A new release version of the PrePoMax contains the following FEM features:
improvement of the .dat file reader
bug fixes
And the following user features:
coordinate systems are now highlighted when selected during the creation of BCs and loads
history (for regeneration) was updated to include commands for analysis and commands for postprocessing; regeneration can now be done for the model commands only (the commands for analysis and commands for postprocessing are not regenerated) or for all commands
support for command line parameters was improved
additional .exe file called PrePoMaxCmd.exe was added, which starts PrePoMax as a command line application
both PrePoMax executable files can be used with command line parameters, but PrePoMaxCmd.exe should be used for running PrePoMax in batch mode
-g, --showGui; (Default: Yes) Show Graphical User Interface [Yes | No]. No can only be used for regeneration.
-r, --regenerate; A .pmx file name to be used for regeneration. A work directory -w can be defined for regeneration. If no work directory is defined the current directory is used . All files needed during regeneration (geometry, mesh) must be located in the work directory.
-u, --unitSystem; Unit system type to be used when importing [M_KG_S_C | MM_TON_S_C | M_TON_S_C | IN_LB_S_F | UNIT_LESS].
-w, --workDirectory; A directory path to be used as work directory.
To do some automation, PrePoMax can be run from the command line using the -r switch.
Let’s assume the user prepared a “test.pmx” file where a geometry file “geom.step” is imported, meshed, and analyzed, and some results are post-processed and exported to “results.csv” file using a results history exporter. Now, the user can change the geometry and repeat all the PrePoMax commands automatically.
The new geometry must first be created and saved in the file with the same name as when the .pmx file was created (“geometry.step”). Then, the .pmx file and all files imported while creating the .pmx file (“geometry.step”) must be copied into one folder. For example, “C:\Temp”. Then, using the command line (cmd), the folder should be changed to the folder containing all files (“C:\Temp”) and the command:
C:\path to PrePoMax folder\PrePoMax.exe -r test.pmx
can be used to rerun all commands saved in the .pmx file with the new geometry and create a new “results.csv” file with updated results. The GUI is shown by default to check for errors and can be hidden using the -g switch. If debugging is needed, closing PrePoMax at the end of regeneration can be prevented by using the -x switch.
C:\path to PrePoMax folder\PrePoMax.exe -r test.pmx -g No
I am currently working on some Python scripts to automate FEA and optimize material properties with PrePoMax. I’m trying to understand the scope of the regeneration command versus the way my optimization script currently works, which is directly modifying an .inp file and rerunning.
If I regenerate my .pmx file using a new geometry file, is there a way I can programmatically generate a new .inp file that has the updated mesh information of the new geometry? I need to do this, as the only way I know of to perform optimization on something like material properties is to iteratively modify the .inp file. I am working towards a system where I can run multiple variations of my geometry via regeneration, and optimize each geometry variation for material properties.
If you need to only change the material, changing the .inp file is the way to go. You can even split the .inp file into multiple files using the *Include keyword and have all materials in a separate, easy-to-handle file.
The regenerate can be started from the command line. All the commands in the history (Edit → Edit History) will be regenerated. If the .pmx model is prepared in a way that it exports selected results using the Results → History Outputs → Export, the regeneration will recreate this file. This is an easy way to change the model, run the analysis, and then export the selected results only.
If you need to change the material properties, you can use Tools → Parameters to define some constants and use them in the material definitions (use equations; start with equal = symbol). Then you can change the parameter value by specifying the p switch on the command line. To overwrite parameter a use [a=1.2] after the switch. To overwrite parameters a and b, use quotations and semicolons [“a=1.2; b=31.4”].
Unfortunately, the File → Export function is not recorded in the command history (I will add it), so you cannot conveniently export a model to the .inp file using the regeneration. What you can do is, instead or running the analysis while preparing the .pmx model, you only use Analysis → Check Model. This feature will also create the .inp file and run a fast model check only instead of running the analysis.
I’m struggling to get my command line arguments to behave. Right now I am only using flags -p, -r, -w, and -t with their respective values. I believe this should regenerate the indicated .pmx file, but I’m not seeing the intended behavior. Do I still need to use the -i or -f flag to make regeneration run properly from the command line?
To regenerate, you only need -r followed by the pmx file name. Then, you use -w if you want to change the work directory and -p to change the parameter values.