I have worked on 2D electromagnetic simulaiton of motor or transformer. I met with a problem. When the magnetic property model of core with muti-axis magnetic anistropy in 2D situation is adopted, the FEM programm with node element converges very difficult. The converge situation have not improve greatly when the solver is changed from Newton-Raphson method to Fixed point method.
Is there some good advise? Is that helpful,if I change my node element method to the edge element in 2D situation?