Process and control loop performance monitoring through detection of abrupt parameter changes

Abstract

The paper is concerned with process and controller loop performance monitoring through model validation and detection of abrupt changes of model parameters. The local approach for detection of abrupt changes is adopted in the paper as a computational framework for the change detection. This approach has the ability to detect small changes quite effectively. To enhance its robustness in the presence of changing disturbance dynamics, a local approach based on the output error model is proposed. The effectiveness and robustness of the proposed method are demonstrated through Monte Carlo simulations.