The Application of a Model Predictive Control-Based Approach to CO2 Capture Processes: Towards Operational Cost Minimisation

Tuesday, October 18, 2011: 2:35 PM
209 A/B (Minneapolis Convention Center)
Alicia Arce, MATGAS Research Center, Barcelona, Spain, Niall Mac Dowell, Chemical Engineering, Imperial College London, London, United Kingdom, Nilay Shah, Centre for Process Systems Engineering, Department of Chemical Engineering, Imperial College London, London, United Kingdom and Lourdes F. Vega, MATGAS Research Center and Carburos Metálicos, Air Products Group, Barcelona, Spain

Amine-based chemisorption of CO2 is a promising near term option for the decarbonisation of large, fixed-point CO2 emission sources[1]. However, the operational expenditure (OPEX) associated with this technology imposes a significant energy penalty on the power plant. As it is the solvent which determines the thermodynamic and kinetic efficiency of the process, the design of advanced solvents provides a real opportunity to reduce the OPEX. Typically, when one refers to the dynamic operation of CCS-type processes, it is the transient behaviour of the start-up and shut-down of this process which is considered. However, the so-called steady-state operation of power-plants is itself dynamic, i.e., the flowrate, temperature and composition of the inlet flue-gas can vary in real time. Thus, a given solvent and mode of process operation which may be optimal for a particular operating regime, may be sub-optimal for another operating regime. Consequently, the implementation of advanced control strategies present an important opportunity for the intensification of these processes resulting in a significant reduction in the lifetime operational expenditure associated with CCS systems.

This work focuses on the design of an explicit model predictive controller (MPC)[2] for the real-time control of solvent composition and process operation as successfully implemented in other applications such as [3].To achieve this, we integrate molecular-based fluid theories[4] with high-fidelity process models[5]. Based on this model, an approximate model is developed and a model predictive controller is formulated for the reduced model where our multi-parametric algorithms are applied to derive a suitable and robust explicit MPC controller. By incorporating the explicit controller expressions (control laws) in the original model, a validation step is then carried out.

In this way, we present a unified-systems based methodology for the OPEX reduction of solvent-based post-combustion CO2 capture processes.


[1] IPCC, 2005: IPCC Special Report on Carbon Dioxide Capture and Storage. Prepared by Working Group III of the IPCC, Cambridge University Press, Cambridge, United Kingdom and New York, USA

[2] Pistikopoulos, E.N., Bozinis, N., Dua, V., Perkins, J. & Sakizlis, V. (2004). Improved Process Control, European Patent EP1399784

[3] Arce A., del Real A. J., Bordons C. and  Ramírez D. R. IEEE Transactions on Industrial Electronics, 57(6), 1892-1905, 2010.

[4] Mac Dowell, N. et al., Ind. Eng. Chem. Res., 49(4), 1883-1899, 2010

[5] Mac Dowell, N. et al., ESCAPE 20, 2010

Extended Abstract: File Not Uploaded
See more of this Session: CO2 Capture, Control and Sequestration
See more of this Group/Topical: Sustainable Engineering Forum