Model-Based Fault Diagnosis and Fault Tolerant Control for Safety-Critical Chemical Reactors: A Case Study of an Exothermic Continuous Stirred-Tank Reactor

In safety-critical chemical reactors with potential hazards,reactionkinetics and heat transfer parameters are usually known, and a mathematicalmodel is available. It is then meaningful to base fault detectionand isolation algorithms on the first-principles model as opposedto statistics, so that physically meaningful residual signals aregenerated from material and/or energy balances not closing, leadingto reliable fault diagnosis. Additionally, to maintain the safetyof the entire system, it is necessary to take appropriate controlaction based on the mathematical model and the identified faults,to minimize their impact and thus ensure safe operation. In the presentwork, these ideas will be formulated and illustrated through a continuousstirred-tank reactor case study involving the liquid-phase oxidationof alkylpyridine with hydrogen peroxide. The proposed fault tolerantcontrol strategy monitors the DSM (distance of the system state fromthe boundary of the dynamic safe set) and the estimate of the faultsize, and when they cross a certain limit as a result of an abnormalevent, the manipulated input is switched. Simulation results showthe effectiveness of the proposed fault tolerant control strategyin dealing with cooling system failure.