Design optimisation of submersible permanent magnet synchronous motor by combined DOE and Taguchi approach

Abstract

This study presents an optimisation procedure for a submersible permanent magnet synchronous motor (PMSM) to achieve the maximum efficiency and minimum cogging torque simultaneously. Numerous geometric parameters can be used to define submersible PMSM. To identify the most significant parameters for optimisation, the fractional factorial design out of design of experiment (DOE) is employed for the screening activity, considering the interaction effects into account. The manufacturing process of the motor causes dimension tolerances, shape and position tolerances, installation errors and deviations of material properties. To ensure a good robustness for the performance of the submersible PMSM, Taguchi's robust design method is employed. The screened significant parameters are taken as the control factors. The noise factors include inner diameter of stator, the air-gap length, the remanence of the magnets and the dimension tolerances of magnets. Numerical results demonstrate that this method is an effective optimisation design procedure to improve the performance of submersible PMSM, which is finally proved by the prototype test.