Skip to content

Add regressions for all convective numerical schemes for NEMO#1885

Merged
WallyMaier merged 11 commits intodevelopfrom
feature_nemo_ausm_regressions
Jan 18, 2023
Merged

Add regressions for all convective numerical schemes for NEMO#1885
WallyMaier merged 11 commits intodevelopfrom
feature_nemo_ausm_regressions

Conversation

@WallyMaier
Copy link
Contributor

@WallyMaier WallyMaier commented Jan 12, 2023

I am adding regressions for all the convective numerical schemes in NEMO using the implicit solver. All cases run on an inviscid wedge at the moment, but I am happy to make a new folder in the regression.

Related Work

This will allow us to better track our changes and is the start of a more comprehensive set of regression tests for the Nemo solver coming in the future (boundary conditions, etc.)
This directly relates to #1773 to eventually refactor the AUSM schemes.

PR Checklist

  • I am submitting my contribution to the develop branch.
  • My contribution generates no new compiler warnings (try with --warnlevel=3 when using meson).
  • My contribution is commented and consistent with SU2 style (https://su2code.github.io/docs_v7/Style-Guide/).
  • I have added a test case that demonstrates my contribution, if necessary.
  • [] I have updated appropriate documentation (Tutorials, Docs Page, config_template.cpp), if necessary.

@WallyMaier
Copy link
Contributor Author

WallyMaier commented Jan 12, 2023

As mentioned, these all use the same geometry, and I happy to discuss inclusion of different test cases.

Also, I noticed that mutation++ does have a regression case, do we not compile M++ for regression builds?

@WallyMaier WallyMaier changed the title [WIP] Add regressions for all convective numerical schemes for NEMO Add regressions for all convective numerical schemes for NEMO Jan 12, 2023
@WallyMaier WallyMaier requested a review from jtneedels January 12, 2023 18:56
jtneedels
jtneedels reviewed Jan 12, 2023
View reviewed changes
Copy link
Contributor

@jtneedels jtneedels left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This looks good to me, definitely important to get test coverage on these schemes!

Sounds like all these tested cases run, do you have a comparison of convergence history? I'd be curious if they have similar performance on this test case, and would be good to confirm correct behavior past 10 iters.

@WallyMaier
Copy link
Contributor Author

WallyMaier commented Jan 12, 2023

I do not have a good grasp on the performance of each scheme. I fear that may be a much time-intensive task, particularly since these all are implicit at the moment.

@jtneedels
Copy link
Contributor

jtneedels commented Jan 12, 2023

I do not have a good grasp on the performance of each scheme. I fear that may be a much time-intensive task, particularly since these all are implicit at the moment.

I'm thinking just run this test case out to convergence for each scheme then comparing convergence histories. That way if any don't converge, we know we're not grandfathering in broken cases/schemes.

@WallyMaier
Copy link
Contributor Author

WallyMaier commented Jan 18, 2023

Here is a quick summary for the inviscid wedge case using different NEMO schemes. There is clearly work to be done in the convergence/robustness world.
nemo_scheme_regressions.pdf

@WallyMaier WallyMaier merged commit 1016eef into develop Jan 18, 2023
@WallyMaier WallyMaier deleted the feature_nemo_ausm_regressions branch January 18, 2023 20:02
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@pcarruscag pcarruscag pcarruscag approved these changes

@jtneedels jtneedels jtneedels approved these changes

Assignees

No one assigned

Labels

changelog:chore PR: merged

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

3 participants

@WallyMaier @jtneedels @pcarruscag