FTC strategies in model predictive control of a dearomatisation process

Abstract

The behavior of an industrial dearomatization process operated under model predictive control (MPC) is studied in the presence of incipient faults. Different automatic fault-tolerant control (FTC) strategies are studied utilizing both a dynamic chemical engineering process simulator and the process MPC. The focus is on FTC strategy directionality resulting from incipient faults such as drifting, and on management of fault consequence risks such as loss of production. The directionality, early detection, and detection reliability aspects in applying appropriate real-time FTC strategies are particularly pronounced in this process and economy environment. The inherent accommodation properties and model information in the studied MPC can readily be exploited as a vehicle to realize the types of FTC strategies required by the users.