Multi-physics and multi-objective optimization of a permanent magnet-assisted synchronous reluctance machine for traction applications

Abstract

This contribution addresses the rotor design process of a Permanent Magnet-assisted Synchronous Reluctance Machine by adopting a multi-physics and multi-objective optimization algorithm. A Finite Element (FE) approach is employed to determine the electromagnetic and structural responses during the optimization. In particular, a detailed FE structural modeling is used, which often is based on simplifications and inaccuracies in the available literature. A genetic algorithm is adopted, with the objectives being the maximization of the mean torque, the minimization of the torque ripple and the minimization of the stress in the rotor. A parametric analysis of the geometric features precedes the optimization to establish the design variables which mostly affect the machine performance, and thus to reduce the computational cost of the optimization. The presented methodology consists of a useful tool for the final stages of the design process, and provides a rotor with a torque ripple reduced by 15.1% compared to an existing design used as a benchmark, while the mean torque and the maximum stress remain the same as the original configuration.