used to substitute into the Master Dataset. Each parameter has a number of properties associated with it.

Adding a New Parameter A new parameter can be manually entered in two ways. Information can be entered into the bottom row of the Parameters Table or the Insert button can be used to insert a new row into the table where information can then be added.

CMOST can also automatically copy all parameters that are present in the Master Dataset to the Task file. To do this, click on the Import button. If the Master Dataset is very large, it may take a while for CMOST to read and find all parameters. See Importing Parameters from the Master Dataset for more information on how to import parameters from Master Dataset.


Name The name of the CMOST Parameter listed in the Master Dataset should be entered into the Name column. The name entered in the Parameters Table must match the name of the parameter in the Master Dataset exactly. Names are case sensitive. See Appendix B: Names in CMOST for more information about using names in CMOST.

Default Value The default value should produce the original value that is entered in the Base Dataset. This value is only used when the Active checkbox is not checked for a parameter. To edit a default value in the studio, click on the cell and enter in the new value. If a more complicated CMOST Formula is used, the default value may not be equal to the original value in the dataset. For example, if the following CMOST Formula was entered into the Master Dataset:

this[1]=LOG10(Keq) The default value for the parameter Keq would be 10 since this value would produce the original value of 1 when the CMOST Formula is evaluated.

Active The active checkbox determines whether or not CMOST will vary specific parameter when substituting into the Master Dataset. If the Active checkbox is checked, CMOST will assign a Candidate Value to the parameter. If the box is unchecked, CMOST will assign the Default Value to the parameter for every run that is created.

Type There are three different data types that the parameters can be defined as:

Real Integer Text

Depending on the simulator keyword that the parameter is being used with, different types should be chosen.

Real should be chosen for parameters that can have decimal values such as parameters representing porosity or permeability. This is the default type.

Integer should be chosen for a parameter that cannot have decimal values such as parameters representing rock type or a block location.

Text should be used for parameters that have text values. The values should always be enclosed in double quotation marks (for example OPEN). The Import option will automatically assign the text type to any parameter which has a default value enclosed in double quotation marks in the Master Dataset.


Source Source is used to tell CMOST how to assign values to a parameter. There are two different types of sources for values that CMOST will use for a parameter:

Candidate Values List Formula

The Candidate Value List refers to the items entered into the Candidate Value table located on the bottom part of the studio. If this option is selected, values will be taken from this list to assign to the parameter. This is the default.

Formulas can be entered for a parameter. The section for entering the formula will appear if the source is changed to Formula for a parameter. Any other parameter can be used in the formula, as well as any of the CMOST functions. See Validating Errors.

There are a variety of different ways to check errors in CMOST Studio. It is important to check for errors and warnings before running a task since a task cannot be started unless all errors have been fixed. A task can be started if there are warnings, but it is recommended that all warnings are examined before starting.

Task Readiness The main way to check errors in CMOST Studio is to use the Task Readiness display on the General Properties page.

The Task Readiness display shows if there are any errors on the pages listed. As pages are completed, the Xs will turn to checkmarks if there are no errors. If errors are encountered, the pages cell can be expanded to give a description of the error.

If there are any warnings, they will appear as exclamation marks. The description of the warning can also be viewed by expanding the cell with the warning.


Validate Task File Menu Item Another way to check for errors or warnings is to access the Validate Task File menu item.

This can be found under the Tools menu or by clicking on the Validate icon in the Tool bar. When this item is selected, it will give a text output of errors or warnings found on each page

If no errors are encountered, the status will be shown as Ready.

Treeview Icons If there are errors with the task files, an error or warning display will appear on the treeview icon where the problem was encountered.

A description of the error can be found by right clicking on the icon then selecting Show Page Status. A similar display to the Validate Task File dialog will appear.


Experimental Design Errors Specific errors relating to experimental design can be found on the Experimental Design page in the Error Messages display. This field is updated when first entering the Experimental Design page or when the Check Error button is clicked.

CMOST Formula Editor for more details on entering formulas for parameters.

