A parallel computing for magnetostatic problems with iterative domain decomposition method

A numerical analysis with an iterative domain decomposition method is performed for magnetostatic problems, which is based on the initial step of an iterative scheme without the Lagrange multiplier. The magnetic vector potential, which is considered as an unknown function, is descretized by the Nedelec element of simplex type. The iterative domain decomposition method is combined with the Conjugate Gradient (CG) procedure, and the Hierarchical Domain Decomposition Method (HDDM), which has been shown effective for structural problems, is adopted for the parallel computing. Numerical results show that the CG procedure converges, and that the computed magnetic flux density is suitable. Moreover a magnetostatic problem where the number of degrees of freedom is about one million can be solved by using HDDM.