A comparative study of neural network structures in identification of nonlinear systems

被引：49

作者：

Efe, MO ^{[1
]}

Kaynak, O ^{[1
]}

机构：

[1] Bogazici Univ, Mechatron Res & Applicat Ctr, TR-80815 Istanbul, Turkey

来源：

MECHATRONICS | 1999年 / 9卷 / 03期

关键词：

D O I：

10.1016/S0957-4158(98)00047-6

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

This paper investigates the identification of nonlinear systems by neural networks. As the identification methods, Feedforward Neural Networks (FNN), Radial Basis Function Neural Networks (RBFNN), Runge-Kutta Neural Networks (RKNN) and Adaptive Neuro Fuzzy Inference Systems (ANFIS) based identification mechanisms are studied and their performances are comparatively evaluated on a three degrees of freedom anthropomorphic robotic manipulator. (C) 1998 Elsevier Science Ltd. All rights reserved.

引用

页码：287 / 300

页数：14

共 8 条

[1]

[Anonymous], 1990, IEEE T NEURAL NETWOR

[2] ON THE APPROXIMATE REALIZATION OF CONTINUOUS-MAPPINGS BY NEURAL NETWORKS [J].

FUNAHASHI, K .

NEURAL NETWORKS, 1989, 2 (03) :183-192

[3] MULTILAYER FEEDFORWARD NETWORKS ARE UNIVERSAL APPROXIMATORS [J].

HORNIK, K ;

STINCHCOMBE, M ;

WHITE, H .

NEURAL NETWORKS, 1989, 2 (05) :359-366

[4]

Jang J.-S.R., 1997, NEUROFUZZY SOFT COMP

[5] NEURO-FUZZY MODELING AND CONTROL [J].

JANG, JSR ;

SUN, CT .

PROCEEDINGS OF THE IEEE, 1995, 83 (03) :378-406

[6] Runge-Kutta neural network for identification of dynamical systems in high accuracy [J].

Wang, YJ ;

Lin, CT .

IEEE TRANSACTIONS ON NEURAL NETWORKS, 1998, 9 (02) :294-307

[7]

[No title captured]

[8]

[No title captured]

← 1 →