Statistical Process Monitoring and Disturbance Isolation in Multivariate Continuous Processes

Quick detection of out-of-control status and diagnosis of disturbances leading to the abnormal process operation are crucial in minimizing product quality variation. Multivariate statistical techniques are used in developing methodology for detection of abnormal process behavior and diagnosis of disturbances causing poor process performance. The methodology is illustrated by monitoring the Tennessee Eastman plant simulation benchmark problem. Twenty-one different disturbances are introduced to the plant. Most of the disturbances can be diagnosed correctly, the success rate being higher for step and ramp disturbances than noise disturbances.