|Table of Contents|

The Structure of Fault tolerant Control System Based on Neural Networks(PDF)


Research Field:
Publishing date:


The Structure of Fault tolerant Control System Based on Neural Networks
LiJun WangZhiquan SunJinsheng
School of Information,NUST,Nanjing 210094
neur al netw orks fault-tolerant technique stability reco nst itut io n
Reliabil ity is an important index to weight the performance of a cont rol system. Fault-to ler ant contr ol is an effect ive means o f r aising the r eliabil ity of a system. Neural netw orks ar e g ood candidates for per forming a variety of reflex ive funct ions in faulttolerant cort ro l system because they ar e no nlinear , can be fault-to ler ant themselves, and can learn f rom exper ience. T his paper analyses the dev elo ping situation and the main proper ties of neural netw orks, and proposes a ty pe of st ructure for fault-tolerant co nt ro l system based o n neural netw orks. The feasibility is studied fr om reliable stabil ity , sy stem reconfigur at ion and integrity design. Neural netw or ks are an ef fective means of performing intelligent fault-to ler ant control.


1 葛建华, 孙优贤. 容错控制系统的分析与综合. 杭州: 浙江大学出版社, 1994
2 Narendra K S, Mukho padhy ay S. Intelligent contr ol using neur al netw or ks. IEEE Control System M agzine, 1992, 12( 3) : 11~18
3 Naidu S R, Zafir io n E. Use o f neur al netw or ks fo r sensor failur e det ect ion in a contr ol sy stem.IEEE Co ntr o l Sy st em Mag zine, 1990, 10( 3) : 49~55
4 Hunt K J, Sbar baro D. Neural netw or ks for contr ol sy stems—— a sur vey . Automat ica, 1992,28( 6) : 1083~1112


Last Update: 2013-04-11