Evaluation of CO₂ Absorption by Amino Acid Salt Aqueous Solution Using Hybrid Soft Computing Methods

Area of Science:

• Chemical Engineering
• Computational Chemistry
• Environmental Science

Background:

• Amino acid salt (AAs) aqueous solutions are emerging as effective media for carbon dioxide (CO2) absorption.
• Efficiently predicting CO2 absorption capacity is crucial for optimizing capture technologies.

Purpose of the Study:

• To develop and evaluate hybrid machine learning models for predicting CO2 loading capacity in AAs aqueous solutions.
• To compare the performance of different machine learning approaches for this prediction task.

Main Methods:

• Developed four hybrid machine learning models: least squares support vector machine with simulated annealing, Radial Basis Function Neural Network (RBF-NN), particle swarm optimization-adaptive neuro-fuzzy inference system, and hybrid adaptive neuro-fuzzy inference system.
• Utilized a dataset of 626 CO2 and AAs equilibrium data points.
• Input features included CO2 partial pressure, temperature, AAs concentration, and molecular descriptors of AAs.

Main Results:

• The developed models accurately predicted CO2 loading capacities.
• Radial Basis Function Neural Network (RBF-NN) demonstrated superior performance compared to the other evaluated models.
• Graphical and statistical analyses confirmed the accuracy and reliability of the model predictions.

Conclusions:

• Hybrid machine learning models, especially RBF-NN, offer a robust and accurate method for predicting CO2 absorption in AAs aqueous solutions.
• These findings can aid in the design and optimization of CO2 capture processes using AAs.