Hi,
I am trying to refine a mesh gradually for only a certain region of my model. The picture shows the region where I would want this gradual refinement to occur with it being the a average edge length at the border of this region and then gradually refine to a very small edge length in the region where the diameter is the smallest. What would be the best way of going about this? I have tried using the sphere regional refinement but that has not produced very desirable results.
Thanks,
Ricardo
Gradual Mesh Refinement
-
Ricardo Roopnarinesingh
- Posts: 15
- Joined: Fri Sep 17, 2021 9:56 am
Gradual Mesh Refinement
- Attachments
-
- Stenosis_Region.png (221.73 KiB) Viewed 259 times
-
David Parker
- Posts: 1712
- Joined: Tue Aug 23, 2005 2:43 pm
Re: Gradual Mesh Refinement
Hi Ricardo,
The sphere regional refinement does create a finer mesh in the region of the stenosis. Why is that not a desirable result? Another way to refine a mesh is to use adaptive meshing (http://simvascular.github.io/docsMeshin ... etgenadapt).
Cheers,
Dave
The sphere regional refinement does create a finer mesh in the region of the stenosis. Why is that not a desirable result? Another way to refine a mesh is to use adaptive meshing (http://simvascular.github.io/docsMeshin ... etgenadapt).
Cheers,
Dave
-
Ricardo Roopnarinesingh
- Posts: 15
- Joined: Fri Sep 17, 2021 9:56 am
Re: Gradual Mesh Refinement
Hi Dave,
This is what my results were looking like with spheres, I wanted a tapering mesh to account for the outer regions of the stenosis. I had to use three spheres with different edge lengths centered at the same point to accomplish this but the mesh has many strange spots. I tried using radius-based meshing but I am getting
"Number of Free Edges on Surface: 144
Number of Non-Manifold Edges on Surface: 38
There are bad edes on surface!
Terminating meshing!"
as an error. What would you recommend to complete the tapering or would it just be worth changing the mesh only at the smallest diameter point of the stenosis instead of tapering. The pictures show the three spheres I used at the same time with different edge length and radius for each. The smallest sphere corresponds to the smallest edge length. The mesh in the picture is the completed mesh using these three spheres.
Thanks
This is what my results were looking like with spheres, I wanted a tapering mesh to account for the outer regions of the stenosis. I had to use three spheres with different edge lengths centered at the same point to accomplish this but the mesh has many strange spots. I tried using radius-based meshing but I am getting
"Number of Free Edges on Surface: 144
Number of Non-Manifold Edges on Surface: 38
There are bad edes on surface!
Terminating meshing!"
as an error. What would you recommend to complete the tapering or would it just be worth changing the mesh only at the smallest diameter point of the stenosis instead of tapering. The pictures show the three spheres I used at the same time with different edge length and radius for each. The smallest sphere corresponds to the smallest edge length. The mesh in the picture is the completed mesh using these three spheres.
Thanks
- Attachments
-
- Sphere_3.png (169.38 KiB) Viewed 207 times
-
- Sphere_2.png (194.97 KiB) Viewed 207 times
-
- Sphere_1.png (214.22 KiB) Viewed 207 times
-
David Parker
- Posts: 1712
- Joined: Tue Aug 23, 2005 2:43 pm
Re: Gradual Mesh Refinement
Hi Ricardo,
That's an interesting meshing pattern, I wonder if the sphere refinement has a bug of that is just the way it works.
I'm not sure what the problem might be with radius-based meshing, seems that the original surface is bad.
Probably your best option is to use adaptive meshing. The mesh refinement will then be based on the flow characteristics in the stenosis region which is what you want.
Another option would be to create a separate face for the stenosis region. The easiest way to do this would be to write a Python script to set the ModelFaceID data array in the SV Project/Models model .vtp file. You might could identify the stenosis region using the centerlines geometry MaximumInscribedSphereRadius data array.
Cheers,
Dave
That's an interesting meshing pattern, I wonder if the sphere refinement has a bug of that is just the way it works.
I'm not sure what the problem might be with radius-based meshing, seems that the original surface is bad.
Probably your best option is to use adaptive meshing. The mesh refinement will then be based on the flow characteristics in the stenosis region which is what you want.
Another option would be to create a separate face for the stenosis region. The easiest way to do this would be to write a Python script to set the ModelFaceID data array in the SV Project/Models model .vtp file. You might could identify the stenosis region using the centerlines geometry MaximumInscribedSphereRadius data array.
Cheers,
Dave