Dynamic modelling, identification and preliminary control analysis of an amine-based post-combustion CO2 capture pilot plant

Abstract

Solvent-based post combustion CO2 capture (PCC) is considered as a mature technology for dealing with CO2 emissions from fossil-fired power plants. In this study, a mathematical black box model is developed to analyse the dynamic responses of a PCC pilot plant. The model identification reported the dynamics of variables of the key units in the plant, the absorber, rich/lean heat exchanger and desorber. Pilot plant dynamic data were used to develop a data-driven model for each unit operation. Individual models were integrated to produce a simplified 4 × 3 PCC process model of the PCC plant. The fastest dynamic with a time constant ranging from 2 to 3 min featured in the relationship between power plant flue gas flow rate and CO2 concentration in the absorber off gas. Whereas, the slowest response with a process time constant between 9 and 27 min occurred in CO2 concentration at the top of the stripper due to changes in reboiler heat duty. Preliminary control analysis using relative gain array (RGA) analysis suggested that carbon capture efficiency, CC (%), and energy performance, EP (MJ per kg of CO2 captured), can be controlled by manipulating the lean solvent flow rate and reboiler heat duty, respectively. The proposed control structure was tested and tracked CC and EP random set point step changes in the range between 7–25 h and 4–5 h respectively. This study contributes to understanding transient variable behaviours in PCC plants; concurrent with current industrial requirements for controllability and flexible operation of PCC plants, in response to the dynamics of power plant load, electricity and carbon prices.