Multiobjective Optimization of the PMSM With the Same Number of Poles and Slots Considering Dynamic Response and Torque Performance

For several applications, such as robotic joints, it is necessary to simultaneously optimize the torque density, torque ripple, and dynamic response capability of permanent magnet synchronous machines (PMSMs). However, the method for evaluating the dynamic response capability based on the structural parameters of the machine and its co-optimization with the torque performance is not clear. Therefore, this article takes the PMSM with the same number of poles and slots as the research object and optimize these three objectives simultaneously. First, a generalized electromagnetic performance prediction model for this type of machine with different topologies is proposed based on the improved subdomain model. In addition, the evaluation model of the dynamic response capability of the PMSM is established by the mechanical and voltage equations. Based on the proposed model, the sensitivity of each parameter to the torque density, torque ripple, and dynamic response capability of the machine is analyzed, and the NSGA-II algorithm is used to optimize these three objectives simultaneously. Finally, the performance of the selected case is verified based on finite elements analysis, which is improved in all three objectives.

I. Introduction

Permanent magnet synchronous machines (PMSMs) have a wide range of applications and development prospects in many fields, such as electric vehicles, aerospace, and rail transportation, due to their compact structure, high power density, and high efficiency [1], [2], [3]. Different applications have different requirements for the electromagnetic characteristics of machines. For instance, the main requirements for machines in robotic joints include high torque density, low torque ripple, and high dynamic response capability [4], [5]. Among them, the method of evaluating the dynamic response capability based on the structural parameters is not clear. Therefore, it seems challenging to simultaneously optimize the torque performance and dynamic response capability of PMSMs based on structural parameters.