Bjm/stiffgas

[c0sm0-kramer]

PR Summary

Added Stiffened Gas EoS according to reference Élaboration des lois d'état d'un liquide et de sa vapeur pour les modèles d'écoulements diphasiques, International Journal of Thermal Sciences, O Le Métayer and J Massoni and R Saurel, https://doi.org/10.1016/j.ijthermalsci.2003.09.002. Built with cmake 3.26.4 and make 3.82, compiled and tested with gcc 10.3.0.

This allows for the use of a stiffened gas EoS.

PR Checklist

• Adds a test for any bugs fixed. Adds tests for new features.
• Format your changes by using the make format command after configuring with cmake.
• Document any new features, update documentation for changes made.
• Make sure the copyright notice on any files you modified is up to date.
• After creating a pull request, note it in the CHANGELOG.md file
• If preparing for a new release, update the version in cmake.

[Yurlungur]

Overall looks pretty good! Thanks for adding this. A few minor nitpicks below. In addition, some bigger picture things:

1. We should test this on GPUs. the Darwin CI pipeline will do that for us, but if you have access to a GPU machine it's worth trying to run the tests there yourself too.
2. We try to add the PR to the CHANGELOG.md file
3. It would be good to add documentation about this EOS to the models.rst file in doc/sphinx/src. You can see how the others there are done. It doesn't have to be very extensive.
4. Looks like your code isn't formatted. We use clang-format to automatically format our code. If you have clang-format installed, you can run it on singularity-eos automatically in one of two ways: first, you can type make format_singularity-eos in your build directory. Alternatively, if that isn't working for you, there's a script in utils/scripts called format.sh. You can run it to format your code. I recommend running it as:

FMT=path/to/clang-format VERBOSE=1 utils/scripts/format.sh

where FMT points to the path of your clang-format executable.

[jhp-lanl]

Good start Ben!

We should decide how to do entropy. I'd prefer this to match the ideal gas case, but I'm open to arguments as to why this is better. Numerically, I think the ideal gas case is simpler though.

I'd like to see a few sanity tests like:

• Can you create an ideal gas EOS and a stiffened gas EOS with appropriate parameters and have them return the same result?
• Are the asymptotic limits correct? You'll need to address some of the numerical issues above (divide by zero, etc.), but for example $P(\rho, 0)$ or $P(0, T)$ should return $-P_\infty$.
• Check the bulk modulus against a finite difference of the pressure (see the Gruneisen EOS for an example)
• Use a finite difference to approximate $T\left(\frac{\partial S}{\partial T}\right)_V$ and see if that's equal to the constant heat capacity.

EDIT - And of course add documentation for the new model.

[jhp-lanl]

@c0sm0-kramer after you add documentation I think this is good to go.

EDIT - Actually can you add a couple more tests. I'll make a new review

[jhp-lanl]

Looks like it's just about there. A few more changes and it should be good to go.

Ping me on mattermost when you make the changes and I'll run the CI on gitlab and get this merged when that passes

[jhp-lanl]

Looks good to me! I'll run the CI on gitlab and then if that passes we just need @Yurlungur to approve and this is good to go. Thanks a lot @c0sm0-kramer !

[jhp-lanl]

CI failed on GPUs because of missing i declaration. I've moved the declaration of i into the loops in the test so that it's more limited in scope.

@c0sm0-kramer all we need is documentation and this is good to go. Please add the EOS to the models section of the documentation.

[jhp-lanl]

Small nitpick: can you move the stiff gas EOS model up just after the ideal gas model in the documentation rather than at the end? Otherwise this looks good!

[Yurlungur]

This looks very good! Thanks for all your effort in pulling this through.