I’m working right now on the modeling of the surface elasticity of a droplet. The problem has azimuthal symmetry for which the geometry and mesh is the surface of a unit sphere (Only the surface is being modeled, no bulk Material). I used SALOME to create the sphere and with Explode and Partition I extracted the surface and meshed it. Exported it to a .med File which I exported as a .msh with GMSH. Finally I converted the .msh to a .xml with dolfin-convert. This is the method for complex geometries we were taught in a basic FEniCS course.
Additionally I used the mesh created by ROGNES in her paper.
The mesh created by ROGNES produces results with perfect azimuthal symmetry, where as my mesh does look symmetric at first glance but by changing views I can see that it really isn’t. Even while tripling the amount of elements.
My question is then, why do you guys think this happens and how would you proceed to create and mesh such a geometry? Should I use another program or method to create more complicated geometries for FEniCS? This may be a dumb question but I’m not really versed on FEniCS or Open-Source Geometry and Mesh Generators.
Hopefully someone could help me,