Economic MPC of deep cone thickeners in coal beneficiation

Abstract

The thickener is an important operating unit in a coal handling and preparation plant (CHPP). The optimal control and operation of the thickener is critical for the efficiency and economics of a CHPP. In this work, we apply a computationally efficient economic model predictive controller (EMPC) to a deep cone thickener and compare its performance to a proportional‐integral controller and a regular model predictive controller (MPC). Based on a detailed model of the deep cone thickener that is appropriate for controller design purpose, the three control designs are compared on different aspects, including average water recovery rate and controller execution time, via extensive simulations. The robustness of the EMPC is also investigated in the presence of random disturbances in feed flow rate. The results demonstrate that the EMPC has the potential to significantly improve water recovery rate compared to the proportional‐integral control and the regular MPC.