Premium
Modeling Equilibrium Systems of Amine‐Based CO 2 Capture by Implementing Machine Learning Approaches
Author(s) -
Ghiasi Mohammad M.,
AbediFarizhendi Saeid,
Mohammadi Amir H.
Publication year - 2019
Publication title -
environmental progress and sustainable energy
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.495
H-Index - 66
eISSN - 1944-7450
pISSN - 1944-7442
DOI - 10.1002/ep.13160
Subject(s) - diethanolamine , triethanolamine , amine gas treating , solubility , adaboost , carbon dioxide , artificial neural network , computer science , co2 removal , chemistry , artificial intelligence , biological system , machine learning , analytical chemistry (journal) , support vector machine , chromatography , organic chemistry , biology
Precise calculation of carbon dioxide equilibrium solubility in aqueous amine solutions is decisive in the success of establishment or maintenance of amine‐based absorptive carbon dioxide capture processes. To implement the AdaBoost algorithm in conjunction with the classification and regression tree (AdaBoost‐CART) aimed at developing models to accurately estimate the equilibrium absorption of carbon dioxide in ethanolamine solutions, experimental data for monoethanolamine (MEA), diethanolamine (DEA), and triethanolamine (TEA) systems were gathered from the literature. Furthermore, neural‐based models were developed using the collected databank as the basis of comparison. The results of the presented models were compared to the results of the available models in the literature. It was found that the proposed AdaBoost‐CART models for the investigated amine systems present more precise and reliable outputs compared to the results of the neural‐based and literature models. In a respective order, the introduced AdaBoost‐CART models for MEA, DEA, and TEA solutions show average absolute relative deviation percent of 0.51, 2.76, and 1.41 which indicate their reliability and superiority over other models. © 2019 American Institute of Chemical Engineers Environ Prog, 38:e13146, 2019