Validation of Signals Using Principal Component Analysis

Today’s process control industry, which is extensively automated, generates huge amounts of process data from the sensors used to monitor the processes. These data if effectively analyzed and interpreted can give a clear picture of the performance of the underlying process and can be used for its proactive monitoring. With the great advancements in computing systems, a new genre of process monitoring and fault detection systems are being developed which are essentially data-driven. The objective of this research is to examine the effectiveness of Principal Component Analysis (PCA) in validation of the signals from a given topology. PCA is applied on the set of steady-state signals to obtain a model that represents the algebraic relationships among the variables. The model so developed is used for the generation of residuals which further help in Fault Detection and Isolation (FDI). A Global Test (GT) statistic is used to detect the presence of anomalies in the data. The scheme is so tuned that the scheme should indicate no fault when the steady state is not altered and the presence of fault when the GT statistics exceeds the threshold. The scheme is applied on data of 5-bus voltages in a IEEE 5-bus system and the effectiveness is investigated.