dynmat-assn

A repo for scripts made for the "Dynamic behaviour of Materials" final Project. The name of our assignment is "Assessing Material Performance in Fuselage Impact". Abaqus CAE 2024 required for running the geometry generation, Abaqus 2023 or newer for running the input files.

For access to compiled data contact Alexandru Olareanu

Contributors:

• Alexandru Olareanu
• Pawel Golla
• Lennard Leypold
• Frederic Huwyle

File description

File Name	Description
abq_full_model.py	Script to be run in Abaqus CAE to generate geometry, boundary conditions, job, etc.
abq_script_model.py	Copy of abq_full_model.py but without parameters at the start.
local_job_gen.py	Script to iterate through parameters and create .inp files using abq_script_model.py.
job_gen_log.txt	Log file with the names of all the generated jobs.
euler_job_submission.py	Script that loads all the jobs into the queue on euler.
euler_job_status.py	Script that reads the .sta files of the jobs.
euler_odb_extract.py	Script that extracts pictures and CSV files from ODB Databases.
euler_sendToBox.py	Script that sends files straight to the storage box through ssh.
euler_pullFromBox.py	Script that sends files straight from the storage box to euler through ssh.
euler_batch_odb_extract.py	Automates the extraction job submission process
local_plots_from_csv.py	Script that generates plots from CSV files on euler.

Usage of scripts:

• Set working directory and other parameters in abq_script_model.py
• Decide what parameters you want to iterate trough and which ones are fixed in local_job_gen.py
• Run local_job_gen.py (about 10 seconds per input file generated)
• Check job_gen_log.txt to make sure everything was generated
• Copy .inp files to euler cluster (or any machine with SLURM queue management)
• Load abaqus module on euler
• Run euler_job_submission.py to submit all the .inp files that have a certain naming scheme, check generated log
• Check queue state with squeue
• Check status files by running euler_job_status.py on euler
• After sims are done, run euler_odb_extraction with abaqus CAE then copy those new files back to local machine. ( Supports filter param)
• Optionally run euler_sendToBox.py to send files matching some naming scheme straight to the storage box
• Optionally run local_plots_from_csv.py to create some quick graphs out of the newly generated .csv files

An entire assortment of Matlab scripts is the used for post-processing

Useful euler commands and things to know:

navigate to /cluster/scratch/yourKurzel
module load abaqus/2023
python3 --version
python3 euler_job_submission.py
sbatch -n 4 -t 1-0 --mem-per-cpu 4G --wrap "abaqus job=Job-1 double cpus=4 scratch=\$TMPDIR"
squeue
scancel <job_ID>
scancel --state=PENDING
scontrol show jobid –dd <job_ID>
python3 euler_job_status.py
abaqus cae noGUI=euler_odb_extract.py -- Job-5