In this paper we develop a fault detection and isolation method based on data-driven approach. Data-driven methods are effective for feature extraction and feature analysis using statistical techniques. In the proposal, the Principal Component Analysis (PCA) method is used to extract the features and to reduce the data dimension. Then, the Kullback-Leibler Divergence (KLD) is used to detect the fault occurrence by comparing the Probability Density Function of the latent scores. The faulty sensor is isolated thanks to a linear combination of the original measurements with binary coefficients denoted as Z-decomposition. The proposed approach is experimentally verified with vibration signals used for monitoring bearings in electrical machines.

Additional Metadata
Keywords Bearing faults, Fault detection, Fault isolation, Kullback-Leibler Divergence, Vibration signals
Persistent URL dx.doi.org/10.1109/SDPC.2017.65
Conference 2017 International Conference on Sensing, Diagnostics, Prognostics, and Control, SDPC 2017
Citation
Delpha, C. (Claude), Diallo, D. (Demba), Wang, T. (Tianzhen), Liu, J, & Li, Z. (Zelig). (2017). Multisensor Fault Detection and Isolation Using Kullback Leibler Divergence: Application to Data Vibration Signals. In Proceedings - 2017 International Conference on Sensing, Diagnostics, Prognostics, and Control, SDPC 2017 (pp. 305–310). doi:10.1109/SDPC.2017.65