首页 | 本学科首页   官方微博 | 高级检索  
     

基于动态神经网络的非线性系统建模及其控制
引用本文:李峰,李树荣,刘先广. 基于动态神经网络的非线性系统建模及其控制[J]. 中国石油大学学报(自然科学版), 2002, 26(2)
作者姓名:李峰  李树荣  刘先广
作者单位:石油大学信息与控制工程学院,山东东营,257061
基金项目:国家基础研究发展规划项目 (G19980 3 0 2 0 0 ),山东省自然科学基金资助项目 (Q96G0 2 15 0 )
摘    要:结合已知机理信息构造动态神经网络 ,进行了非线性动态系统的建模 ,给出了权值调整算法。利用获得的模型 ,设计了反馈线性化控制器。由训练好的网络在线提供反馈线性化所需要的信息。为了解决模型失配问题 ,采用内模控制结构来引入模型的误差反馈 ,以消除稳态误差。文中给出了仿真实例。

关 键 词:动态神经网络  非线性系统建模  反馈线性化  内模控制

Nonlinear dynamics modeling and controlling based on dynamic neural network
LI Feng,LI Shu rong and LIU Xian guang. College of Information and Control Engineering in the University of Petroleum,China,Dongying. Nonlinear dynamics modeling and controlling based on dynamic neural network[J]. Journal of China University of Petroleum (Edition of Natural Sciences), 2002, 26(2)
Authors:LI Feng  LI Shu rong  LIU Xian guang. College of Information  Control Engineering in the University of Petroleum  China  Dongying
Affiliation:LI Feng,LI Shu rong and LIU Xian guang. College of Information and Control Engineering in the University of Petroleum,China,Dongying 257061
Abstract:A nonlinear dynamics is modeled using a dynamic neural network in combination with known mechanistic properties of the process. The corresponding algorithm is given. The well trained neural network is used as a system model for feedback linearization in differential geometric control. An internal model control(IMC) structure is utilized to cope with the model plant mismatch, which guarantees the steady output of the plant in offset free. A simulation example shows the effectiveness of the strategy.
Keywords:dynamic neural network  nonlinear system modeling  feedback linearization  internal model control
本文献已被 CNKI 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号