A Process Monitoring Scheme for Uneven-Duration Batch Process Based on Sequential Moving Principal Component Analysis

The batch process contains complicated production mechanism, which has posed a huge challenge for condition monitoring. Motivated by the problems of multiphase (MP) partition in time sequence, uneven-duration batches with out-of-sync trajectories, this brief presents a novel process monitoring scheme of MP partition, modular modeling, and online monitoring. The sequential moving principal component analysis (SMPCA) is first proposed to perform the MP partition of uneven-duration batches in time sequence. The update of the SMPCA exactly explains the dynamic MP characteristics of the sampling data and the immediacy of the local SMPCA models. Moreover, the essential trend of process change is also explained by extracting the local SMPCA models' feature space and establishing the similarity evolution index for critical variables. Subsequently, the modular modeling based on the subphase partition is conducted to solve the modeling problem of uneven-duration batches with out-of-sync trajectories. Then, a monitoring technique including the differentiated monitoring and secondary monitoring is introduced to further effectively abate fault alarm rate of the monitoring. The performance and advantage of the process monitoring scheme proposed are explained through a typical case and comparative experimental analysis.