Skip to content

Python and Paraview postprocessing tools for use with NREL's SOWFA

License

Notifications You must be signed in to change notification settings

NotDrJeff/sowfatools

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

sowfatools

This package (not yet published) contains a miscellaneous assortment of postprocessing functionality for use with CFD simulations peformed in NREL's SOWFA code. It's FAR from complete!

I'm writing it for my own use primarily, so it's very specialised to my simulation cases. But maybe if your're working on a similar case you'll find something useful here. Then again, maybe not!

It would be great to produce a standardised set of postprocessing scripts and functions so that results from these cases can be quickly compared with other published studies. But I think that is a bit fanciful right now. If you do happen to stumble on this and want to contribute, crack ahead.

PYTHON

Here's how you can set up the same python environment as me:

I'm using conda 22.9.0. If you're using a HPC cluster, you may have restrictive file quotas in your home directory. In this case you'll need to set your package and environment paths to somewhere else (becuase they end up having a LOT of files!)

You may need to run conda init the first time your using it.

To get paraview, you'll need the conda-forge channel to your condarc file with the command conda config --add channels conda-forge

Then you can create an evironment (I call mine sowfapy) with python 3.11 and the neccesary packages like this:

conda create -n sowfapy python=3.11 numpy scipy matplotlib sympy paraview

You can activate the environment with conda activate sowfapy

PARAVIEW

Getting paraview (pvbatch) working in parallel sucessfully is another kettle of fish. But downloading the precompiled binaries and using the mpi binary included with that seems to work (with some limitations).

On my system, I submit a SLURM job script with the following command:

if [ -f $1 ]; then
    ~/OpenFOAM/ParaView-5.11.1/bin/mpiexec -n 16 ~/OpenFOAM/ParaView-5.11.1/bin/pvbatch "$@" 2>&1
fi

Submit the script with the python filename as the first argument. All other arguments will then be passed to the script.

Good luck!