东北大学学报:自然科学版 ›› 2014, Vol. 35 ›› Issue (12): 1673-1677.DOI: 10.12068/j.issn.1005-3026.2014.12.001

• 信息与控制 •    下一篇

基于RBF神经网络的非线性磁悬浮系统控制

赵石铁, 高宪文, 车昌杰   

  1. (东北大学 信息科学与工程学院, 辽宁 沈阳110819)
  • 收稿日期:2013-11-20 修回日期:2013-11-20 出版日期:2014-12-15 发布日期:2014-09-12
  • 通讯作者: 赵石铁
  • 作者简介:赵石铁(1977-),男,朝鲜平壤人,东北大学博士研究生; 高宪文(1955-),男,辽宁盘锦人,东北大学教授,博士生导师.
  • 基金资助:
    国家自然科学基金重点项目(61034005).

Control of a Nonlinear Magnetic Levitation System Based RBF Neural Network

ZHAO Shi-tie, GAO Xian-wen, CHE Chang-jie   

  1. School of Information Science & Engineering, Northeastern University, Shenyang 110819, China.
  • Received:2013-11-20 Revised:2013-11-20 Online:2014-12-15 Published:2014-09-12
  • Contact: ZHAO Shi-tie
  • About author:-
  • Supported by:
    -

摘要: 磁悬浮系统是一个典型的不确定、非线性系统.由于磁悬浮系统的复杂性很难建立精确的数学模型,采用RBF神经网络(RBFNN)对非线性磁悬浮系统进行辨识,再根据神经网络自适应控制原理设计了非线性磁悬浮系统的神经网络自适应状态反馈控制器与自适应PID控制器,并利用MATLAB进行了仿真.仿真结果表明,神经网络自适应控制能很好地控制本磁悬浮系统;神经网络自适应控制器对于此非线性磁悬浮系统位置具有良好的控制效果,该控制系统具有较好的稳态特性和控制特性.

关键词: RBF神经网络, 自适应控制, 状态反馈, 磁悬浮系统

Abstract: Magnetic levitation system is a typical nonlinear and uncertain system, because it must be combined with a controller which has good control performance to be applied in various occasions. The identified nonlinear magnetic levitation system using of the Radial Basis Function neural network(RBFNN)was proposed. The neural network adaptive state feedback controller and adaptive PID controller of magnet levitation system was designed based on the neural network adaptive control principle. Besides, a simulation of the system was proposed by using MATLAB, and the result showed that neural network adaptive controller had a good effect on this nonlinear system. In addition,this control system had a preferable stability and control property.

Key words: RBF neural networks, adaptive control, state feedback, magnetic levitation system

中图分类号: