An Easy to Interpret Fault Detection Approach to Multivariable Statistical Process Monitoring

Chemical process plant safety, production specifications, environmental regulations, operational constraints and plant economics are some of the main reasons driving an upward interest in research and development of more methods for process monitoring and control. Although there have been a large number of industrial applications of Multivariable Statistical Process Control (MSPC) reported in the literature, there have been far fewer documented cases where MSPC systems have been applied in stored data with their results interpreted by plant operators, rather than MSPC experts. In this paper an easy to interpret method to diagnose process fault is proposed. The method uses the concepts of normal region and benefits from the data projection capability of principal component analysis. A two CSTRs case study is invoked to show the effectiveness of the proposed approach.

Keywords:

Multivariable Statistical Process Monitoring (MSPC) , Principal Component Analysis (PCA) , Fault Detection , CSTR