Statistical Process Monitoring with External Analysis and Independent Component Analysis

Abstract

Extracting essential variables, which drive a process, from measured variables and monitoring them will improve the process monitoring performance. In the present work, a new multivariate statistical process control (MSPC) method based on independent component analysis (ICA) is proposed for realizing this concept. In addition, to cope with changes in operating conditions such as load changes, SPC techniques are integrated with external analysis. The simulation results have clearly shown the superiority of the procesed ICA-based MSPC over univariate SPC and conventional MSPC using principal component analysis (PCA) and also the usefulness of external analyisis.