Parameter Sensitivity Analysis and Robust Design Approach for Flux-Switching Permanent Magnet Machines

Abstract

Parameter sensitivity analysis is usually required to select the key parameters with high sensitivity to the optimal goal before the optimization is carried out, especially for flux-switching permanent magnet (FSPM) machines where lot of design parameters should be considered. Unlike the traditional studies on parameter sensitivity, which are generally experience- or statistics-based, and are time-consuming, this paper proposes a parameter sensitivity analysis method of a FSPM machine based on a magnetic equivalent circuit (MEC), which enables the parameters’ sensitivities to be evaluated by their exponential in the nondimensionalized equations, thus providing a fast and accurate way to obtain the parameter sensitivities. Thereafter, the influences of modular manufacturing methods on magnetic performances are discussed, and the robust design approach for the FSPM machine is introduced, which aims to achieve the best machine stability and robustness by setting boundaries on design dimensions when taking into account the manufacturing tolerances. Experimental validations are also presented.