Fabrication modeling of industrial CO2 ionic liquids absorber by artificial neural networks

The fabrication of industrial CO2 blended solution absorber was modeled by artificial neutral network. First the generated model had been statistically evaluated and then its ability of prediction was confirmed by validation test. The validated model was used to predict the desirable density and relative importance of the fabrication's effective variables. In conclusion, the importance included x[H2O], 36.18%, x[gua], 25.37%, x[MDEA], 25.34% and temperature, 13.11% which showed none of them is negligible as well as the density (g cm(-3)) was validated by further experiment that showed the actual density, 1.101, was quite close to the predicted value, 1.017. (C) 2014 The Korean Society of Industrial and Engineering Chemistry. Published by Elsevier B.V. All rights reserved.