Track The Track checkbox determines whether or not CMOST will show the value of the parameter on the Job Summary page of the CMOST results file. If the Track checkbox is checked, CMOST will show the parameter data in the Job Summary table. If the box is unchecked, the data will not be shown.

Candidate Values The Candidate Value table is where values are entered for CMOST to substitute into the Master Dataset. This section will differ depending on the Task Type that is chosen.

The specific candidate value that is chosen for a run is different for each Task Type. Sensitivity Analysis and Uncertainty Assessment values for each run are determined by the Experimental Design. History Match and Optimization values for each run are determined by the Optimization Method that is used.

Sensitivity Analysis For a Sensitivity Analysis Task, this section will be labeled as Sample Value. Two different values will be required to be entered for each parameter using Candidate Value Table as the source. The values entered for low and high should be within the reasonable range for the property represented by the parameter. The low and high values should be different.

History Match and Optimization An unlimited number of entries can be added to the Candidate Value table for History Match and Optimization tasks. However, it should be noted that it will take longer and longer for the optimizer to arrive at a solution for the task as more candidate values are added. The values added should represent a range of acceptable values that can be substituted for a certain parameter.

Candidate Value entries can be removed by clicking on the table entry number (the grey cell next to entry that needs to be removed), then using the delete key on the keyboard.

For each parameter, candidate values must be arranged in either ascending or descending order.


Uncertainty Assessment For an Uncertainty Assessment Task, this section will be labeled as Dataset Value. Three different values will be required to be entered for each parameter using Candidate Value Table as the source. The value entered for low should represent a value near the lower limit of practical values for that parameter. High should represent a value that is near the upper limit of practical values for that parameter. Middle should be a value somewhere in the middle of that range.

For all task types except Sensitivity Analysis, if the parameter type is Text, an equivalent numerical value will have to be added with the candidate value. If there is a value that fits with the text value, that number should be entered.

Copying Candidate Values If candidate values for several parameters are the same, the values can be copied from one parameter to another parameter or group of parameters. To do this, right click on the parameter that the values are to be copied from then select Copy Candidate Values to Other Parameters.

This will bring up the Candidate Value Copying dialog:

For Sensitivity Analysis and Uncertainty Assessment tasks, all candidate values will have to be copied. For History Matching and Optimization tasks, the different candidate values can be chosen to be copied by using the checkboxes in the Candidate Value table.


The Parameters table will be filled with all parameters that are the same Type as the candidate value that is being copied. The parameters that the candidate values will be copied to can be chosen by checking the corresponding checkbox for each parameter.

NOTE: If the parameters that the values are being copied to already have values entered, the preexisting values will be replaced by the new copied values. Values will NOT be appended.

File Button If a parameter is selected to represent an include file in the Master Dataset, the File button can be used to browse for the files that will be included. The parameter type must be selected as Text for this button to be activated. Only one file may be selected at a time to add to the task file. See Using Include Files for more information on how to use include files with a Master Dataset.

Generate Button The Generate button can be used to quickly populate the Candidate Value table with a range of values. This button is only available for History Matching and Optimization tasks. The parameter type must be either Real or Integer for this button to be active.

Once the button has been clicked, the Value Generator dialog will appear:

There are three different methods available for generating values:


Arithmetic Sequence Geometric Sequence Sampling from Distribution (Not recommended unless there is a specific reason to

use this method.)

For each of the methods, the number of values that should be entered into the Task File is determined using the Number of Values box.

NOTE: It will take longer and longer for the optimizer to arrive at a solution for the task as more candidate values are added.

The Arithmetic Sequence produces values that are evenly spaced between them (for example, 0.25, 0.50, 0.75, 1.00, 1.25, and 1.50). The minimum and maximum values can be any value.

The Geometric Sequence creates a sequence of values that follow a pattern of multiplying a fixed number from one term to the next (for example, 0.25, 0.5, 1, 2, and 4). The minimum and maximum values can be any non-zero value and both values must be the same sign. (i.e. Minimum*Maximum > 0)

The Sampling from Distribution method creates a random sampling of values based on the distribution that is entered. There are five types of distributions available for sampling:

Normal Uniform Triangle Lognormal Universal

See Appendix D: Probability Density Functions for more information on each of the distribution types.

