Online model-based optimization and control for the combined optimal operation and runaway prediction and prevention in (fed-)batch systems. (22nd December 2015)

• Record Type: Journal Article
• Authors: Rossi, Francesco; Copelli, Sabrina; Colombo, Andrea; Pirola, Carlo; Manenti, Flavio
• Is Part Of: Chemical engineering science. Volume 138(2015)
• Page Start: 760
• Page End: 771
• Publication Date: 2015-12-22
• Subjects: Non-linear model predictive control -- Real-time dynamic optimization -- Online runaway prediction and prevention -- Safe and profitable management of (fed-)batch processes; Chemical engineering -- Periodicals; Génie chimique -- Périodiques; Chemical engineering; Periodicals; Electronic journals; 660
• DOI: 10.1016/j.ces.2015.09.006
• Languages: English
• ISSNs: 0009-2509

Abstract

Abstract: An advanced model-based strategy for the online optimization and/or optimal control of (fed-)batch systems, the BSMBO&C, is applied to discontinuous equipment that can be subject to runaway phenomena. The aim is that of showing that it is simultaneously possible to provide profitable online optimization and/or optimal control policies, predict possible future hazardous situations in advance and automatically decide whether and when to (optimally) stop a production cycle in order to prevent safety risks (unavoidable runaways). All these actions are performed online, thus allowing for the effect of any external perturbation. A fed-batch process based on a well-known example of runaway reaction, i.e. the oxidation of 2-octanol to 2-octanone with nitric acid in aqueous solution, is selected as a benchmark to test the real attainability of simultaneous process profitability and safety. Highlights: An advanced model-based strategy for online optimization and optimal control is broached. The strategy is effective to handle batch systems subject to runaway conditions. It allows to provide profitable online optimization and predict potential hazardous situations It stops the batch process in advance when no operational solutions are possible.