Modeling and Control of CO₂ Capture Plant using Model Predictive Control

Abstract

The increasing concentration of atmospheric carbon dioxide (CO₂) necessitates the development of efficient carbon capture technologies to mitigate climate change. Post-combustion capture using Monoethanolamine (MEA) is a mature technology; however, its operational efficiency is hindered by process nonlinearities and multivariable interactions. This research focuses on enhancing the control of a CO₂ absorption column by implementing a Model Predictive Control (MPC) strategy. A dynamic process model was first developed and validated using Aspen Plus Dynamics. Subsequently, a linear state-space model was derived from this high-fidelity model through system identification techniques. The performance of the MPC was then benchmarked against a conventional Proportional-Integral (PI) control scheme in both setpoint tracking and disturbance rejection scenarios. The results demonstrate that the MPC provides superior control, exhibiting faster settling times, minimal overshoot, and more effective disturbance rejection compared to the PI controller. The enhanced performance, quantified by a lower Integral Squared Error (ISE), highlights the capability of MPC to handle complex process dynamics and improve the stability and efficiency of CO₂ capture plants.