Fault diagnosis based on black-box models with application to a liquid-level system

Author

Palma, L.B. ; Coito, F.V. ; Silva, R.N.

Author_Institution

Dept. de Engenharia Electrotecnica, Univ. Nova de Lisboa, Monte De Caparica, Portugal

Volume

2

fYear

2003

fDate

16-19 Sept. 2003

Firstpage

739

Abstract

This paper proposes an on-line robust approach to fault detection and isolation (FDI) of dynamic systems. This FDI approach is based on black-box models: artificial neural networks (ANNs) and the autoregressive with exogenous input (ARX) models. ANNs are used as observers and pattern classifiers, and adaptive ARX models are used as observers. The generalized likelihood ratio (GLR) algorithm is used for change detection. Process faults are considered, and the robust FDI problem is also addressed. The approach is applied to a laboratory set-up tank system under closed-loop control.

Keywords

autoregressive processes; closed loop systems; fault diagnosis; neural nets; observers; pattern classification; artificial neural networks; autoregressive with exogenous input models; black-box models; closed-loop control; dynamic system; fault detection and isolation; fault diagnosis; generalized likelihood ratio algorithm; laboratory set-up tank system; liquid-level system; Automatic control; Control systems; Fault detection; Fault diagnosis; Isolation technology; Manufacturing automation; Mathematical model; Neural networks; Power system modeling; Robustness;

fLanguage

English

Publisher

ieee

Conference_Titel

Emerging Technologies and Factory Automation, 2003. Proceedings. ETFA '03. IEEE Conference

Print_ISBN

0-7803-7937-3

Type

conf

DOI

10.1109/ETFA.2003.1248772

Filename

1248772