Update domain.py

[keiyamamo]

Two main reasons for this PR

1. There was a problem restarting the problem. The problem should be fixed with this PR. (I tested restarting and it works now)
2. domain.py was quite complicated to understand without enough documentation. Therefore, I added args/returns as well as underlying assumption for some of the variables that was not explicitly stated.

Please review my comments and feel free to add comments on them!

Kei

[keiyamamo]

Hi Daivd,

Good point! I did test 1, 3, 5, 7, but not 2, 4, 6. Do you have any problems with multiple solid/fluid regions that I can test?

Best, Kei

[dbruneau-mie]

Hey Kei, here is a simulation with mutliple fluid regions in the Aneurysm Workflow FSI: https://github.com/dbruneau-mie/Aneurysm_Workflow_FSI/blob/master/simulations/stenosis_FC/pulsatile_vessel.py

[dbruneau-mie]

And in the solver itself there is a cylinder with 2 solid regions https://github.com/KVSlab/turtleFSI/blob/master/turtleFSI/problems/Test_Cylinder/problem_aneu_2solid.py

[keiyamamo]

Hi @dbruneau-mie

Thanks for pointing out! I just reverted the change and will re-open the pull-request after I test the changes. We should consider adding multiple fluid/solid regions as a part of test...

Best,

[dbruneau-mie]

Yes, I think we need more tests for the solver. A restart test would also be very useful

[keiyamamo]

I did work on that, but found that the problem is multi-folded, meaning we need to fix time loop, drag/lift evaluation and not-designed-to restart. But I will work on this now that it turned out that it is important we have well designed test to avoid manually performing many kinds of test.