Optimal Design of SPMSM for UAV Using Adjusted Evolution Strategy

This paper proposes the optimization algorithm, an adjusted evolution strategy (AES) and verifies performance of the AES applying the optimal design of the outer-rotor surface mounted permanent magnet synchronous motor (SPMSM) for unmanned aerial vehicle. The proposed algorithm improved disadvantage of the conventional evolution strategy, which requires long computation time when solving the optimization problem, utilizing an exponential function of the annealing factor. In addition, the AES further enhances the convergence characteristic by combining with the compass segment method which is deterministic algorithm. This paper shows that the superiority of AES over NGA and ES in three test functions, with 30% to 70% reduction in the number of function calls, and an improvement of 5%p to 30%p in the convergence accuracy. Finally, the AES is applied to the optimization design of the outer-rotor SPMSM, and this paper derived the optimal model of the SPMSM to increase average torque and reduce torque ripple that causes vibration in unmanned aerial vehicle. As a result, proposed algorithm successfully derived the optimal model and verified the robustness. Therefore, the results of this study are expected to be widely applied to the multimodal optimization problem of various electrical machines utilizing FEM.