Monitoring Chemical Processes For Early Fault Detection Using Multivariate Methods

By Dr Frank Westad, Chief Scientific Officer, CAMO Software

Multivariate Statistical Process Monitoring (MSPM) has been established as a valuable tool for ensuring reliable product quality in the process industry. However, many organizations today are still not fully utilizing its potential to make significant improvements in their production environment. The MSPM approach to process monitoring involves the use of multivariate models to simultaneously capture the information from as few as two process variables, up to thousands. The methodology provides means for increased process understanding, fault detection and on-line prediction, all typical tasks for the process engineer and production manager.

With MSPM approaches, it is possible to not only control the final product quality data, but also all of the available process variable data in terms of the underlying systematic variations in the process. The variables measured in a process are often correlated to a certain extent, e.g. when several tempertures are measured in a distillation column. This means that the events or changes in a process can be visualized in a smaller subspace that may give a direct chemical or physical interpretation. If we want to keep such a process ”in control”, traditional univariate control charts – due to the covariance or interaction between variables - may not assure this efficiently.