RIS valve mesh scale factor bug fix #439
Conversation
|
@hanzhao2020 Why is there no Issue open for this ? |
|
@ktbolt Thanks for the reminder. An issue has been created and linked to this pull request. |
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## main #439 +/- ##
=======================================
Coverage 67.51% 67.51%
=======================================
Files 170 170
Lines 32622 32623 +1
Branches 5741 5741
=======================================
+ Hits 22024 22025 +1
Misses 10459 10459
Partials 139 139 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
aabrown100-git
left a comment
aabrown100-git
left a comment
There was a problem hiding this comment.
@hanzhao2020 I tested this with my left heart FSI case with mesh_scale_factor != 1, and it looks like the SDF fields are being computed correctly. The nonlinear convergence is pretty bad though. I also noticed that convergence is bad for the pipe_uris_3d/ test case as well. Do you know why?
|
@aabrown100-git I think it's because the added resistive forces are similar to penalty terms, affecting the conditioning of the linear system. Using smaller resistance values can improve the nonlinear convergence a bit, but the convergence is still pretty slow. |
aabrown100-git
left a comment
aabrown100-git
left a comment
There was a problem hiding this comment.
Suggesting some changes!
aabrown100-git
left a comment
aabrown100-git
left a comment
There was a problem hiding this comment.
Looks good to me now, thanks.
* Fixed bug for using uris mesh max eNoN and scale factor. * Fixed bugs for RIS valve with mesh scale factor other than 1 * Changed default RIS valve signed distance function to a large value --------- Co-authored-by: Aaron Brown <[email protected]>
3 participants
Resolves #440
Current situation
The current RIS code always uses the unit cm when using the signed distance function to determine if the valve is opening or closing. These parameters should be consistent with the mesh scale factor.
Release Notes
All signed distance function related parameters are using consistent units now.
Testing
This update is tested by the URIS tests.
Code of Conduct & Contributing Guidelines