微特电机 ›› 2018, Vol. 46 ›› Issue (11): 46-50.doi: 1004-7018(2018)11-0046-05

• 设计分析 • 上一篇    下一篇

基于SVM与NSGA-II的外转子永磁力矩电机多目标优化

李振1,2,陈进华2,张驰2,刘亮1   

  1. 1. 上海大学,上海 200072
    2. 中科院宁波材料技术与工程研究所浙江省机器人与智能制造装备技术重点实验室,宁波 315201
  • 收稿日期:2017-09-13 出版日期:2018-11-28 发布日期:2018-11-28
  • 作者简介:李振(1992—)男,硕士研究生,研究方向为永磁同步电机设计优化。
  • 基金资助:
    国家自然科学基金-浙江两化融合联合基金(U1509202);宁波市重大专项(2017B10014);浙江省基础公益研究计划(LGG18E070007)

Multi-Objective Optimization of External-Rotor Permanent Magnet Torque Motor Based on SVM and NSGA-II

Li Zhen1,2, Chen Jin-hua2, Zhang Chi2, Liu Liang1   

  1. 1. Shanghai University,Shanghai 200072,China
    2. Laboratory of Robotics and Intelligent Manufacturing Equipment Technology,Ningbo Institute of Materials Technology and Engineering,CAS,Ningbo 315201,China
  • Received:2017-09-13 Online:2018-11-28 Published:2018-11-28

摘要:

建立了外转子永磁力矩电机的有限元分析模型,并以转矩密度和转矩波动为优化目标,建立了电机多目标优化问题的数学模型。通过电机结构参数样本空间设计,利用支持向量机(support vector machine,SVM)对电机进行非线性回归建模,利用带精英策略的快速非支配排序遗传算法(nondominated sorting genetic algorithm II,NSGA-II)进行参数寻优。通过有限元仿真验证,NSGA-II与变权数加权的网格搜索法寻优的对比结果,表明了NSGA-II寻优的准确性、有效性及全局性。

关键词: 外转子永磁力矩电机, 多目标优化, 转矩密度, 转矩波动, 支持向量机, 带精英策略的快速非支配排序遗传算法

Abstract:

Finite element model (FEM) of external-rotor permanent magnet torque motor was presented. The mathematical model of multi-objective optimization problem was established based on torque density and torque ripple. Nonlinear regression modeling of the motor was set up, which is based on the sample space of motor structure parameters, and the parameter optimization of the motor is based on the nondominated sorting genetic algorithm II (NSGA-II). After the finite element simulation, the results of NSGA-II and variable weighted grid search method were compared, which shows that the NSGA-II search is accurate, effective and global.

Key words: external-rotor permanent magnet torque motor, multi-objective optimazation, torque density, torque ripple, support vector machines (SVM), improved non-dominated sorting genetic algorithm (NSGA-Ⅱ)

中图分类号: