Optionally, using FileGen to prepare Eclipse input data

Once a project has been selected (see Project Selection), the active project will be displayed in the title bar of the TransGen Control Menu.

For a new TransGen Project, the input Eclipse format data required from the reservoir simulation model might need to be restructured using FileGen.  
This optional application is designed to make the data files more manageable for the user. It filters data contained in very large Eclipse input files into a number of smaller files based on essential keywords and also creates an initial project file which is then used by WizGen.

The FileGen icon on the Control Menu will only be functional if the Allow access to FileGen option is toggled "on" in the Project Information window.

Click on the third icon in the Control Menu to launch FileGen.

FileGen will search through one or several Eclipse data files (following INCLUDE statements) to locate the data TransGen requires and places the data in *.DATA files in the TransGen Project directory or in a subdirectory which it creates with the name <Project name>_INPUT.  

HINT:- See Essential Eclipse Data in the section on TransGen data requirements for a list of Eclipse keywords that TransGen recognises and needs.

The first stage is to open the Eclipse files - click on the Open button to display the Choose input files window as shown below.

Click on Add to open the Choose file name window and navigate to the relevant directory containing the Eclipse files to be used as input.

Select a file and click on the Apply button to add that Eclipse file to the Choose input files window.
Repeat until all the required files have been added; then press OK to add the last one and dismiss the Choose file name window.

Any files that have been added in error can be removed from the input file list by selecting them and pressing Remove.

When the list of Eclipse files is complete, press OK to close the Choose input files window.  
Then click on the Merge at the top of the main FileGen menu to follow the links via INCLUDE keywords to secondary files, and create a single temporary input file in the /tmp subdirectory.

The next stage is to find the data associated with each Eclipse keyword and either place it in the <Project>.TGDATA file, or in an associated include file.  The latter is best for the large data blocks.  By default, COORD, ZCORN, ACTNUM, NTG,  PERMX, PERMY, PORO, MULTX, MULTY, VSH, SWAT and FAULTS data are placed as secondary files by FileGen in the <Project>._INPUT subdirectory, with INCLUDE statements in the <Project>.TGDATA file.  

To change the location where FileGen places the data associated with a particular keyword, click the Browse... button for the particular keyword.

This launches a file selection box.  It might, for example, be useful to place the poroperm data in a separate subdirectories if there are several realisations.

FileGen is designed to allow you to structure the data in a tidy and efficient manner.  
NOTE:- FileGen makes the directories NEW_INPUT and NEW_OUTPUT by default,  but you may have made others in addition outside FileGen.  (If a directory does not exist, this is not a problem - FileGen will prompt with a red warning message to ask you to make the directory when it tries to write the files).

First select the directory in which to wish to place the data, e.g. F64_POROPERM.
Add the new filename for the data (e.g. PERMX.DATA) and click on OK.
Repeat to place PERMY and PORO data in appropriately named files in the same new F64_POROPERM subdirectory.

Once the location for the input data has been defined, click on Search, and FileGen will search for the data.  If the Eclipse source files are very large, this may take some time.

FileGen works down the list of keywords,  indicating its progress in the 'State' column:  

Data that has been found will be identified by a tick

Data that is absent will be marked by a cross, and the filename will be greyed out when it has finished searching.  

Data that FileGen is actively looking for is shown by an eye symbol

and lastly; data that FileGen still has to search for are shown by a question mark.

NOTE:- The way data is read in from an Eclipse run file is not always straightforward. Data may be missing because a necessary file was omitted from the list of input files or the data was input in another way.  EQUALS blocks can be used to set the value of any input array,  and COPY may be used to transfer the contents of one array into another.   ADD and MULTIPLY may also be used to modify the values.

Multiple occurences of Keywords (shown by the Number field) should be investigated.

It is possible that an Eclipse input file was read twice on its own or through an include statement.  In this case, the redundant data should be omitted.

Alternatively, the data for the Eclipse parameter arrays may be defined in several places.  For example, the PERMX data may initially have been set to a certain value and modified by one or more subsequent PERMX block each perhaps with an associated BOX keyword.    In this case, the order in which the PERMX blocks are read could be very important.

When the data search is complete, the data can be written to the TransGen input files by clicking on the Write button (as shown below).  
As files are successfully written, the State marker will change as shown below.  
Note what data has been successfully written (HINT:- See Essential Eclipse Data in the section on TransGen data requirements for a list of Eclipse keywords that TransGen recognises and needs) and click on Quit to exit FileGen.

Once the relevant TransGen input files have been created, use the WizGen module to create a model runfile.