Dynamic Simulation and Model Predictive Control of Post-Combustion CO2 Capture Process with Advanced Flash Stripper

Abstract

Amine-based post-combustion CO2 capture is an option that is readily applicable to power plants to mitigate their negative environmental impact. However, high operating cost due to significant regeneration energy stands as the main obstacle for its widespread adoption. Advanced flash stripper (AFS) is an intensified process designed to reduce the energy consumption of the conventional stripper. The capture process with AFS can reduce the regeneration energy by about 20%, but the complex heat integration scheme in AFS may reduce its operational flexibility and controllability. According to a previous study on regulatory control of the AFS process, it is revealed that achieving a major control objective (e.g.CO2 capture rate) is highly challenging with decentralized PID controllers. This study aims to suggest an effective control structure for the CO2 capture process with AFS under a number of operational scenarios. For this, a dynamic model for the capture process with AFS using 30wt% Monoethanolamine (MEA) solvent is built in the commercial software platform of gPROMS, to examine dynamic characteristics of the process. Step response analysis is performed using the developed model and the result shows that the process exhibits slow dynamics with strong interactions among its variables, due to the heat integration and recycle. Then, the use of a multivariate control technique like model predictive control (MPC) is considered to handle the complex process dynamics. Closed-loop dynamic behavior with different control structures of MPC and decentralized PID controllers are simulated and compared. Our results indicate that the MPC can give superior control performance, robustness and convenience of operation over the decentralized control, under the various studied scenarios of changes in the inlet flue gas flow rate and CO2 capture rate set-point.