Robust multi-objective optimisation of an axial-radial flux hybrid excitation permanent magnet synchronous motors based on improved sequential Taguchi method

This paper proposes a novel axial-radial flux hybrid excitation permanent magnet synchronous motor (ARFHE-PMSM), which has better space utilisation and torque density than traditional radial flux PMSM (RFPMM) and axial flux PMSM (AFPMSM). Due to the complex mechanical structure of ARFHE-PMSM, the computation burden is higher and larger than RFPMM and AFPMSM. Therefore, the sequential Taguchi optimisation method is proposed to improve the performance of ARFHE-PMSM and reduce computation burden. The torque, torque ripple, core loss and permanent magnet volume of ARFHE-PMSM are optimisation targets. In the optimisation process, the design parameters are divided into non-sensitive and sensitive parameters based on the sensitivity analysis results. The non-sensitive parameters are first optimised through the finite element model and then the sensitive parameters are optimised. The motor's manufacturing errors are considered in the optimisation process, and the Gaussian membership function and characteristic index are utilised to transform the multi-objective optimisation problem into a single-objective optimisation problem. The iterative Taguchi method narrows the optimisation scope until the convergence conditions are met. The optimal combination of different design parameters of the motor is obtained by analysing the experimental results. Compared with the initial design, the electromagnetic performance and robustness of the ARFHE-PMSM have been significantly increased.