Once the PDF has been defined, click Random Sample to randomly choose values using that distribution. This can be repeated several times if desired.

After all of the information is filled out for the Value Generator dialog, click OK. The values will be transferred to the parameter.

NOTE: If duplicate values are created by the random sample, they must be removed before continuing. To do this, select the item number of a duplicate value in the Preview Values table then use the Delete key on the keyboard.


Prior Probability Density Functions Prior Probability Density Functions are only available for Uncertainty Assessment tasks. The distribution given in this section should represent the probability that specified values can occur. There are several different distribution types available for CMOST:

Normal Fixed Uniform Triangle Lognormal Universal

More information for each of the Prior Probability Density Functions is provided in the Appendix D: Probability Density Functions section.

All of the associated Prior Probability Density Function data entries must be filled out for each parameter.

Deleting a Parameter To delete a parameter, the entire row of the parameter must be selected. To do this, click on the grey cell to the left of the parameter name. The entire row will then become highlighted and the Delete Button will become enabled. Click on the Delete button to delete the row. The Delete key on the keyboard can also be used.

Multiple Parameters can be removed by clicking on the grey cell to the left of a Parameters Name and dragging the mouse up or down. Shift and Ctrl functionality is also available. The Delete button or Delete key on the keyboard can then be used to delete the rows.

Moving Parameters in Table To move a Parameter up or down in the Parameters Table, select any cell in the row of the parameter that needs to be moved, then click on the Up or Down arrow buttons. Parameters may be listed in any order.

Editing a Master Dataset The Master Dataset and Task File can be accessed from the Parameters page to view or to edit with other programs. The files can either be opened in Builder or with a text editor. To open the Task File in Builder, click on the Builder button. This will open the Task File in Builder with the CMOST Project dialog opened (This feature will be available for 2009 Builder release). To open the Master Dataset in a text editor, click on the Edit button.


NOTE: If there is no text editor associated with the master dataset file type (CMM), a dialog will appear to set the file type association.

Importing Parameters from the Master Dataset CMOST can automatically copy all parameters that are present in the Master Dataset to the Task file. To do this, click on the Import button (It may take a while for CMOST to read large Master Dataset files). If the this[OriginalValue] syntax is used in the Master Dataset, the Default Value will also be copied from the file. CMOST will assume that the Default Value is equal to the original value when importing parameters from the Master Dataset. The imported default values should be checked since this may not always be the case.

The parameter Type should also be checked for any errors. When importing, CMOST will set any parameters that have original values with quotation marks to Text. All other parameters will be set to Real. Parameters that should be Integers will have to be changed manually after importing.

Active and Track checkboxes are automatically checked when importing parameters. The Source will be set to Candidate Value List by default.

Candidate Values and Prior Probability Density Functions will still have to be entered after the parameters have been imported.

Objective Functions Objective functions are what CMOST uses to analyze the data that is retrieved from the simulation output files. For a History Matching or Optimization task, a global objective function is required. For all task types, at least one local objective function is required. Each local objective function must contain at least one objective function term.

For History Matching and Optimization, both global objective function and local objective functions are used to rank results of simulation jobs and determine how to modify parameter values to improve simulation results.

For Sensitivity Analysis and Uncertainty Assessment, local objective functions are the responses you want to investigate for a simulation model by changing parameter values.


Global Objective Function Global Objective functions are only used for History Matching and Optimization tasks.

Name The Global Objective Function can be changed to any acceptable name to give it a better description. See Appendix B: Names in CMOST for more information about using names in CMOST.

Method For a History Matching task, the global objective function is always the weighted average of all the Local Objective functions. For an Optimization Task, the Global Objective Function can be determined using one of three different methods:

Sum (Recommended) Weighted Average Use the First


The Sum method adds all objective functions together to produce a Global Objective Function. The weighted average will average all values using the Weight set for each Local Objective Function.

Error! Objects cannot be created from editing field codes.

Here, LOFi is local objective function; wi is weight.

