Optimum Design of Rotor for High-Speed Switched Reluctance Motor Using Level Set Method

Abstract

This paper presents a novel rotor structure optimized by using the level set method (LSM) to improve the static torque characteristics of a two-phase 4/2 high-speed switched reluctance motor. The magnetic equivalent circuit approach is used to analyze the influence factors on the static torque. The inner material boundary contained in the rotor saliency, which will be optimized, is implicitly represented by an embedded level set function. The evolution of the inner material boundaries is driven by the normal velocity derived through sensitivity analysis and adjoint variable computation. Optimal rotor configuration produces the nonuniform distribution of magnetic flux density in the air gap at aligned position, which can make the ratio of maximum and minimum inductances increased. High-permeability material is applied to enhance the mean torque. In order to achieve the optimal static torque, the strategy combining finite-element electromagnetic computation with LSM is conducted to optimize the rotor saliency at different rotor positions. The comparison of the optimized rotor configuration using LSM with that using density-based method suggests that the optimized rotor structure obtained using this presented method is easier to implement.