Statistics Pattern Analysis Based Fault Detection and Diagnosis

Statistics pattern analysis (SPA) is a new multivariate statistical monitoring framework proposed by the authors recently. It addresses some challenges that cannot be readily addressed by the commonly used multivariate statistical methods such as principal component analysis (PCA) in monitoring batch processes in the semiconductor industry. It was later extended to the monitoring of continuous processes using a moving window based approach. In this work, we explore the potential of SPA in fault diagnosis. Specifically, we derive variable contributions based on the fault detection indices to generate contribution plots for fault diagnosis. The superior performance of the proposed method is demonstrated using the challenging Tennessee Eastman process (TEP), and compared with the commonly used contribution plots based on PCA and dynamic PCA (DPCA).