首页 | 本学科首页   官方微博 | 高级检索  
     

一种自适应CMAC神经网络控制器的设计与仿真
引用本文:李辉. 一种自适应CMAC神经网络控制器的设计与仿真[J]. 系统仿真学报, 2005, 17(9): 2233-2235,2243
作者姓名:李辉
作者单位:重庆大学高电压与电工新技术教育部重点实验室;重庆大学电气工程学院,重庆,400044
摘    要:为了消除常规前馈型CMAC神经网络控制器的过学习和振荡现象,基于常规CMAC的基础上,提出了一种自适应CMAC神经网络的控制器结构。该控制器以系统动态误差和给定信号量作为CMAC的激励信号,并与自适应神经网络控制器相并联构成系统的复合控制。仿真实例表明,提出的自适应CMAC神经网络控制器具有良好的鲁棒性、抗干扰能力和自适应能力,是解决非线性和不确定性对象控制问题的一种简便有效的控制算法。

关 键 词:CMAC神经网络 自适应控制 仿真 鲁棒性
文章编号:1004-731X(2005)09-2233-03
收稿时间:2004-09-23
修稿时间:2004-09-232005-02-05

Design and Simulation of Adaptive CMAC Controller
LI Hui. Design and Simulation of Adaptive CMAC Controller[J]. Journal of System Simulation, 2005, 17(9): 2233-2235,2243
Authors:LI Hui
Affiliation:The Key Laboratory of High Voltage Engineering and Electrical New Technology, Ministry of Education; College of Electrical Engineering, Chongqing University, Chongqing 400044, China
Abstract:In order to eliminate the excess self-learning and chattering phenomena of the traditional feed-forward cerebellar model articulation controller (CMAC) neural network,based on the conventional structure of CMAC,a self-adaptive CMAC controller is proposed.This cooperative CMAC adopts the dynamic errors and desired values of the controlled plants for the input vectors,which is parallel to the adaptive neuron controller.The simulation results show that this proposed controller is of perfect self-adaptive control properties and robustness against the parameters variation and the external disturbance.It is a simple convenient and effective control scheme for the control of nonlinear and uncertain system.
Keywords:CMAC neural network   self-adaptive control   simulation   robustness
本文献已被 CNKI 维普 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