MULTIVARIABLE CONTROL OF AN INDUSTRIAL GAS PHASE COPOLYMERIZATION REACTOR

Advanced control is used to avoid large temperature offsets for a batch copolymerization gas phase reactor used for test of catalysts. A nonlinear state-space model is developed to represent the reactor behavior in terms of mass and energy balances. A growth model describes the gas phase copolymerization and includes an activation-deactivation model for the catalyst. The multivariable control system includes three manipulated inputs, three controlled outputs and two measurements, temperature and pressure. The strategy combines nonlinear geometric control laws where states are either estimated by extended Kalman filtering or predicted from the kinetic model. This advanced control has been implemented with success on an industrial reactor.