Stellarator module update

[jjwalkowiak]

Description

Main changes:

• coil minor radius scaling for stellarators
• corrected stress equation for vacuum vessel
• corrected coil height (this was actually more on the pre-process side)
• stellarator module refactoring

Checklist

I confirm that I have completed the following checks:

• My changes follow the PROCESS style guide
• I have justified any large differences in the regression tests caused by this pull request in the comments.
• I have added new tests where appropriate for the changes I have made.
• If I have had to change any existing unit or integration tests, I have justified this change in the pull request comments.
• If I have made documentation changes, I have checked they render correctly.
• I have added documentation for my change, if appropriate.

[timothy-nunn]

The pre-commit quality jobs are failing again. You can install pre-commit as a hook by following our guide here https://ukaea.github.io/PROCESS/development/pre-commit/. This will flag any non-compliant code and automatically update it to follow the style guide (in most cases)

[timothy-nunn]

I am happy with these changes to the unit test file provided you and your team are happy that the numbers have changed for a legitimate reason.

[timothy-nunn]

Who created test: tests/unit/test_stellarator.py::test_stbild ? Either I misunderstand something, or the reference values are contradicting each other. f_st_rmajor=0.99099099099099097, f_st_aspect=1, f_st_rminor=0.99125889880147788, How can it be, that minor and major radius have different scaling, but aspect ration is not changed?

I believe this was an automatic test. When we converted PROCESS from Fortran to Python, we ran regression tests and scraped data from functions we were converting. This allowed us to test that we hadn't changed the codes' behaviour between Fortran and Python

[timothy-nunn]

Please also ensure the unit test failures are fixed and that the regression tests stop failing with an error

I solved most issues with units test, but with regression tests it will be more complicated.

Just to clarify that your regression tests don't need to pass, there are several failure mechanisms that a regression test my experience:

• An error occurs when the input file is run (this is what is happening on this branch). This type of test failure is not ok and must be fixed.
• An error is raised because the MFile is different from the reference MFile. This is ok, as long as you and the reviewers are happy with the changes. When the PR is merged, it will update the reference MFile.

[jjwalkowiak]

Please also ensure the unit test failures are fixed and that the regression tests stop failing with an error

I solved most issues with units test, but with regression tests it will be more complicated.

Just to clarify that your regression tests don't need to pass, there are several failure mechanisms that a regression test my experience:

* An error occurs when the input file is run (this is what is happening on this branch). This type of test failure is not ok and must be fixed.

* An error is raised because the MFile is different from the reference MFile. This is ok, as long as you and the reviewers are happy with the changes. When the PR is merged, it will update the reference MFile.

I think it would be best to create new regression test (input+stella_config). I will deal with this after my vacation.

[timothy-nunn]

Please also ensure the unit test failures are fixed and that the regression tests stop failing with an error

I solved most issues with units test, but with regression tests it will be more complicated.

Just to clarify that your regression tests don't need to pass, there are several failure mechanisms that a regression test my experience:

* An error occurs when the input file is run (this is what is happening on this branch). This type of test failure is not ok and must be fixed.

* An error is raised because the MFile is different from the reference MFile. This is ok, as long as you and the reviewers are happy with the changes. When the PR is merged, it will update the reference MFile.

I think it would be best to create new regression test (input+stella_config). I will deal with this after my vacation.

Hey @jjwalkowiak happy with this solution... as long as the stellarator is well tested I am happy for files to be updated or for new ones to be made.

[jjwalkowiak]

I updated stellarator input file for regression test. stella_config seems to work. On my machine I get: "4626 differences: the reference MFile contains different values for some of the variables. See the warnings for a breakdown of the differences.".
So I think it works as expected. I just didn't found the reference MFile.

[timothy-nunn]

@jjwalkowiak Could you please fix the merge conflicts so that I can run the tests and then hopefully we are not too far off this being ready.

[jjwalkowiak]

Yes, this is on my TODO list. ETA 24h. wt., 17 lut 2026 o 11:12 Timothy ***@***.***> napisał(a):

…

*timothy-nunn* left a comment (ukaea/PROCESS#4082) <#4082 (comment)> @jjwalkowiak <https://github.com/jjwalkowiak> Could you please fix the merge conflicts so that I can run the tests and then hopefully we are not too far off this being ready. — Reply to this email directly, view it on GitHub <#4082 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ARQVQDA6IIL7SWBBQ5R4BBD4MLSSNAVCNFSM6AAAAACT2UM2SCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTSMJTGYZDSOJWHA> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[timothy-nunn]

Hi @jjwalkowiak

There is still an error for one of the Stellarator test files:

FAILED tests/regression/test_process_input_files.py::test_input_file[helias_5b] - RuntimeError:  An error occured while running PROCESS: float division by zero

[jjwalkowiak]

Mea culpa. I used pytest tests/regression -k stellarator, so helias was skipped. śr., 18 lut 2026 o 11:02 Timothy ***@***.***> napisał(a):

…

*timothy-nunn* left a comment (ukaea/PROCESS#4082) <#4082 (comment)> Hi @jjwalkowiak <https://github.com/jjwalkowiak> There is still an error for one of the Stellarator test files: FAILED tests/regression/test_process_input_files.py::test_input_file[helias_5b] - RuntimeError: An error occured while running PROCESS: float division by zero — Reply to this email directly, view it on GitHub <#4082 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ARQVQDF4K2WSVI47GKZQ7IL4MQ2CTAVCNFSM6AAAAACT2UM2SCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTSMJZHA2DSMRUHE> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[jjwalkowiak]

I will fix it ASAP śr., 18 lut 2026 o 11:38 Jędrzej Walkowiak ***@***.***> napisał(a):

…

Mea culpa. I used pytest tests/regression -k stellarator, so helias was skipped. śr., 18 lut 2026 o 11:02 Timothy ***@***.***> napisał(a): > *timothy-nunn* left a comment (ukaea/PROCESS#4082) > <#4082 (comment)> > > Hi @jjwalkowiak <https://github.com/jjwalkowiak> > > There is still an error for one of the Stellarator test files: > > FAILED tests/regression/test_process_input_files.py::test_input_file[helias_5b] - RuntimeError: An error occured while running PROCESS: float division by zero > > — > Reply to this email directly, view it on GitHub > <#4082 (comment)>, or > unsubscribe > <https://github.com/notifications/unsubscribe-auth/ARQVQDF4K2WSVI47GKZQ7IL4MQ2CTAVCNFSM6AAAAACT2UM2SCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTSMJZHA2DSMRUHE> > . > You are receiving this because you were mentioned.Message ID: > ***@***.***> >

[jjwalkowiak]

Ok, now it should work (hopefully)

[timothy-nunn]

Two very minor things and then I am happy once @ym1906 has had a quick look

[jjwalkowiak]

Done. Sorry for this revert madness, I am just too lazy to do this correctly :p

[timothy-nunn]

Happy with the code and changes. Will approve once @ym1906 has approved

[ym1906]

These changes look good to me. There's nothing alarming in the regression tests, and further to previous comments there are not significant modelling changes.