基于TSK型递归模糊神经网络的永磁直线同步电机位置控制研究  

Position control of PMLSM based on TSK-type recurrent fuzzy neural network

作　　者：熊渊琳[1] 方宝英[2] XIONG Yuan-lin;FANG Bao-ying(Electrical Engineering Department, Jiangsu Maritime Institute,Nanjing 211170,China;School of Optical-Electrical and Computer Engineering,University of Shanghai forScience and Technology, Shanghai 200093,China)

机构地区：[1]江苏海事职业技术学院电气学院,江苏南京211170 [2]上海理工大学光电信息与计算机工程学院,上海200093

出　　处：《机电工程》2019年第4期413-417,共5页Journal of Mechanical & Electrical Engineering

基　　金：国网江苏电力公司科技项目(5210EF17001A);江苏省科技攻关资助项目(12010203036);江苏省教育科学"十二五"规划课题资助项目(B-b/2015/03/076)

摘　　要：针对基于磁场定向控制的永磁直线同步电机(PMLSM)伺服系统的位置精准控制问题,提出了一种TSK型递归模糊神经网络(TSKRFNN)控制方法。在考虑了系统易受参数变化、外部扰动和摩擦力等不确定性因素影响的基础上,建立了含有不确定性因素在内的PMLSM动态数学模型;利用TSKFRNN对系统同时进行了实时在线的结构学习和参数学习,提高了系统抑制不确定性因素的鲁棒性,保证了系统的动态跟踪性能。实验及研究结果表明:与模糊神经网络PID控制方法相比,TSKFRNN可以有效辨识电机参数,抑制系统的不确定性对系统伺服性能的影响,提高了系统的鲁棒性和跟踪性能。In order to solve the problem of position precision control of permanent magnet linear synchronous motor (PMLSM) servo system based on field-oriented control, a TSK-type recurrent fuzzy neural network (TSKRFNN) control method was proposed. Considering that system was susceptible to uncertainties such as parameter changes, external disturbances and frictions, a PMLSM dynamic mathematical model with uncertainties was established.The TSKRFNN was used to do structure learning and parameter learning of the system at the same time.The system could be automatically increased the neuron resistance to external disturbance and improved the robustness of the system,ensured the dynamic performance of the system.Experimental results show that,compared with the fuzzy neural network type PID, the proposed method can identify the parameters of PMLSM, suppress uncertainties of the system and improve the robust performance and tracking performance of the system effectively.

关 键 词：永磁直线同步电动机 不确定性因素 TSK型递归模糊神经网络 鲁棒性 跟踪性 

分 类 号：TM301.2[电气工程—电机] TP273[自动化与计算机技术—检测技术与自动化装置]