-
Notifications
You must be signed in to change notification settings - Fork 68
Compiling the CFAST Validation Guide
This wiki describes how to compile the CFAST Validation Guide, which consists of running the CFAST verification and validation cases, plotting the results with Python, and compiling the guide with LaTeX.
-
CFAST (the latest CFAST installation)
-
Git client
-
Python
-
LaTeX, preferably PDF-LaTeX
Note that you should not need a Fortran or C compiler to perform this task.
The process has a number of prerequisite steps:
-
Install CFAST on your computer
-
Fork and clone the CFAST GitHub repository onto your computer
The CFAST executable used for running the validation cases is located in cfast/Utilities/for_bundle/Bin. Copy the CFAST executable you want to use to this folder and name it cfast.exe. If you do not have a Fortran compiler, you can find cfast (linux) or cfast.exe (Windows) in the installation folder.
On Windows only: Add the program background.exe (available in the CFAST installation) to a folder that is in your path, i.e. so that the scripts can find it.
On Windows and linux: Add the executable VandV_Calcs.exe to a folder that is in your path. This program is located in cfast/Source/VandV_Calcs.
Each sub-folder under the Validation folder of the Repository represents an experimental test series. Within each folder, you will find a folder called Experimental_Data and a number of CFAST input files, one for each validation case. For some data sets, there are a sufficient number of cases that each case is located in a folder of its own. However, it is preferred that you keep all the CFAST input files in the same folder.
The Verification folder is organized in a similar way, but there is no experimental data.
In Windows, the CFAST validation suite is located in cfast/Validation. To run all the cases under Windows, type:
runall ALL
This batch script runs all of the CFAST cases included in the validation suite. Allow 15 to 30 minutes for the suite to run, depending on your computer speed.
Windows: in the same Validation folder, run the batch file runvv.bat. This program processes a few special cases.
Windows: cd to the Verification folder and run the batch file runall.bat. This runs the verification cases.
cd to cfast/Utilities/Python and run the Python scripts called CFAST_validation_script.py and then run the script called CFAST_verification_script.py. These two scripts create all the plots that are included in the CFAST Validation Guide. If you are not adding cases to the V&V suite, this is all you need to do.
If you are adding cases, in cfast/Utilities/Python there are two files called CFAST_validation_dataplot_inputs.csv and CFAST_verification_dataplot_inputs.csv which contain all of the information to make the plots you find in the CFAST Validation Guide. The entries in CFAST_validation_dataplot_inputs.csv are organized by the type of output quantity. The first rows are for HGL (Hot Gas Layer) temperature. Most compartment fire validation cases include a comparison of HGL temperature and depth. The entries in CFAST_verification_dataplot_inputs.csv are simply in alphabetical order.
A detailed explanation of the columns within the dataplot input file is found in the wiki.
Go to the folder called cfast/Manuals/Validation_Guide run pdflatex on the file Validation_Guide.tex. You can do this at the command prompt or using a LaTeX editor. If you are working from the command prompt, you will probably have to run the sequence of commands to fully compile the document:
pdflatex Validation_Guide.tex
biber Validation_Guide
pdflatex Validation_Guide.tex
pdflatex Validation_Guide.tex
On a linux machine, there is a script called make_guide.sh, and on a Windows machine make_guide.bat that does this for you.