基于BP神经网络优化自抗扰PMSM高精度速度控制  

PMSM Speed Control Based on Active Disturbance Rejection Control Optimized by BP Network

作　　者：付文强 赵东标[1] 赵世超 FU Wen-qiang;ZHAO Dong-biao;ZHAO Shi-chao(College of Mechanical and Electronic Engineering,Nanjing University of Aeronautics and Astronautics,Nanjing 210016,China)

机构地区：[1]南京航空航天大学机电学院,南京210016

出　　处：《微特电机》2020年第12期50-54,共5页Small & Special Electrical Machines

基　　金：国家重点基础研究发展计划项目(973计划,2014CB046501)。

摘　　要：永磁同步电机(PMSM)是一个复杂非线性系统,其速度精度与控制器的性能紧密相关。为了提高永磁同步电机调速系统的速度精度及抗干扰能力,提出一种BP神经网络优化自抗扰控制(ADRC)的方法。以永磁同步电机的实际转速和期望转速作为速度控制器的输入,将BP神经网络嵌入扩张状态观测器(ESO)中,利用BP网络拟合部分扰动,从而降低ESO观测器的计算复杂度,提高观测精度。仿真结果表明:该方法具有良好的可行性,提高了控制器的观测精度,控制器的速度精度和鲁棒性得到一定的提升。Permanent magnet synchronous motor(PMSM)is a complex nonlinear system.Its speed accuracy has a great relationship with the performance of the controller.In order to improve the anti-disturbance ability and the speed accuracy of PMSM speed regulation system,a novel control method was proposed based on active disturbance rejection control(ADRC)optimized by back propagation(BP)neural network.The speed controller was designed with the given speed and actual speed as the input signals.The BP network was embedded in the extended state observer(ESO)to reduce the computational complexity of the observer and improve the accuracy of the observer.Simulation results demonstrated that the proposed method had good feasibility and that ADRC observation precision was improved.The speed accuracy and robustness of the controller were further improved.

关 键 词：永磁同步电机 BP神经网络 扩张状态观测器 自抗扰控制 速度控制 

分 类 号：TM351[电气工程—电机]