Multivariate statistical process monitoring and fault diagnosis based on an integration method of PCA-ICA and CSM

摘要

In this paper, an approach for multivariate statistical process monitoring and fault diagnosis based on an improved independent component analysis (ICA) and continuous string matching (CSM) is presented, which can detect and diagnose process fault faster and with higher confidence level. The trial on the Tennessee Eastman process demonstrates that the proposed method can diagnose the fault effectively. Comparison of the method with the well established principal component analysis is also made.