About the HFSS simulation process
"If you are performing an adaptive analysis, HFSS uses the current finite element solution to estimate the regions of the problem domain where the exact solution has strong error. Tetrahedra in these regions are refined."
I would want to know the next:
does HFSS compare the result obtained using current finite element solution with?
Regards.
The highest error areas are defined, as the areas were the electromagnetic field varied the most. Thus HFSS didn?t use other solutions to calculated the local errors.
HFSS initially generate the mesh, then calculate the S-parameters, then refine mesh, calculate S-parameter again, compare the S-parameter difference, if the difference satisfy the stop condition which you set by deta-S, then HFSS treat the mesh is ok, and do the calculation and sweep with this mesh file, if the difference dont satisfy the stop condition, it will do further mesh refinement, except the number of iterations exceed the maxium iteration rounds you specified, in this case, HFSS will use the mesh file got from the last round to proceed the calculation.
Best Regards,
The comparison is done between the last two adaptations.
Itai
Does HFSS divide the model in to sevel regions? And in adaptive process, how did HFSS know which sub region has the most error?
the adaption sotppage condition is based on the convergence of S-parapemeters of the ports you defined. No matter how many subregions being generated, the integration of field component on the surface of the port still can be done.
HFSS dont know which subregion should have the finest mesh to redeuce the error, the meshing process by trial, sometimes it can only based on the aspect ratio of the model to generate mesh. In case of complex object in our model, we use maual mesh generation to get fast convegence of the adaptive meshing.
Best Regards.