Real-time monitoring for a process with multiple operating modes

Abstract

The real-time monitoring of a chemical process with multiple operating modes is a challenging problem. The frequent changes of operating modes require frequent updates of the monitoring models, which lead to frequent pauses in the real-time monitoring activities. This paper proposes a monitoring methodology for a process with multiple operating modes, based on hierarchical clustering and a super PCA model. The case studies show that the super PCA model performs better than a single PCA model for all operating modes, or local PCA models developed for each operating mode.