电工技术学报  2020, Vol. 35 Issue (2): 266-273    DOI: 10.19595/j.cnki.1000-6753.tces.181656
电机与电器 |
永磁直线伺服系统的自适应改进Elman神经网络积分反推控制
赵希梅, 付东学, 金洋洋
沈阳工业大学电气工程学院 沈阳 110870
Adaptive Modified Elman Neural Network Integral Backstepping Control for Permanent Magnet Linear Servo System
Zhao Ximei, Fu Dongxue, Jin Yangyang
School of Electrical Engineering Shenyang University of Technology Shenyang 110870 China
全文: PDF (34216 KB)   HTML
输出: BibTeX | EndNote (RIS)      
摘要 针对永磁直线同步电动机(PMLSM)伺服系统存在的参数变化、外部扰动和摩擦力等不确定性因素,该文采用了积分反推控制和自适应改进Elman神经网络相结合的控制方案。首先,针对PMLSM伺服系统的非线性特性,利用积分反推控制方法,通过逐步修正算法来设计虚拟控制函数,实现系统的全局调节和位置跟踪;其次,设计自适应改进Elman神经网络来估计系统中存在的不确定性,且利用基于Lyapunov函数的自适应律推导出神经网络的在线参数学习律,使系统具有适应时变特性的能力,克服不确定性对系统的影响,从而提高系统的鲁棒性;最后,实验结果表明所提出的控制方案是有效的,明显提高了系统的跟踪性能和鲁棒性能。
服务
把本文推荐给朋友
加入我的书架
加入引用管理器
E-mail Alert
RSS
作者相关文章
赵希梅
付东学
金洋洋
关键词 永磁直线同步电动机积分反推控制改进Elman神经网络不确定性    
Abstract:The permanent magnet linear synchronous motor (PMLSM) servo system has the uncertainties such as parameter variations, external disturbance and friction. Therefore, a control scheme combining integral backstepping control and adaptive modified Elman neural network was adopted. Firstly, for the nonlinear characteristics of PMLSM servo system, the integral backstepping control method was used to design the virtual control function through the step-by-step modification algorithm to realize the global adjustment and position tracking of the system. Secondly, an adaptive modified Elman neural network was designed to estimate the uncertainty in the system, and the on-line parameter learning law of the neural network was derived by the adaptive law based on Lyapunov function. Thus, the system had the ability to adapt to the time-varying characteristics and overcome the influence of uncertainty on the system, thereby improving the robustness of the system. Finally, the experimental results show that the proposed control scheme is effective, and the tracking performance and robust performance of the system are obviously improved.
Key wordsPermanent magnet liner synchronous motor    integral backstepping control    modified Elman neural network    uncertainties   
收稿日期: 2018-10-22      出版日期: 2020-01-17
PACS: TP273  
  TM351  
基金资助:辽宁省自然科学基金计划重点项目(20170540677)和辽宁省教育厅科学技术研究项目(LQGD2017025)资助
通讯作者: 赵希梅 女,1979年生,教授,博士生导师,研究方向为电机控制、机器人控制、智能控制等。E-mail: zhaoxm_sut@163.com   
作者简介: 付东学 男,1992年生,博士研究生,研究方向为电机控制、智能控制。E-mail: 18629974521@163.com
引用本文:   
赵希梅, 付东学, 金洋洋. 永磁直线伺服系统的自适应改进Elman神经网络积分反推控制[J]. 电工技术学报, 2020, 35(2): 266-273. Zhao Ximei, Fu Dongxue, Jin Yangyang. Adaptive Modified Elman Neural Network Integral Backstepping Control for Permanent Magnet Linear Servo System. Transactions of China Electrotechnical Society, 2020, 35(2): 266-273.
链接本文:  
https://dgjsxb.ces-transaction.com/CN/10.19595/j.cnki.1000-6753.tces.181656          https://dgjsxb.ces-transaction.com/CN/Y2020/V35/I2/266