If Use the First is selected, the Global Objective Function will be equal to the first Local Objective Function listed in the Local Objective Functions Table.

Local Objective Functions Local Objective Functions are used in all task types. There are three types of local objective functions:

History Match Error (For History Matching only) Discounted Value (Optimization, Sensitivity Analysis, Uncertainty Assessment) Raw Simulation Result (Optimization, Sensitivity Analysis, Uncertainty Assessment) See Appendix E: Objective Functions for more information about how these objective functions are calculated.

Adding a New Local Objective Function To add a new Local Objective Function, click on the Insert button. For each task type except for History Matching, a menu will appear. There will be two options for creating a Local Objective Function:

Discounted Value Raw Simulation Result

Local Objective Functions using Raw Simulation Results will analyze data by taking simulation results without any alterations. If Discounted Value is selected, simulation results will be transformed using the discounting function. This option should be used if you want to calculate Net Present Value using simulation results. See Discounted Value for more information about how data is transformed.


Name The name of the Local Objective Function should be entered into the Name column. Names for each Local Objective Function must be unique. Do not use the same name for more than one Objective Function. See See Appendix B: Names in CMOST for more information about using names in CMOST

Active The active checkbox determines whether or not CMOST will analyze data using that specific Local Objective Function. If the Active checkbox is checked, CMOST will use the function. Otherwise, the Local Objective Function will not be used.

Weight The Weight will give local objective functions more or less emphasis in the Global Objective function if Weighted Average is selected for the Calculation Method. The higher the Weight relative to the other Local Objective Functions, the more emphasis the objective function will have on the Global Objective Function.

This value has no effect on Sensitivity Analysis or Uncertainty Assessment tasks.

Category The Category lists the type of local objective function. This will be the same as the option that was chosen when adding the new Local Objective Function. If the task is History Matching, the Category will be listed as History Match Error. See Appendix E: Objective Functions for more information about how each option is calculated.

Formula The Formula column lists how each Local Objective Function uses its Objective Function Terms to produce a value. For History Match Error and Discounted Value, this column is read only. For Raw Simulation Result, the formula needs to be entered manually. See Appendix E: Objective Functions for more information on how other Local Objective Function types are calculated.

NOTE: It is recommended that Objective Function Terms are entered before adding a Formula to a Raw Simulation Result Objective Function.

To enter a formula for a Raw Simulation Result Objective Function, click the Formula cell in the Local Objective Functions table. A text box and button will appear. The Formula can be typed into the text field or the button can be clicked to bring up a dialog for editing formulas.


The variables listed will be the Objective Function Terms that have been added for that Local Objective Function. See Validating Errors.

There are a variety of different ways to check errors in CMOST Studio. It is important to check for errors and warnings before running a task since a task cannot be started unless all errors have been fixed. A task can be started if there are warnings, but it is recommended that all warnings are examined before starting.

Task Readiness The main way to check errors in CMOST Studio is to use the Task Readiness display on the General Properties page.

The Task Readiness display shows if there are any errors on the pages listed. As pages are completed, the Xs will turn to checkmarks if there are no errors. If errors are encountered, the pages cell can be expanded to give a description of the error.

If there are any warnings, they will appear as exclamation marks. The description of the warning can also be viewed by expanding the cell with the warning.


Validate Task File Menu Item Another way to check for errors or warnings is to access the Validate Task File menu item.

This can be found under the Tools menu or by clicking on the Validate icon in the Tool bar. When this item is selected, it will give a text output of errors or warnings found on each page.

If no errors are encountered, the status will be shown as Ready.

Treeview Icons If there are errors with the task files, an error or warning display will appear on the treeview icon where the problem was encountered.


A description of the error can be found by right clicking on the icon then selecting Show Page Status. A similar display to the Validate Task File dialog will appear.

Experimental Design Errors Specific errors relating to experimental design can be found on the Experimental Design page in the Error Messages display. This field is updated when first entering the Experimental Design page or when the Check Error button is clicked.

