基于响应面法的电动汽车用电励磁同步电机多目标优化设计  

Multi-objective Optimization Design of Electric Vehicle Excitation Synchronous Motor based on Response Surface Methodology

作　　者：周鹏飞 赵南南[1] 乔鹏博 李骋 王议磊 ZHOU Pengfei;ZHAO Nannan;QIAO Pengbo;LI Cheng;WANG Yilei(Xi'an University of Architecture and Technology,Xi'an 710055,China)

机构地区：[1]西安建筑科技大学,西安710055

出　　处：《大电机技术》2025年第2期18-27,共10页Large Electric Machine and Hydraulic Turbine

基　　金：国家自然科学基金青年项目(52206109)。

摘　　要：针对车用电励磁同步电机输出转矩较低的问题,本文以一台800V电励磁同步电机为研究对象,采用了一种响应面模型与多目标遗传算法相结合的方法对电机转矩性能进行优化。首先,研究了转子偏心对齿槽转矩的影响,并对电机的齿槽转矩进行了初步优化;其次,通过田口正交优化方法进行灵敏度分析,确定了电机初始优化目标;再次,利用响应面法构建平均转矩、转矩脉动及齿槽转矩的代理优化模型,并采用非支配排序遗传算法进行多目标优化设计;最后,对采用响应面-正交法和本文方法的优化结果进行了对比分析,验证了本文方法的有效性。优化后的平均转矩提高了11.27%,转矩脉动降低了5.74%,齿槽转矩降低了53.05%。To improve the torque performance of electric excitation synchronous motors for automotive applications,this paper focuses on an 800V EESM and employs a method that combines response surface methodology with a multi-objective genetic algorithm.Firstly,the influence of rotor eccentricity on cogging torque was studied,and a preliminary optimization of the motors cogging torque was performed.Secondly,sensitivity analysis was conducted using the Taguchi orthogonal optimization method to determine the initial optimization objectives for the motor.Thirdly,surrogate models for average torque,torque ripple,and cogging torque were constructed using the RSM,and multiobjective optimization design was carried out using the non-dominated sorting genetic algorithm.Finally,a comparative analysis of the optimization results obtained from the RSM-orthogonal method and the proposed method was conducted,validating the effectiveness of the proposed method.The average torque is increased by 11.27%,torque ripple is decreased by 5.74%,and cogging torque is reduced by 53.05%.

关 键 词：电励磁同步电机 转矩性能 田口法 响应面模型 多目标遗传算法 

分 类 号：TM341[电气工程—电机]