Prediction of CO2 absorption by nanofluids using artificial neural network modeling

This study focused on providing a model for prediction of CO2 absorption by nanofluids in closed vessel absorber system, for the first time. A 6-input artificial neural network model was presented over 165 extracted experimental data related to CO2 absorption by nanofluids. The used nanofluids were containing spherical nanoparticles of SiO2, Al2O3, Fe3O4 and TiO2 dispersed in water, diethanolamine solution, propylene carbonate and sulfinol as base fluids, respectively. The effective parameters of temperature (T), initial pressure of CO2 (p), time (t), density of nanoparticles (.np), average diameter of nanoparticles (dnp) and mass concentration of nanoparticles (f) were considered as input variables of the network, and the amount of CO2 absorption (a) was chosen as target. The optimal ANN model was obtained in neuron 9. The mean square error (MSE), mean absolute error (MAE), and correlation coefficient (R) were found to be 0.0000236, 0.326 and 0.9996, respectively, for all data. These results showed a good accuracy and performance of developed ANN model in predicting of CO2 absorption.