Advancing Control Paradigms in Heat-Integrated Distillation Columns: An MPC Perspective

This paper presents a Model Predictive Control (MPC) strategy to control the product compositions within a Heat-Integrated Distillation Column (HIDiC) for separating Benzene and Toluene. The controller was designed to maintain Benzene and Toluene compositions at desired setpoints of 99.5% and 0.5% respectively. The MPC controller is compared with a proportional integral (PI) controller tuned using the loop shaping method. Mean absolute error (MAE) and integral absolute error (IAE) were employed as performance indexes. The findings of this study unequivocally demonstrate the superior performance of the MPC controller over the traditional PI controller. A precise setpoint tracking with insignificant overshot was achieved for both the top and bottom compositions as is evident in the MAE and IAE values. Robust control and high purity separation was achieved, demonstrating potential for industrial implementations.