Nonlinear System Identification Based on Recurrent Wavelet Neural Network

被引：0

作者：

Zhao, Fengyao ^{[1
]}

Hu, Liangming ^{[1
]}

Li, Zongkun ^{[1
]}

机构：

[1] Zhengzhou Univ, Sch Water Conservancy & Environm, Zhengzhou 450002, Peoples R China

来源：

SIXTH INTERNATIONAL SYMPOSIUM ON NEURAL NETWORKS (ISNN 2009) | 2009年 / 56卷

关键词：

Elman network; Recurrent wavelet neural network(RWNN); Extended kalman filter(EKF); Nonlinear dynamical system; Identification;

D O I：

暂无

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Based on Elman network, the recurrent wavelet neural network (RWNN) is presented, and the extended kalman filter of RWNN is given in this paper. The recurrent wavelet neural network (RWNN) can be used in the nonlinear system identification successfully. Practical example shows that RWNN has a faster convergence as well as a better precision in calculation, and a good result on the nonlinear system identification is got, which means it has a broad prospect on application.

引用

页码：517 / 525

页数：9

共 9 条

[1]

[Anonymous], INTERNATIONAL JOINT

[2] FINDING STRUCTURE IN TIME [J].

ELMAN, JL .

COGNITIVE SCIENCE, 1990, 14 (02) :179-211

[3]

LIANG F, 2000, J GUILIN I ELECT TEC, V20, P18

[4]

MATTHEWS MB, 1990, P INT NEUR NETW C PA, V1, P115

[5] Training wavelet networks for nonlinear dynamic input-output modeling [J].

Oussar, Y ;

Rivals, I ;

Personnaz, L ;

Dreyfus, G .

NEUROCOMPUTING, 1998, 20 (1-3) :173-188

[6] NEUROCONTROL OF NONLINEAR DYNAMICAL-SYSTEMS WITH KALMAN FILTER TRAINED RECURRENT NETWORKS [J].

PUSKORIUS, GV ;

FELDKAMP, LA .

IEEE TRANSACTIONS ON NEURAL NETWORKS, 1994, 5 (02) :279-297

[7] New fuzzy wavelet neural networks for system identification and control [J].

Srivastava, S ;

Singh, M ;

Hanmandlu, M ;

Jha, AN .

APPLIED SOFT COMPUTING, 2005, 6 (01) :1-17

[8]

Williams R., 1992, NUCCS921 NE U

[9]

YUAN X, 1998, WATER RESOURCES HYDR, V3, P38

← 1 →