DocumentCode :
3638871
Title :
Nonlinear system control based on Neural Networks with Adaptive Predictive strategy
Author :
Mikel Larrea;Eloy Irigoyen;Vicente Gómez;Fernando Artaza
Author_Institution :
Department of Systems Engineering and Automatic Control, Intelligent Control Research Group, University of the Basque Country (UPV/EHU), ETSI, 48013 Bilbao, Spain
fYear :
2010
Firstpage :
1
Lastpage :
7
Abstract :
This paper presents an Adaptive Predictive Control strategy based on Neural Networks for nonlinear systems. In order to train the Neural Network controller, an identification of the system is carried out by the Neural Network Identifier. This second Neural Network provides the training terms related to the nonlinear system dynamics. In this way it is possible to train the Neural Network controller online. The simulation results show a correct online adaptation of the NN controller and the validity of the proposed control strategy.
Keywords :
"Artificial neural networks","Nonlinear systems","Training","Mathematical model","Control systems","Equations","Adaptation model"
Publisher :
ieee
Conference_Titel :
Emerging Technologies and Factory Automation (ETFA), 2010 IEEE Conference on
ISSN :
1946-0740
Print_ISBN :
978-1-4244-6848-5
Electronic_ISBN :
1946-0759
Type :
conf
DOI :
10.1109/ETFA.2010.5641348
Filename :
5641348
Link To Document :
بازگشت