On the MPC of the Methanol Synthesis in a Simulated Moving Bed

Abstract

This paper is focused on the design of a Model Predictive Control (MPC) algorithm to control and optimize the methanol synthesis in a Simulated Moving Bed (SMB) reactor. First, the advantages that can be obtained when the process in carried out in this reactor configuration are summarized; then, the control algorithm is described. A simplified model based on Artificial Neural Networks (ANN) is used to calculate the control actions. The influence of the tuning parameters of the algorithm is studied by means of mathematical simulation, thus resulting in the best controller configuration. Finally, examples of the performance of the controlled system are provided, thus demonstrating the effectiveness of the proposed tool.