English

应用科学学报 ›› 2002, Vol. 20 ›› Issue (1): 66-69.

• 论文 • 上一篇    下一篇

一类非线性大系统的间接自适应分散神经网络控制

胡寿松, 张锐   

  1. 南京航空航天大学自动化学院, 江苏南京 210016
  • 收稿日期:2000-06-22 修回日期:2000-11-20 出版日期:2002-03-31 发布日期:2002-03-31
  • 作者简介:胡寿松(1937-),男,浙江慈溪人,教授,博导;张锐(1974-),男,江苏泰州人,博士生.
  • 基金资助:
    国家自然科学基金(69974021);航空科学重点基金(98z51002)资助项目

Indirect Adaptive Decentralized Neurcontrol for a Kind of Nonlinear Large Scale System

HU Shou-song, ZHANG Rui   

  1. AutomationInstitute, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China
  • Received:2000-06-22 Revised:2000-11-20 Online:2002-03-31 Published:2002-03-31

PDF

84

被引次数

3

摘要： 针对仿射非线性大系统,采用分散控制的思想,利用模糊神经网络辨识方法对各个子系统建立NARMA模型,运用自适应控制、滑模控制方法消除各子系统间的非线性关联项及不确定的外部干扰,并使用神经网络的辨识结果来调节控制器参数,形成间接自适应分散神经网络控制器.最后,将本文提出的控制方法运用于互连双倒立摆仿真中,结果表明所提出的算法是有效的.

关键词: 非线性大系统, 分散控制, 神经网络, 自适应控制

Abstract: An indirect adaptive decentralized neurcontrol is proposed for a kind of nonlinear large scale systems by adopting a decentralized control mode. Fuzzy neural network is used to identify the subsystem for building NARMA model and adjusting the parameter of the controller. Besides, the adaptive control and sliding-model-control method is used to extinguish nonlinear interconnection between each subsystems and disturbances with unknown bounds. The indirect adaptive decentralized neurcontrol controller is verified by its application to a double inverted pendulum system, the results of simulation confirm the effectiveness of the proposed algorithm.

Key words: decentralized control, neural network, nonlinear large scale system, adaptive control

中图分类号: