基于IPSO-BPNN的电机控制方法研究  

Research on Motor Control Method Based on IPSO-BPNN

作　　者：梁策 张兵 朱建阳[1] LIANG Ce;ZHANG Bing;ZHU Jianyang(Faculty of Mechanical Engineering,Wuhan University of Science and Technology,Wuhan Hubei 430081,China;School of Mechanical Engineering,Dongguan University of Technology,Dongguan Guangdong 523808,China)

机构地区：[1]武汉科技大学机械工程学院,湖北武汉430081 [2]东莞理工学院机械工程学院,广东东莞523808

出　　处：《机床与液压》2025年第7期81-87,共7页Machine Tool & Hydraulics

摘　　要：永磁同步电机是一个典型的非线性多变量强耦合系统,会受外部扰动、参数摄动和磁场非线性等因素的影响。针对这一问题,提出一种使用改进粒子群算法优化BP神经网络的PID控制器(IPSO-BPNN-PID)。通过引入自适应变异与随机权重对粒子群算法进行优化,以提升算法的全局搜索能力与收敛速度。利用IPSO算法优化神经网络的初始权值,提升了神经网络的学习速度;并结合神经网络的非线性逼近能力,对PID进行在线调节,以提高PID的响应速度和精度。建立PMSM双闭环调速系统,并采用优化后的IPSO-BPNN算法对PID控制器参数进行在线整定。结果表明:与标准粒子群算法相比,改进后的粒子群算法适应度更佳,收敛速度比标准PSO算法快24%;IPSO-BPNN-PID控制器的平均响应速度分别比PID控制器和BPNN-PID控制器提高了53.57%、19.77%,平均超调量比BPNN-PID控制器低41.67%,表明提出的IPSO-BPNN-PID控制器显著提升了PMSM驱动系统的响应速度和动态抗扰动能力等性能。Permanent magnet synchronous(PMSM)motor is a typical nonlinear multi-variable strongly coupled system,which is affected by external disturbances,parameter perturbations and magnetic field nonlinearity.In order to solve this problem,an improved particle swarm optimization optimization PID controller for back propagation neural network(IPSO-BPNN-PID)was proposed.The particle swarm optimization was optimized by introducing adaptive mutation and random weights to improve the global search ability and convergence speed of the algorithm.The IPSO algorithm was used to optimize the initial weight of the neural network,and the learning speed of the neural network was improved.Combined with the nonlinear approximation ability of the neural network,the PID was adjusted online to improve the response speed and accuracy of the PID.A PMSM double closed-loop speed regulation system was established,and the optimized IPSO-BPNN algorithm was used to set the parameters of the PID controller online.The results show that compared with the standard particle swarm algorithm,the improved particle swarm optimization has better adaptability,and the convergence speed is 24%faster than that of the standard PSO algorithm.The average response speed of the IPSO-BPNN-PID controller is 53.57%and 19.77%faster than that of the PID controller and the BPNN-PID controller,respectively,and the average overshoot is 41.67%lower than that of the BPNN-PID controller,indicating that the proposed ISO-BPNN-PID controller significantly improves the performance of the PMSM drive system,such as the response speed and dynamic anti-disturbance ability.

关 键 词：永磁同步电机 粒子群算法 神经网络控制器 电机控制方法 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置] TM341[自动化与计算机技术—控制科学与工程]