COUNTERPROPAGATION NEURAL NETWORKS FOR FAULT-DETECTION AND DIAGNOSIS

Authors
VORA N TAMBE SS KULKARNI BD
Citation
N. Vora et al., COUNTERPROPAGATION NEURAL NETWORKS FOR FAULT-DETECTION AND DIAGNOSIS, Computers & chemical engineering, 21(2), 1997, pp. 177-185
Citations number
26
Categorie Soggetti
Computer Application, Chemistry & Engineering","Engineering, Chemical","Computer Science Interdisciplinary Applications
Journal title
Computers & chemical engineering → ACNP
ISSN journal
00981354
Volume
21
Issue
2
Year of publication
1997
Pages
177 - 185
Database
ISI
SICI code
0098-1354(1997)21:2<177:CNNFFA>2.0.ZU;2-D
Abstract
This paper shows the application of a counterpropagation neural networ k (CPNN) to detect single faults and their magnitudes. The performance of CPNN has been evaluated by considering a variety of faults occurri ng in a nonisothermal continuous stirred tank reactor (CSTR). The resu lts presented here indicate that CPNN provides an attractive alternati ve to error-back-propagation (EBP) networks due to its faster teaming ability for fault detection and diagnosis. Copyright (C) 1996 Elsevier Science Ltd