CMOST Formula Editor for more information about adding formulas to Objective Functions

Units The units that should be displayed with the objective functions can be listed in this cell. For example, if the objective function represents the cumulative liquid production of a well, bbl may be added as the unit type. This item is optional and does not affect the values that are calculated.

Removing a Local Objective Function To delete a Local Objective Function, the whole row for that function must be selected. To do this, click on the grey cell to the left of the Local Objective Functions name. The whole row will become highlighted and the Delete button will become enabled. Click on the Delete button or use the Delete key on the keyboard to delete the row.

Multiple Local Objective Functions can be removed by clicking on the grey cell to the left of a Local Objective Functions Name and dragging the mouse up or down. Shift and Ctrl functionality is also available. The Delete button or Delete key on the keyboard can then be used to delete the rows.

Repeating an Objective Function Local Objective Functions can quickly be repeated if very similar Local Objective Functions need to be created. Using the Repeat button, all Local Objective Function information will be copied except for the Local Objective Function Name and Objective Function Terms Origin Name.


To repeat a Local Objective Function, the entire row for that function must be selected. To do this, click on the grey cell to the left of the Local Objective Functions name. The whole row will become highlighted and the Repeat button will become enabled.

NOTE: To be able to repeat an Objective Function, all of the Objective Function Terms for that function must have the same Origin Name.

Click on the Repeat button to bring up a dialog for selecting which Origin Names should be repeated. Shift and Ctrl functionality is available for selecting items in the list. Once all the Origin Names that should be repeated are selected click OK.

The same objective function information for all of the selected items will be entered into the table. Any of the information for the repeated Local Objective Functions may be edited.

Moving Local Objective Functions in Table To move a Local Objective Function up or down in the Local Objective Functions Table, select any cell in the row of the objective function that needs to be moved, then click on the Up or Down arrow buttons respectively. Local Objective Functions may be listed in any order.

Objective Function Terms Objective Function Term describes which data should be extracted from simulation results files for calculating objective functions.


Adding a New Objective Function Term There are two ways to add new Objective Function Terms. Information can be added to the bottom row of the Objective Function Terms Table or by clicking on the Insert button and adding information to the new row that is created.

Name The name of the Objective Function Term should be entered into the Name column. Names for each Objective Function can be the same as an Objective Function Term for a different Local Objective Function. However, the same name cannot be used for two Objective Function Terms for the same Local Objective Function. See Appendix B: Names in CMOST for more information about using names in CMOST.

Origin Type The Origin Type is the type of data that will be retrieved for the Objective Function Term. This data can come from the following types:

Wells Groups Specials Sectors Layers Leases Parameters Numerical (Not supported yet)

All Origin Types come from the simulation results file except for Parameters which retrieves parameter information from the task file.

Origin Name The Origin Name is a specific item of the Origin Type. If there are no items available for Origin Type, the Origin Name list will be empty. If this is the case, that Origin Type should not be used.

For example, if the Origin Type is WELLS, the Origin Name would contain a list of all the wells that are present in the dataset.

Property The Property is a specific item for the Origin Name. No information needs to be entered into this column if the Origin Type is SPECIALS or PARAMETERS. Information is required for all other types.


For example, if the Origin Type is WELLS, the Property cell would have a list of well properties such as Cumulative Oil SC or Gas Rate SC.

If Discounted Value is selected for the Local Objective Function, select property rates, not cumulative property amounts.

Start Date Time The Start Date Time is only available for Discounted Value and History Match Error Local Objective Functions. This date represents the date that data should start to be analyzed from. This date must be between the simulation start date and the End Date Time. The default date that is entered is the simulation start date.

End Date Time The End Date Time is only available for Discounted Value and History Match Error Local Objective Functions. This date represents the date that CMOST should stop analyzing data. This date must be between the Start Date Time and the simulation stop time. The default date that is entered is the simulation stop time.

Simulation Date Time The Simulation Date Time is only available for Raw Simulation Result Local Objective Functions. This date is the time that the data for the Objective Function Term will be taken at. This date must be between the simulation start date and simulation stop time. The default date that is entered is the simulation stop time.

