Principal component analysis (PCA) based fault detection method and experimental applications

The fault detection based upon multivariate statistical projection method (such as principal component analysis, PCA) have attracted more and more interest in academic research and engineering practice. PCA methods for fault detection use data collected from a steady-state process to monitor T² and Q statistics with a calculated control limit. In this paper, PCA and statistical control chart (SCC) have been used to detect process operating sensor and actuator faults on an electromechanical system. Hotelling, T², statistic is used calculating the control limits of SCC. Experimental results indicate that the method is effective and available.