Should I shut down my processing plant? An analysis in the presence of faults. (August 2017)

• Record Type: Journal Article
• Authors: Olivier, L.E.; Craig, I.K.
• Is Part Of: Journal of process control. Volume 56(2017)
• Page Start: 35
• Page End: 47
• Publication Date: 2017-08
• Subjects: Adaptive hypothesis test -- Economic performance -- Fault detection -- Fault tolerant -- Monte Carlo -- Model predictive control -- Nonlinear; Process control -- Periodicals; Fabrication -- Contrôle -- Périodiques; Process control; Periodicals; Electronic journals; 660.281
• DOI: 10.1016/j.jprocont.2017.05.005
• Languages: English
• ISSNs: 0959-1524

Abstract

Abstract : Highlights: Develop method to answer the question: should I shut down my processing plant? First test if the plant can be regulated in the presence of a fault, i.e. is the plant regulatable? Regulatability analysis is done through adaptive hypothesis testing. Then test if it is economically optimal to run the plant in the presence of a fault using economic operability analyses. A fault-tolerant (FT) nonlinear MPC is developed that use the nonlinear generalized likelihood ratio method for fault identification. Abstract: An analysis of the economic operability of a processing plant in the presence of faults is presented in this paper. Once a fault has been identified and diagnosed through a suitable fault detection and isolation algorithm, an adaptive hypothesis test is used to determine if the plant can still be operated. Operation may continue through the use a suitable fault tolerant control scheme. Once this analysis is complete, an economic operability analysis is done to determine whether the plant can still operate economically with the faults present, or whether the plant should be shut down to repair the faults. The overall operating profitability of the process may be maximized through this method. The method is illustrated in this article through simulation of a nonlinear grinding mill circuit controlled by a fault tolerant nonlinear model predictive controller in the presence of a variety of actuator faults.