Sensor fault detection based on principal component analysis for interval-valued data

ABSTRACT

Principal component analysis (PCA)-based fault detection and isolation (FDI) is a well-established data-driven diagnosis strategy that has long been praised for its performances. However, it is still not optimal for uncertain systems, mainly since the model uncertainties usually have a significant effect on the reliability of the method. As an alternative solution, modeling with PCA for interval-valued data ensures a better monitoring by apprehending the sensor uncertainties and including them in the modeling phase. This article presents an extension of data-driven PCA fault detection to the case of interval-valued data. The PCA model is built based on the complete information principal component analysis (CIPCA) for interval-valued data, and different fault detection indices are generated based on the squared prediction error (SPE) statistic. A fault detection scheme is proposed based on squared interval norm of residuals vector. The performances of the proposed fault detection scheme are illustrated using a simulation example and a milling machine process, along with a Monte Carlo experiment for validation.

KEYWORDS:

• complete-information PCA
• fault detection
• interval-valued data
• milling machine
• principal component analysis
• SPE statistic

About the authors

Tarek Ait-Izem is a Phd student in automatics and signals at the Badji-Mokhtar Annaba university, Algeria. While conducting the research that led to this article, he was a member of the Laboratory of Automatics and Signals of Annaba (LASA), and was working in collaboration with PRISME laboratory at the INSA-CVL, Bourges, France. His research interests include process modeling and monitoring, multivariate statistics, with a particular focus on PCA based diagnosis for systems subject to uncertainties using interval methods.

M.-Faouzi Harkat received his Eng. degree in automation from Annaba University, Algeria in 1996, his Ph.D. degree from Institut National Polytechnique de Lorraine (INPL), France in 2003. He is now Professor in the Department of Electronic at Annaba University, Algeria. His research interests include fault diagnosis, process modelling and monitoring, multivariate statistical approaches and neural networks.

Messaoud Djeghaba is a Professor with the Laboratory of Automatics and Signals of Annaba (LASA),at the Department of Electronic at Annaba University, Algeria. His research interests include reliability engineering and fault diagnosis.

Frédéric Kratz received the graduate degree in physics from [cole Nationale Suprieure de Physique de Strasbourg], Strasbourg, France, in 1988, the Ph.D degree in automatic control from the Nancy-University, France, in 1991 and French [Habilitation Diriger des Recherches] in electrical engineering from National Polytechnic Institute of Lorraine, Nancy, in 1998. He is currently a professor with PRISME laboratory at the INSA-CVL, Bourges, France. His current research interests include reliability engineering, stochastic degradation modeling and FDI studies in presence of uncertainties.