Page 1 of 1

Using SimVascular volume mesh in OpenFoam

Posted: Wed Jan 11, 2017 3:21 pm
by pushindenis
Hi, everyone.

Is it possible to use a volume mesh created by SimVascular in OpenFoam? And if yes, what actions with mesh should I do (i.e. how to convert mesh to format suitable for OpenFoam)?

Thank you in advance for any hint.

Re: Using SimVascular volume mesh in OpenFoam

Posted: Wed Jan 11, 2017 5:02 pm
by adam
Hi Denis,

I am unfamiliar with the format that OpenFoam takes. If you give me more information about the OpenFoam format, I may be able to assist in the operations necessary to convert to that format. For your info, the mesh output from SimVascular's mesher is a tetrahedral mesh in VTK unstructured grid format (vtu).

SimVascular Development Team

Re: Using SimVascular volume mesh in OpenFoam

Posted: Wed Jan 11, 2017 5:17 pm
by amarsden
Hi Denis,

You would need to write a script to convert the mesh file generated in SimVascular to the format appropriate for SimVascular. We're happy to help or advise on this, but we're not part of the OpenFoam development efforts. Since other users may wish to have the same feature, I'd encourage you to contribute this code back to the project!

Best, Prof. Marsden

Re: Using SimVascular volume mesh in OpenFoam

Posted: Mon May 22, 2017 3:43 am
by rudolf.hellmuth
You can export the STL from SimVascular, and use Blender to split the exit patches of the whole STL into separate STLs. Then you have two options. The first option is to mesh using snappyHexMesh (octree method) using all STL patches for assigning BCs. You can make very good prism boundary layers in snappyHexMesh. The second option is to use gmsh then gmshToFoam. In gmsh you can extrude the wall STL inwards a few times to make the prism boundary layers. Than you close the exits with surfaces and fill the inside with tetrahedra using either the Delaunay or the frontFace method. You can also recombine the tetrahedra into hexahedra, if you feel necessary to decrease the cell count. There is a tutorial in the gmsh wiki that describes this method.

Best regards,
Rudolf