A Dynamic Dual-Response-Surface Methodology for Optimal Design of a Permanent-Magnet Motor Using Finite-Element Method

This paper presents an approach for improving the process of optimal design of permanent-magnet (PM) motors. A finite-element method (FEM) combined with a dynamic dual-response-surface model (dual-RSM) is proposed to compute an objective function. The FEM is employed as a basic tool for the computation of the objective function. In order to build up the dual-RSM model, a radial basis function and a moving least-square approximation are both employed. The results that are obtained from the dual-RSM model are dynamically compared together with a preset goodness-of-fit, which determines whether the accuracies of the dual-RSM are high enough to replace the FEM computation. This strategy makes the FEM computation not always necessary at all the sample points in the optimization process and ensures the accuracy of the solution. For the purpose of showing the effectiveness of the proposed optimization methodology, the optimal design process of a dual PM-excited synchronous motor is taken as an example in this paper.