Yield maximization for a catalytic fixed-bed reactor using model predictive control and moving horizon state estimation

This paper presents the implementation of a model predictive controller (MPC) and a moving horizon estimator (MHE) for a catalytic fixed-bed reactor. For this, the nonlinear distributed parameter reactor model is first approximated by a high-dimensional system of ordinary differential equations. The maximization of the yield of the reactor is obtained by a suitable choice of the dynamic optimization problem for the MPC. A tailored gradient method in connection with an efficient numerical integration scheme is used for a solution in real-time of the dynamic optimization problems.