A robust NMPC scheme for semi-batch polymerization reactors

Abstract

A robust nonlinear model predictive control (NMPC) scheme is proposed for batch processes with multiple types of uncertainties. Recently, economic MPC (eMPC) has attracted significant attention, particularly for batch process control given its flexibility in the cost function while addressing the nonlinear constrained multivariable dynamics seen in most batch processes. However, in the presence of various uncertainties such as parameter errors, external disturbances, and noise, performance of eMPC can deteriorate significantly as it tends to drive the system to limits of constraints. To achieve constraint satisfaction in the presence of common uncertainties, we propose a robust NMPC method based on multistage scenarios, state estimation, and back-off constraints. Performance of the proposed robust NMPC scheme is evaluated through an example of anionic propylene oxide polymerization reactor.