NOTE: You must make sure that the required data at the specified Simulation Date Time are available in SR2 files. Improper use of WSRF keyword in Master Dataset may cause simulator to skip writing data to SR2 output files.

Yearly Interest Rate The Yearly Interest Rate is the rate that is used for discounting simulation results. This column is only available for Discounted Value Local Objective Functions. This value should be entered as a fraction, not a percent. The default value is 0.10 (10%). See Discounted Value for more information how this value is used.

Field History File This option is only available for History Match Error Local Objective Functions. This is the column where the Field History File containing the production data that matches the Property information for the Objective Function Term.

NOTE: There must be data in the field history file for each objective function term. If there is not, analysis will not be completed. Do not add terms which do not have available data.

Different Field History Files can be added for different Objective Function Terms. The ellipsis button can be used to browse for the Field History File.


Measurement Error This option is only available for History Match Error Local Objective Functions. Measurement error is used to indicate the accuracy of the production data. The measurement error for each Objective Function Term can be added to the measurement error cell. A default of 0 is entered otherwise. More information about how Measurement Error is used to calculate the history match error can be found in the History Match Error section.

Term Weight This option is only available for History Match Error Local Objective Functions. The Term Weight gives Objective Function Terms more or less emphasis in the Local Objective Function. The higher the Term Weight relative to the other terms Term Weight, the more emphasis the term will have on the Local Objective Function. Normally, higher term weight should be given to wells that are important (good production, long history, near future development wells) and difficult to match.

Unit Value This column is only available for Discounted Value Local Objective Functions. The unit value should be used primarily for adding a monetary value to a simulation result. Both positive and negative values can be entered. For example, in SAGD processes, unit value for oil rate should be positive and determined from oil price; unit value for injector water rate should be negative and determined from steam generation cost.

Conversion Factor This option is available for Discounted Value and Raw Simulation Value Local Objective Functions. The conversion factor is used to convert from one unit to another. For example, if the units in the dataset were bbl and units of Million bbl were desired, a value of 1E-6 would be entered for the conversion factor. (1 Million bbl/1,000,000 bbl)

Removing an Objective Function Term To delete an Objective Function Term, the whole row for that function must be selected. To do this, click on the grey cell to the left of the Objective Function Terms name. The whole row will become highlighted and the Delete button will become enabled. Click on the Delete button or use the Delete key on the keyboard to delete the term.

Multiple Objective Function Terms can be removed by clicking on the grey cell to the left of an Objective Function Terms Name and dragging the mouse up or down. Shift and Ctrl functionality is also available. The Delete button or Delete key on the keyboard can then be used to delete the rows.

Repeating an Objective Function Term Objective Function Terms can quickly be repeated if very similar terms need to be created. Using the Repeat button, all Objective Function information will be copied except for the Objective Function Terms Name and the Objective Function Terms Origin Name.


To repeat an Objective Function Term, the entire row for that function must be selected. To do this, click on the grey cell to the left of the Objective Function Terms name. The whole row will become highlighted and the Repeat button will become enabled.

NOTE: To be able to repeat an Objective Function Term, the Origin Type of the term cannot be set to PARAMETERS or NUMERICAL. Additional Origin Names must also be available.

Click on the Repeat button to bring up a dialog for selecting which Origin Names should be repeated. Shift and Ctrl functionality is available for selecting items in the list. Once all the Origin Names that should be repeated are selected click OK.

The same objective function term information for all of the selected items will be entered into the table. Any of the information for the repeated Objective Function Terms may be edited.

Experimental Design The Experimental Design page is where detailed information about which candidate values will be used for each run. This page is only available for Sensitivity Analysis and Uncertainty Assessment Tasks. This page should not be completed until the Parameters page has been completed.


Experimental Design Job Patterns Patterns used by CMOST are composed of a series of + and symbols. The + symbol represents the High value for a parameter and the represents the Low value for a parameter. For Uncertainty Assessment tasks, there is a third symbol: 0. The 0 represents the middle value for a parameter. The order of the pattern is the same order that parameters are listed on the Parameters page. Only Active parameters are used in the patterns.

NOTE: If information is changed on the Parameters page, the Experimental Design page has to be updated.

