Application of decision tree learning in modelling CO2 equilibrium absorption in ionic liquids

Over the past few years, many kinds of researches have been devoted to the issue of greenhouse gas (GHG) emissions. Using a solvent-based technology for capturing CO2 as one of the main GHG is a common choice in practical applications. In addition to the well-known organic solvents, ionic liquids (ILs) can be utilized to remove CO2 from atmosphere/industrial streams. This study is aimed at investigating the performance of Classification And Regression Tree (CART) methodology in modeling CO2 solubility in different ILs as a function of system's temperature and pressure as well as the properties of ILs including critical temperature, critical pressure, and acentric factor. To this end, a tree-based model was developed using an extensive database containing more than 5330 experimental data on the solubility of CO2 in 66 ILs. Findings reveal that the proposed model's outcomes are in excellent agreement with the corresponding experimental values. The presented model shows an average absolute relative deviation equal to 0.04% and provides considerably better estimations than the previously published Radial Basis Function Artificial Neural Network (RBF-ANN), Multi-Layer Perceptron (MLP) ANN, Adaptive Neuro-Fuzzy Inference System (ANFIS), and Least Square Support Vector Machine (LSSVM) models.