Add functionality to check mesh inlets and outlets and flatten if necessary

[dbruneau-mie]

Added new script "check_flatten_in_out.py", which takes a .h5 mesh file and checks whether inlets and outlets are flat, then flattens them if they fall below a certain tolerance

[codecov-commenter]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 78.75%. Comparing base (7d490f4) to head (0d4635a).

Additional details and impacted files

@@           Coverage Diff           @@
##           master     #176   +/-   ##
=======================================
  Coverage   78.75%   78.75%           
=======================================
  Files          32       32           
  Lines        3742     3742           
=======================================
  Hits         2947     2947           
  Misses        795      795           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[keiyamamo]

Hi @dbruneau-mie

Thank you for the script! I tried with several geometries and can see that it improves the mesh.

Just one small comment is that solid inlet and outlet is still not flat. I guess this is not really an issue since flat outlet is important for only fluid domain. I was trying to figure out why inlet/outlet become non-flat and found out that it is due to BoundarylayerMesh from VMTK. In short, VMTK computes surface normal and use those vectors to create boundary layer mesh. However, those vectors are skewed at the ends, and cause this problem, as solid domain is constructed with boundary layer mesh.

It is possible to simply merge this, but it makes more sense to incorporate this as a final step of meshing. If it is okay with you, I will copy this function inside src/vasp/automatedPreprocessing/preprocessing_common.py and call it inside main meshing script. Let me know what you think.

[dbruneau-mie]

Hey Kei, could you merge the PR, then integrate the function into the big preprocessing script in a separate commit? I would like the isolated script preserved at some moment in time, and that way I have my name on the commit

…

On Mon, Sept 2, 2024, 4:56 a.m. Kei Yamamoto ***@***.***> wrote: Hi @dbruneau-mie <https://github.com/dbruneau-mie> Thank you for the script! I tried with several geometries and can see that it improves the mesh. Just one small comment is that solid inlet and outlet is still not flat. I guess this is not really an issue since flat outlet is important for only fluid domain. I was trying to figure out why inlet/outlet become non-flat and found out that it is due to BoundarylayerMesh from VMTK. In short, VMTK computes surface normal and use those vectors to create boundary layer mesh. However, those vectors are skewed at the ends, and cause this problem, as solid domain is constructed with boundary layer mesh. Screenshot.2024-09-02.at.10.38.48.png (view on web) <https://github.com/user-attachments/assets/84a2dbe2-c55b-4ce1-89f8-dc4251ca0dde> It is possible to simply merge this, but it makes more sense to incorporate this as a final step of meshing. If it is okay with you, I will copy this function inside src/vasp/automatedPreprocessing/preprocessing_common.py and call it inside main meshing script. Let me know what you think. — Reply to this email directly, view it on GitHub <#176 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ANWLGEJF2QHYDAU3LQ6QEZLZUQR4VAVCNFSM6AAAAABNNGMHHWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDGMRUGE4TCNBWGI> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[keiyamamo]

@dbruneau-mie Sure! I will merge and then create a separate commit.