2002 Volume 38 Issue 11 Pages 958-965
A chemical process has a large number of measured variables, but it is usually driven by fewer essential variables, which may or may not be measured. Extracting such essential 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, the proposed ICA-based MSPC is integrated with external analysis. The influence of changes in external variables, which represent operating conditions, can be removed from operation data by conducting external analysis. To investigate the feasibility of the proposed ICA-based MSPC and external analysis, the fault detection performance of ICA-based MSPC is evaluated and compared with that of univariate SPC and conventional MSPC using principal component analysis (PCA) by applying those methods to several monitoring problems. The simulation results have clearly shown the superiority of the proposed ICA-based MSPC over the other SPC methods and also the usefulness of external analysis.