Hybrid adaptive random search and genetic method for reaction kinetics modelling: CO 2 absorption systems

Carol Toro Molina (1) , Rodrigo Rivera-Tinoco (1) , Chakib Bouallou (1)

1 CEP/Paris

This paper presents a hybrid algorithm to be used for kinetics modelling and based on the Adaptive Random Search Method (ARSM) enhanced by the Genetic Algorithms (GA) theory. This algorithm is validated comparing modelled pressure values to experimental pressure data from Carbonyl sulfide (COS) absorption by aqueous solvents. An excellent fit is shown when COS absorption by N-methyldiethanolamine (MDEA) aqueous solutions is modelled. Preliminary results for the modelling of CO 2 absorption by NH 3 aqueous solutions show an accurate fit and open the door to further studies on more complex reaction mechanisms suggested for the ammonia-carbon dioxide reaction. Besides, the proposed method leads to lower computational time and slightly more accurate calculated pressure values than classic methods such as Downhill simplex.

Mots clés: Absorption system; Adaptive random search; Aqueous solvents; Carbonyl sulfide; Complex reaction mechanism; Computational time; Genetic methods; Hybrid algorithms; Kinetics modelling; Kinetics models; N-methyldiethanolamine; NH 3; Pressure data; Pressure values; Genetic algorithms; Reaction kinetics; Solutions; Carbon dioxide

Carol Toro Molina, Rodrigo Rivera-Tinoco, Chakib Bouallou. Hybrid adaptive random search and genetic method for reaction kinetics modelling: CO 2 absorption systems. Journal of Cleaner Production, 2012, 34, pp.110-115. ⟨10.1016/j.jclepro.2011.11.051⟩. ⟨hal-00769767⟩