Commercial-scale Reactor Research Articles

On-Line Process Monitoring, Control and Supervision for an Industrial Polymerization Process

In industrial processes many key process variables (like conversions, selectivi ties, concentrations, viscosities etc.) can not be measured adequately (accurately and continuously). This paper presents the application of a Kalman filter to obtain continuous monitoring of such a key variable, which was otherwise not continuously measurable. A kinetic process model was developed and corrected with plant data information to estimate the conversion of the polymerization of butadiene into polybutadiene for a commercial-scale reactor. After tuning and testing for robustness a statistical evaluation of the estimator output has been performed, with a positive result. Given the estimator performance the possible applications range from monitoring and control to supervision of the process cooling margin. The applications ultimately chosen will depend on the estimator performance during a period of passive monitoring of both conversion and cooling margin.