Data mining applications for finding golden batch benchmarks and optimizing batch process control

Abstract

This article discusses MPCA (Multi-way Principal Component Analysis) and MPLS (Multi-way Partial Least Squares) have been used to compress the information into low-dimensional spaces and pinpoint the root causes of batch-to-batch difference. From engineering perspective, this paper focuses on applying MPCA and MPLS to data analysis of batch process combining with operation experiences to find “golden batch benchmark” that describes the best operation of historical batches. This work includes data pre-treatment, batch process modelling and chemical reaction initiation status decision. Finally, optimizing control strategies and batch process improvement are also been discussed. Process and control engineers are be able to obtain the valuable data analyzing and control optimization methods for batch process from this study.