Multivariate Statistical Analysis for Detection and Identification of Faulty Sensors Using Latent Variable Methods
In this paper, three latent variable methods are implemented in a multivariate statistical analysis scheme for detecting and identifying faults in a multi-sensor network. The proposed methods are applied to a sensor network monitoring a MDOF dynamical system and the results are presented, compared and discussed.
Pietro VINCENZINI and Fabio CASCIATI
M. R. Hernandez-Garcia and S. F. Masri, "Multivariate Statistical Analysis for Detection and Identification of Faulty Sensors Using Latent Variable Methods", Advances in Science and Technology, Vol. 56, pp. 501-507, 2008