By default, CMOST will keep the output files (*.irf, *.mrf, *.log, *.out, etc.) for all runs. This may consume large amounts of disk space. If this is an issue, it may be advisable to select which experimental design patterns CMOST should keep output files for.

NOTE: If a pattern has already been simulated, the simulation results (IRF) file can be listed in the Existing IRF column. When CMOST reaches that pattern in the design, it will analyze the Existing IRFs data instead of creating a new dataset and running the simulation again. If the cell is clicked, a button will appear that can be used to browse for the IRF file.


Verification Job Patterns This table is only available for Uncertainty Assessment tasks. It is used by CMOST to test the accuracy of the Response Surfaces that it creates. Patterns are created the same as regular experimental design patterns.

Designs Designs can either be user defined or a design can be chosen from a predefined set. CMOST can produce predefined experimental designs with up to 20 parameters for a Sensitivity Analysis Task and up to 8 parameters for an Uncertainty Assessment Task.

Click the Pick a Design button to select a predefined design.

A list of design types will be given. See Appendix F: Experimental Design for more information on the designs. Once a design has been selected, click OK to add the design to the Task file.

Predefined Experimental Design Job Patterns cannot be edited. However, Verification Job Patterns can be edited.

User Custom Designs If a User Custom Design is chosen, the user has the option to select a predefined pattern to work from.


If a blank design is desired, click No when asked if you wish to work with an existing design.

Design Details Various details about the design are listed to the right of the patterns.

Number of Parameters This text field lists the number of Active parameters.

Number of Jobs This text field lists the number of jobs that will be created using the experimental design. Verification jobs are not included for this number.

Resolution The resolution of a predefined design is listed here. Information is only listed for Sensitivity Analysis tasks.

Design Type The type of design that was chosen in the Pick a Design dialog is listed here. User Custom Design is the default. See Appendix F: Experimental Design for more information on the different design types.

Error Messages This box lists any errors that have been found in the experimental design. To update this display, click on the Check Error button located at the bottom of the display.


Influence Matrix The influence matrix gives CMOST optimizers information that will help speed up convergence of the optimization. For each of the Parameters and Local Objective Functions, a value from 0 to 1 is given to represent the effect each parameter will have on the objective function. The more likely a parameter is to affect the Local Objective Function, the higher the value should be. Engineering judgment and understanding of the reservoir model should be applied when modifying the influence matrix. For more information about the optimizers, see the Appendix C: Optimizers.

NOTE: This page must be updated whenever Parameters page and/or Objective Functions page are modified.

By default, all values in the Influence Matrix are 1. This is acceptable for CMOST but it may take longer for a solution to be found.

Values can be entered into the matrix by selecting a cell and entering a value, or one or more cells can be selected and have their values all changed by selecting a value from the Possibility of Influence radio buttons on the right side of the page.


Each value in the influence matrix indicates users estimation on the possibility of whether changing a parameter will affect the result of an objective function:

Definitely Yes = 1 Very Possible = 0.8 Possible = 0.6 Somewhat Possible = 0.4 Doubtfully Possible = 0.2 Definitely No = 0

For example, if a parameter is a global parameter and its adjustment will definitely affect all the objective functions, you may enter 1.0 for all the objective functions. If a parameter is a regional parameter and its adjustment will only affect certain objective functions, you may enter 1.0 for those that will definitely be affected by the parameter. For objective functions that will definitely not be affected by the parameter, 0.0 should be entered. For objective function that could (somewhat possible) be affected by the parameter, 0.4 could be used.

The following influence matrix shows an example which has 8 objective functions, 6 global parameters, and 16 regional parameters.


The first column in the table shows the names of parameters. The remaining columns are parameter-objective function possibility of influence. The example shown is for a history matching problem with 8 objective functions, which are defined for 8 production wells respectively. Objective function W4_HM_Error is defined for Well #4, which is the only well that has bottom-hole pressure measurement. The adjustable parameters include 6 global parameters and 16 regional parameters.

