Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.

Import mesh parts to compose geometry

Please login with a confirmed email address before reporting spam

Dear Sir and Madam We have completed the 3D topology optimization design of the radiator and created mesh parts of the resulting structure. Afterwards, when the geometry is constructed using this mesh part, a blank section appears as shown in the figure. And the formed geometry surface will have sharp structures. How can we solve the problem of blank areas and ensure the integrity of the topology? If you researchers have related approaches and ideas, we can have a wide exchange, thank you very much! I am looking forward to your replies! Best wishes


6 Replies Last Post Nov 8, 2023, 6:40 a.m. EST
Robert Koslover Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 12 months ago Nov 6, 2023, 2:46 p.m. EST
Updated: 12 months ago Nov 6, 2023, 2:43 p.m. EST
  1. There does not seem to be any figure attached.
  2. Are you trying to import a geometry file created with a different program?
-------------------
Scientific Applications & Research Associates (SARA) Inc.
www.comsol.com/partners-consultants/certified-consultants/sara
1. There does not seem to be any figure attached. 2. Are you trying to import a geometry file created with a different program?

Please login with a confirmed email address before reporting spam

Posted: 12 months ago Nov 6, 2023, 9:56 p.m. EST
Updated: 12 months ago Nov 6, 2023, 9:56 p.m. EST
  1. There does not seem to be any figure attached.
  2. Are you trying to import a geometry file created with a different program?
>1. There does not seem to be any figure attached. >2. Are you trying to import a geometry file created with a different program?


Please login with a confirmed email address before reporting spam

Posted: 12 months ago Nov 6, 2023, 9:58 p.m. EST
Updated: 12 months ago Nov 6, 2023, 9:59 p.m. EST
  1. There does not seem to be any figure attached.
  2. Are you trying to import a geometry file created with a different program?

Files created by other programs and mesh parts that come with comsol don't produce good geometry, and the shape of the geometry changes, even if you don't use "Simplified Mesh".

>>1. There does not seem to be any figure attached. >>2. Are you trying to import a geometry file created with a different program? Files created by other programs and mesh parts that come with comsol don't produce good geometry, and the shape of the geometry changes, even if you don't use "Simplified Mesh".

Kristian Ejlebjærg Jensen COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 12 months ago Nov 7, 2023, 5:05 a.m. EST
Updated: 12 months ago Nov 7, 2023, 5:05 a.m. EST

Performing verification using a mesh based geometry (importing the mesh part directly in the meshing sequence) is usually more robust than converting the mesh to a geometry. The Adapt and (new in 6.2) Remesh operations can be used remove bad surface triangles.

In any case, you might get better results by either using Prescribed Material/Void features to extend the selection of the Helmholtz filter and/or by increase the mesh size (while keeping the filter length constant) so that the design is better resolved. This should be a relatively fast process, if you only solve for the Helmholtz filter using a iterative solver.

Best regards,

Kristian E. Jensen

Technical Product Manager, Optimization.

Performing verification using a mesh based geometry (importing the mesh part directly in the meshing sequence) is usually more robust than converting the mesh to a geometry. The Adapt and (new in 6.2) Remesh operations can be used remove bad surface triangles. In any case, you might get better results by either using Prescribed Material/Void features to extend the selection of the Helmholtz filter and/or by increase the mesh size (while keeping the filter length constant) so that the design is better resolved. This should be a relatively fast process, if you only solve for the Helmholtz filter using a iterative solver. Best regards, Kristian E. Jensen Technical Product Manager, Optimization.

Please login with a confirmed email address before reporting spam

Posted: 12 months ago Nov 8, 2023, 4:33 a.m. EST

Performing verification using a mesh based geometry (importing the mesh part directly in the meshing sequence) is usually more robust than converting the mesh to a geometry. The Adapt and (new in 6.2) Remesh operations can be used remove bad surface triangles.

In any case, you might get better results by either using Prescribed Material/Void features to extend the selection of the Helmholtz filter and/or by increase the mesh size (while keeping the filter length constant) so that the design is better resolved. This should be a relatively fast process, if you only solve for the Helmholtz filter using a iterative solver.

Best regards,

Kristian E. Jensen

Technical Product Manager, Optimization.

Dear Sir I have taken your suggestion of importing the mesh part directly in the meshing sequence to validate it, but a new problem has arisen, i.e. "undefined variable: comp2.spf2.upwind_helpx; geometry: geom2; domain: 0". Boundary 3067 is the red face. I don't know how to solve this problem, can you please tell me how to solve this problem. Attached is my model tree for your reference. Thank you very much! Best regards!

>Performing verification using a mesh based geometry (importing the mesh part directly in the meshing sequence) is usually more robust than converting the mesh to a geometry. The Adapt and (new in 6.2) Remesh operations can be used remove bad surface triangles. > > >In any case, you might get better results by either using Prescribed Material/Void features to extend the selection of the Helmholtz filter and/or by increase the mesh size (while keeping the filter length constant) so that the design is better resolved. This should be a relatively fast process, if you only solve for the Helmholtz filter using a iterative solver. > >Best regards, > >Kristian E. Jensen > >Technical Product Manager, Optimization. Dear Sir I have taken your suggestion of importing the mesh part directly in the meshing sequence to validate it, but a new problem has arisen, i.e. "undefined variable: comp2.spf2.upwind_helpx; geometry: geom2; domain: 0". Boundary 3067 is the red face. I don't know how to solve this problem, can you please tell me how to solve this problem. Attached is my model tree for your reference. Thank you very much! Best regards!


Kristian Ejlebjærg Jensen COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 12 months ago Nov 8, 2023, 6:40 a.m. EST

"Domain 0" sounds strange. I suggest contacting support about that.

Best regards,

Kristian E. Jensen

Technical Product Manager, Optimization.

"Domain 0" sounds strange. I suggest contacting support about that. Best regards, Kristian E. Jensen Technical Product Manager, Optimization.

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.