DocumentCode :
2657239
Title :
On identification of partially known dynamic nonlinear systems with neural network
Author :
Brown, Ronald H. ; Ruchti, Timothy L. ; Feng, Xin
Author_Institution :
Dept. of Electr. & Comput. Eng., Marquette Univ., Milwaukee, WI, USA
fYear :
1993
fDate :
25-27 Aug 1993
Firstpage :
499
Lastpage :
504
Abstract :
A method for incorporating a priori information about an uncertain nonlinear system into the structure of a multilayer feedforward artificial neural network is presented. The result is an improved identification model and controller structures suitable for nonlinear system identification and control applications. The known information is incorporated into the activation function of the network output layer. An algorithm is derived for backpropagating the error and updating adjustable parameters within this layer that is consistent with existing supervised learning techniques. The developed technique is applied to the identification of several dynamic systems and compared with existing approaches. The results exhibit a significant improvement in the quality of the identification model and an increase in the rate of convergence. The results further demonstrate that artificial neural networks using a priori information converge faster and more accurately predict the system of interest
Keywords :
backpropagation; feedforward neural nets; identification; nonlinear dynamical systems; uncertain systems; activation function; backpropagation; convergence; dynamic nonlinear systems; feedforward neural nets; identification model; supervised learning; uncertain nonlinear system; Artificial neural networks; Control system synthesis; Control systems; Feedforward neural networks; Multi-layer neural network; Neural networks; Nonlinear control systems; Nonlinear dynamical systems; Nonlinear systems; Supervised learning;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Intelligent Control, 1993., Proceedings of the 1993 IEEE International Symposium on
Conference_Location :
Chicago, IL
ISSN :
2158-9860
Print_ISBN :
0-7803-1206-6
Type :
conf
DOI :
10.1109/ISIC.1993.397663
Filename :
397663
Link To Document :
بازگشت