基于粒子群算法优化BP神经网络的SRM磁链模型  

SRM Flux Linkage Model of Optimizing BP Neural Network Based on PSO Algorithm

作　　者：郝娟 HAO Juan(College of Energy and Electrical Engineering,Hohai University,Nanjing 211100,China)

机构地区：[1]河海大学能源与电气学院,江苏南京211100

出　　处：《机械制造与自动化》2018年第2期130-132,共3页Machine Building & Automation

摘　　要：以一台6/4级SRM电机模型的电流和磁链为输入,转子位置角度为输出,拟合了磁链-电流-角度模型,并在MATLAB/Simulink中进行了仿真实验。在拟合模型时,为了提高训练效率,简化拟合模型,采用了粒子群优化BP神经网络隐含层神经元个数的算法,并进行了仿真实验。结果表明,粒子群算法优化BP神经网络的控制策略具有较高的训练效率。This paper takes the current and flux linkage of 6/4 SRM motor as the input and its rotor position angle as the output to fit the flux-current-angle model and makes an experiment on its simulation in MATLAB/Simulink.To improve the training efficiency and simplify the fitting model,according to PSO algorithm,the number of neurons of double hidden layers of BP neural network is optimized and the results show that it has higher training result.

关 键 词：开关磁阻电机 BP神经网络 粒子群算法 

分 类 号：TM352[电气工程—电机]