Neural networks in virtual reference tuning

被引：27

作者：

Esparza, Alicia ^{[1
]}

Sala, Antonio ^{[1
]}

Albertos, Pedro ^{[1
]}

机构：

[1] Univ Politecn Valencia, Dept Syst Engn & Control, Valencia 46022, Spain

来源：

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE | 2011年 / 24卷 / 06期

关键词：

Neural networks; Virtual reference feedback tuning; Back propagation through time; Model reference control; Data-based controller tuning; Direct controller design; CONTROL DESIGN; IDENTIFICATION; CONTROLLER;

D O I：

10.1016/j.engappai.2011.04.003

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 discusses the application of the virtual reference tuning (VRT) techniques to tune neural controllers from batch input-output data, by particularising nonlinear VRT and suitably computing gradients backpropagating in time. The flexibility of gradient computation with neural networks also allows alternative block diagrams with extra inputs to be considered. The neural approach to VRT in a closed-loop setup is compared to the linear VRFT one in a simulated crane example. (C) 2011 Elsevier Ltd. All rights reserved.

引用

页码：983 / 995

页数：13

共 45 条

[1]

Baruch I, 2007, STUD FUZZ SOFT COMP, V208, P95

[2] Least costly identification experiment for control [J].

Bombois, X. ;

Scorletti, G. ;

Gevers, M. ;

Van den Hof, P. M. J. ;

Hildebrand, R. .

AUTOMATICA, 2006, 42 (10) :1651-1662

[3]

Butler H., 1991, IEEE Control Systems Magazine, V11, P57, DOI 10.1109/37.103358

[4] Direct nonlinear control design: The virtual reference feedback tuning (VRFT) approach [J].

Campi, MC ;

Savaresi, SM .

IEEE TRANSACTIONS ON AUTOMATIC CONTROL, 2006, 51 (01) :14-27

[5] An application of the virtual reference feedback tuning method to a benchmark problem [J].

Campi, MC ;

Lecchini, A ;

Savaresi, SM .

EUROPEAN JOURNAL OF CONTROL, 2003, 9 (01) :66-76

[6] Virtual reference feedback tuning: a direct method for the design of feedback controllers [J].

Campi, MC ;

Lecchini, A ;

Savaresi, SM .

AUTOMATICA, 2002, 38 (08) :1337-1346

[7] A new approach for multimodel identification of complex systems based on both neural and fuzzy clustering algorithms [J].

Elfelly, Nesrine ;

Dieulot, Jean-Yves ;

Benrejeb, Mohamed ;

Borne, Pierre .

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE, 2010, 23 (07) :1064-1071

[8]

GEVERS M, 2005, EUROPEAN J CONTROL, V11, P1

[9]

GEVERS M, 1993, JOINT DESIGN IDENTIF, P111

[10] Nonlinear system identification: From multiple-model networks to Gaussian processes [J].

Gregorcic, Gregor ;

Lightbody, Gordon .

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE, 2008, 21 (07) :1035-1055

← 1 2 3 4 5 →