Please use this identifier to cite or link to this item:
Title: Detection, isolation and reconstruction of faulty sensors using principal component analysis
Authors: Bose, M S R K
Kumar, G Sathyendra
Venkateswarlu, Ch
Keywords: Sensor;principal component analysis;controlled CSTR system
Issue Date: Jul-2005
Publisher: CSIR
IPC Code: G01N; G21C17/00
Abstract: A strategy based on principal component analysis (PCA) is presented for detection, identification and reconstruction of faulty sensors. In this strategy, sensor fault detection is carried out by using multivariate statistics, faulty sensors are isolated using principal component score contributions and reconstruction of faulty sensors is accomplished through the analysis of fault direction vector. The performance of the strategy is evaluated by applying to a closed-loop controlled CSTR system. The simulation results demonstrate the ability of the strategy for detection, identification and reconstruction of single and multiple faulty sensors.
Page(s): 430-435
ISSN: 0975-0991 (Online); 0971-457X (Print)
Appears in Collections:IJCT Vol.12(4) [July 2005]

Files in This Item:
File Description SizeFormat 
IJCT 12(4) 430-435.pdf256.97 kBAdobe PDFView/Open

Items in NOPR are protected by copyright, with all rights reserved, unless otherwise indicated.