微特电机 ›› 2020, Vol. 48 ›› Issue (4): 21-25.

• 设计分析 • 上一篇    下一篇

基于多目标遗传算法的电机噪声优化

张耀, 杜荣法, 赵新超, 叶乾杰   

  1. 卧龙电气驱动集团有限公司,上虞 312300
  • 收稿日期:2019-12-30 出版日期:2020-04-28 发布日期:2020-04-24
  • 作者简介:张耀(1989--),研究方向为永磁同步电机仿真技术,振动噪声技术

Motor Noise Optimization Based on Multi-Objective Genetic Algorithm

ZHANG Yao, DU Rong-fa, ZHAO Xin-chao, YE Qian-jie   

  1. Wolong Electric Group Co.,Ltd.,Shangyu 312300,China
  • Received:2019-12-30 Online:2020-04-28 Published:2020-04-24

摘要: 基于J-MAG软件建立了多目标优化有限元模型,将定转子冲片拓扑结构中影响电机性能的关键尺寸,位置进行参数化;采用软件内嵌的遗传优化算法,进行了电机单目标性能的优化,将转矩脉动,输出转矩,齿槽转矩作为本次电机优化的目标函数;分析了不同性能指标之间的关联性.通过各指标加权,综合考虑转矩脉动,输出转矩,齿槽转矩对电机振动噪声性能的影响,得出最优的设计方案;通过对比实验,验证了该优化方案的有效性.

关键词: 多目标, 参数化, 遗传算法, 优化设计

Abstract: Based on J-MAG, a multi-objective optimization finite element model was established, parameterized the key dimensions and positions the stator and rotor laminations, that affect the motor performance in the topology.Used the genetic optimization algorithm embedded in the software, the single-objective performance of the motor was optimized, and the torque ripple, output torque and tooth groove torque were taken as the objective functions of the motor optimization.The correlation between different performance indexes was analyzed.Used the weighted index, the influence of torque ripple, output torque and tooth groove torque on the vibration and noise performance of the motor was comprehensively considered, the optimal design scheme was obtained.The effectiveness of the optimization scheme was verified by experiment.

Key words: multi-objective, parameterization, genetic algorithm, optimal design

中图分类号: