电工技术学报  2021, Vol. 36 Issue (6): 1229-1237    DOI: 10.19595/j.cnki.1000-6753.tces.191238
电机及其系统 |
基于递归径向基神经网络的永磁直线同步电机智能二阶滑模控制
王天鹤, 赵希梅, 金鸿雁
沈阳工业大学电气工程学院 沈阳 110870
Intelligent Second-Order Sliding Mode Control Based on Recurrent Radial Basis Function Neural Network for Permanent Magnet Linear Synchronous Motor
Wang Tianhe, Zhao Ximei, Jin Hongyan
School of Electrical Engineering Shenyang University of Technology Shenyang 110870 China
全文: PDF (26298 KB)   HTML
输出: BibTeX | EndNote (RIS)      
摘要 针对永磁直线同步电机(PMLSM)易受系统参数变化、外部扰动、摩擦力等不确定性因素影响的问题,采用二阶滑模控制(2OSMC)和递归径向基神经网络(RRBFNN)相结合的智能二阶滑模控制(I2OSMC)方法来提高系统控制性能。利用2OSMC削弱传统滑模控制中的抖振问题,提高了系统的位置跟踪精度。但由于难以估计系统中不确定性因素的边界,从而无法实现2OSMC的最佳性能,因此,引入RRBFNN对不确定性因素进行估计。由于RRBFNN具有较快的学习能力,可通过在线训练网络参数,进而提高系统的鲁棒性。实验结果表明,所提出的控制方法切实可行,能够有效地抑制不确定性因素对系统的影响,使系统具有较高的位置跟踪精度和较强的鲁棒性能。
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
王天鹤
赵希梅
金鸿雁
关键词 永磁直线同步电机不确定性因素二阶滑模控制递归径向基神经网络    
Abstract:The permanent magnet linear synchronous motor (PMLSM) is susceptible to uncertainty factors, such as system parameter variation, external disturbance and friction. Thus, an intelligent second-order sliding mode control (I2OSMC) method combining second-order sliding mode control (2OSMC) and recurrent radial basis function neural network (RRBFNN) is used to improve system control performance. The design of 2OSMC weakens the chattering problem in the traditional sliding mode control and improves the position tracking accuracy of the system. However, because it is difficult to estimate the boundary of the uncertainty factors in the system, the optimal performance of 2OSMC cannot be achieved. Therefore, the RRBFNN is introduced to improve the robustness of the system, which has faster learning ability and can train the network parameters online. The experimental results show that the proposed control method is feasible and can effectively suppress the influence of uncertainty factors on the control system, so that the system has higher position tracking accuracy and stronger robust performance.
Key wordsPermanent magnet linear synchronous motor    uncertainty factors    second-order sliding mode control    recurrent radial basis function neural network   
收稿日期: 2019-09-24     
PACS: TM351  
  TP273  
基金资助:辽宁省自然科学基金计划重点资助项目(20170540677)
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、鲁棒控制等。E-mail: zhaoxm_sut@163.com   
作者简介: 王天鹤 男,1993年生,硕士研究生,研究方向为电机控制、智能控制等。E-mail: wangtianhech@163.com
引用本文:   
王天鹤, 赵希梅, 金鸿雁. 基于递归径向基神经网络的永磁直线同步电机智能二阶滑模控制[J]. 电工技术学报, 2021, 36(6): 1229-1237. Wang Tianhe, Zhao Ximei, Jin Hongyan. Intelligent Second-Order Sliding Mode Control Based on Recurrent Radial Basis Function Neural Network for Permanent Magnet Linear Synchronous Motor. Transactions of China Electrotechnical Society, 2021, 36(6): 1229-1237.
链接本文:  
https://dgjsxb.ces-transaction.com/CN/10.19595/j.cnki.1000-6753.tces.191238          https://dgjsxb.ces-transaction.com/CN/Y2021/V36/I6/1229