Application of three artificial intelligence techniques for operational problem solving in a CO2 capture process system

A good understanding of the key process parameters and their intricate relationships is critical for improving effectiveness and efficiency of the post combustion CO₂ capture process. The knowledge can help operators with prediction, control and decision-making. Although some critical parameters of the CO₂ capture process, such as reboiler heat duty, have been discussed in the previous research, their significances of influence and the nature of their relationships that affects efficiency of the CO₂ capture processes are not studied. This paper presents a study on exploring the key parameters of the amine-based post combustion CO₂ capture process system at the International Test Centre of CO₂ Capture (ITC) located in Regina, Saskatchewan of Canada. Three artificial intelligence (AI) techniques of sensitivity analysis (SA), artificial neural network (ANN), and neuro-fuzzy modeling were applied for modeling the historical data to identify the relationships among the key parameters. The knowledge obtained in this data modeling study can be useful for tackling the challenges in operation of the process system.