Novel Switch Stabilizing Model Predictive Control Strategy Applied in the Control of a Simulated Moving Bed for the Separation of Bi-Naphthol Enantiomers

Abstract

The control of cyclic processes is an open issue in the literature because of their very peculiar dynamic behavior. Thus, deeper studies about this problem are necessary in order to promote an efficient operation of these processes at the industrial level. This work presents a nominally stabilizing MPC controller, also known as infinite horizon model predictive control (IHMPC) applied in the control of a simulated moving bed (SMB) process where bi-naphthol enantiomers are separated. A novel advanced control strategy is presented in the field of SMB process control, while grounds are provided for further developments in the field of cyclic process dynamics and control. The IHMPC performance is evaluated in terms of feasibility and robustness in a realistic nonlinear plant-model mismatch scenario. It is tested in terms of both unmeasured disturbance rejection and conflicting output-tracking scenarios. The results presented indicate that even though the conventional finite horizon-based MPC controller is able to control the process around its optimal point, far from this condition, the conventional MPC loses the process track. On the other hand, the IHMPC controller performed well the process control in all conditions evaluated, far and close to the design conditions. This demonstrates the ability of an IHMPC-like stabilizing control law not only for controlling the system but also for improving the robustness in plant-model mismatch scenarios by considering an infinite prediction horizon.