Fault detection in processes with multiple operation modes using switch-PCA and analysis of grade transitions

In this paper Principal Components Analysis (PCA) is used for detecting faults in a plant with multiple operation modes. PCA reduces the dimensionality of the original historical data by projecting it onto a lower dimensionality space. It obtains the principal causes of variability in a process. If some of these causes change, it can be due to a fault in the process. The classical PCA approach confuses changes in operation mode with faults. This paper deals with this problem using a switching structure to deal with different operation modes. The transitory states between the operation modes are dealt with using a PCA approach used in batch processes. The implemented method has been proved in a simulated two-communicated-tanks plant.