Control of nonlinear dynamic systems using recurrent wavelet neural network

摘要：

This paper investigates a recurrent wavelet neural network (RWNN) for control of nonlinear dynamic systems.The RWNN includes the basic ability of the wavelet neural network (WNN) such as fast convergence and localization properties.Besides,the RWNN has a property,unlike the WNN,that it can store the past information of the network.We use the real time recurrent learning (RTRL) algorithm with random initial conditions to train the network.Finally,the RWNN controller is applied to two control problems.Experimental results verify that a favorable tracking response can be achieved by the RWNN controller and the proposed controller is quite effective in control of nonlinear dynamic systems.It is notable that the used controller has simple structure.This reduces the number of wavelet functions and the number of on-line adjustable parameters.

关键词： recurrent wavelet neural network real time recurrent learning nonlinear system control

作者: Afsaneh Ghadirian Maryam Zekri

作者单位: Department of Electrical and Computer Engineering Isfahan University of Technology,IUT Isfahan,Islamic Republic of Iran

会议类型: 国际会议

会议名称: 2010 4th International Conference on Intelligent Information Techonlogy Application(第四届智能信息技术应用国际学术研讨会 IITA 2010)

会议地点: 秦皇岛

会议语种:英文

页码: 60-63

在线出版日期: 2010-11-05（万方平台首次上网日期，不代表论文的发表时间）