Hi all,
I have been attempting to mesh segmented CT data for use in Elmer, and am wondering at the best route to take. I am currently segmenting the CT data and exporting the segments to large .STL files, and then post-processing for intersections and smoothness in a mesh modelling package.
My ultimate goal is to use those surfaces as boundary surfaces against which a mesher will generate tet meshes for electrostatics simulations, but the problem I’m running into is that both GMSH and NETGEN are using the topology of the .STLs, which with the number of elements they have is destroying my available compute resources.
I’ve used OpenFOAM a bit, and the snappyHexMesh routine of generating a box mesh, and refining by snapping to .STL surfaces is more what I’m looking for.
I’ve done a literature and tutorial search for both GMSH and Salome workflows to accomplish meshing complex surfaces, and haven’t been able to find anything that shows how to mesh large, arbitrary geometries.
Does anyone have advice for a workflow that would help me generate volume meshes that conform to organic, arbitrary shapes for use in Elmer? I’ve seen some work done on expanded metal foams, so I know it has to be possible.
For Salome, I’ve tried decimating my .STLs down to the minimum acceptable resolution, importing, and generating a solid using the “Build Solid” tool. But when I try to mesh using the 1D-2D-3D NETGEN algorithm and an appropriate hypothesis, it hangs.