A Case Study on Online Estimation of Polystyrene Formation Runaway Reaction Process Parameters. (December 2022)

• Record Type: Journal Article
• Authors: Abburi, Hari Priya; Sengupta, Angan
• Is Part Of: Journal of loss prevention in the process industries. Volume 80(2022)
• Publication Date: 2022-12
• Subjects: Model prediction -- Batch reactor -- Thermal runaway -- Critical time -- Non-linear observer; Chemical industries -- Safety measures -- Periodicals; 660.2804
• DOI: 10.1016/j.jlp.2022.104873
• Languages: English
• ISSNs: 0950-4230

Abstract

Abstract: Batch reactors with polymerisation reaction and ineffective cooling arrangements are prone to thermal runaway. In the present case study, we have investigated such a thermal runaway scenario in case of thermally initiated free radical styrene polymerisation reaction inside adiabatic and non-adiabatic batch reactors. We found that our developed process model for the well-mixed batch reactor along with the reported kinetic model in the literature; aptly describes the thermal runaway polystyrene formation reaction conditions (critical temperature for runaway, Tcr and the critical time for runaway, tcr ). We have also deployed the well-studied Extended Kalman Observer (EKO) to provide online estimates of the runaway reaction process parameters, by using reaction temperature as the only measurement to the observer. It is noted that the non-linear EKO with this one temperature measurement provides accurate estimates of all runaway process parameters from an early time interval. We have also shown that an accurate online estimate of the critical process variables of the polystyrene formation reaction by using the EKO algorithm helps in effective manipulation of initiator concentration that increases the operational safety of the non-adiabatic batch reactors by delaying the attainment of Tcr at a higher monomer conversion. Highlights: Developed process model for batch reactors with available reaction kinetic model, closely resembling experiment data. Use of developed models for model prediction of thermal runaway measures (Tcr and tcr ) in polystyrene polymerisation under adiabatic and non-adiabatic conditions. Use of efficient EKO in conjunction with model prediction to provide early estimates of all process variables that triggers polystyrene runaway reaction. Proposing online estimation technique as effective control strategy to prevent thermal runaway conditions.