Hello all,
I need help and I hope somebody with experience of using iterative method for large 3D problem would be willing to give me help.
I wish to calculate large 3D periodic problems in the RF module but i can not seem to get the calculation to converge properly with GMRES although i hear its the solver i should try.
As i read through the help, i see that there should be special conditions i should take care of when meshing since the solver include geometric multigrid algorithm somewhere. Can somebody tell me how to prepare a mesh suitable for geometric multigrid in large 3D problems easily?
I attach my mock comsol 4.3 mph file, is such meshing adequate for the geometric multigrid algorithm?
I may have completely misdiagnosed the bug in my calculation, i hope somebody would be willing to tell me that if its so.
Many thanks in advance
I need help and I hope somebody with experience of using iterative method for large 3D problem would be willing to give me help.
I wish to calculate large 3D periodic problems in the RF module but i can not seem to get the calculation to converge properly with GMRES although i hear its the solver i should try.
As i read through the help, i see that there should be special conditions i should take care of when meshing since the solver include geometric multigrid algorithm somewhere. Can somebody tell me how to prepare a mesh suitable for geometric multigrid in large 3D problems easily?
I attach my mock comsol 4.3 mph file, is such meshing adequate for the geometric multigrid algorithm?
I may have completely misdiagnosed the bug in my calculation, i hope somebody would be willing to tell me that if its so.
Many thanks in advance