The first parameter initial reservoir pressure INIP is a global parameter, and its adjustment will most likely affect all of the 8 objective functions. Therefore, the value of 0.8 (Very Possible) is given for all objective functions except for the objective function ME04, which is given the value of 1.0 because well bottom-hole pressure match will definitely be affected by INIP.

The seventh parameter BKROCW_1 is a regional parameter, and its adjustment will definitely affect objective function W1_HM_Error. Therefore, the value of 1.0 is given for W1_HM_Error. Moreover, Well #1 is adjacent to well #2. Adjustment on the former may affect the latter. Therefore, 0.4 (Somewhat Possible) is given for W2_HM_Error. For all the other objective functions, 0.0 is used.

CMOST does not require accurate values for the table. However, the table will greatly affect the convergence behavior of CMG DECE optimization if there are many adjustable parameters and objective functions. Therefore, the user should make sure the influence matrix is properly set up. The basic requirement is that global and regional parameters should be differentiated.

Editing the Influence Matrix Values for the influence matrix can be entered by selecting one or more cells and selecting the corresponding Possibility of Influence located at the right side of the grid. Entire rows or columns can be copied or pasted by clicking on the row or column title to select the whole row or column then typing ctrl+c (copy) or ctrl+v (paste). If copying and pasting between influence matrices in different task files, the number of entries in each row or column must match.

The ordering of the influence matrix can be changed by clicking on the row or column title to select the entire row or column then clicking on one of the arrow buttons located at the edge of the grid to move the selection.

Result Observers Result Observers are simulation outputs that CMOST will cache in its results file.

Specific data that should be tracked should be defined in the Result Observers Table. There are two types of Result Observers that can be added to a task file:


Time Series Simulation Result Observers Fixed Date Simulation Result Observers

Times Series Simulation Results Observers collect data for all times during the simulation runs. Fixed Date Simulation Result Observers only collect data at one point for each simulation. Fixed Date Simulation Result Observers are only available for History Matching or Optimization Tasks.

Adding a New Result Observer A new Result Observer can be added by adding information to the bottom row of a Result Observers Table. The Insert button can also be used to add a new entry.

Name The name of the Result Observer should be entered into the Name column. Two Result Observers of the same type cannot have the same name. However, a Time Series Simulation Result Observer can have the same name as a Fixed Date Simulation Result Observer. See Appendix B: Names in CMOST for more information about using names in CMOST.


Origin Type The Origin Type is the type of data that will be retrieved for the Result Observer. This data can come from the following types:

Wells Groups Specials Sectors Layers Leases Parameters Numerical (not supported yet)

All Origin Types come from the simulation results file except for Parameters which retrieves parameter information from the task file.

Origin Name The Origin Name is a specific item of the Origin Type. If there are no items available for Origin Type, the Origin Name list will be empty. If this is the case, that Origin Type should not be used.

For example, if the Origin Type is WELLS, the Origin Name would contain a list of all the wells that are present in the dataset.

Property The Property is a specific item for the Origin Name. No information needs to be entered into this column if the Origin Type is SPECIALS or PARAMETERS. Information is required for all other types.

For example, if the Origin Type is WELLS, the Property cell would have a list of well properties such as Cumulative Oil SC or Gas Rate SC.

Number of Points This option is only available for Times Series Result Observers. Number of Points tells CMOST how many data points should be cached in the CMOST Results File. The points will be spread evenly throughout the entire date range of the simulations. If more detailed results are desired, more points can be chosen.

NOTE: Graphs will take longer to load if many points are displayed.


If only a rough estimate is required, fewer points can be chosen. CMOST can cache fewer points than are in the simulation output file. This can be useful if disk space is a consideration.

NOTE: Values entered here will have no effect on the number of points in the simulation output files.

Multiple cells can be changed at once by selecting the Number of Points cells that should be changed then entering a new value into the last cell that was selected.

Field History File This option is only available for Time Series Simulation Result Observers. If there is production history data available for a specific Result Observer, the field history file can be added. This is usually

Cmost Manual 2009

Documents

rights reserved cmost

cmost studio

cmost version

cmost tasks

cmost formulas

cmost formula samples

base dataset

base files