抄録
This paper presents a multi-objective optimization using genetic programming (GP). In this method, an analysis model is regarded as a tree structure. This tree structure is divided into polygons and material information. Therefore, this method can be applied GP. Additionally, authors take into consideration about shape complexity. The target of electromagnetic device is an interior permanent magnetic synchronous motor (IPMSM). This method optimizes flux barriers in IPMSM by two-dimensional finite element method (2D-FEM). The purpose of this paper is an investigation of design optimization by multi-objective optimization. Authors reveal the pareto optimal solutions of characteristics for average torque and torque ripple of IPMSM.
