Artificial Intelligence in Carbon Capture: Modeling CO2 Absorption Using Amine Blends

Abstract

Abstract Carbon dioxide (CO2) from burning fossil fuels and other modernistic industrial processes is the main contributor to the greenhouse gas impact and global warming. The substantial quantities of CO2 produced by this exercise emphasize the pressing need to produce effective processes for removing it from gas products in order to prevent hazardous emissions into the atmosphere. This work an aim is to develop a predictive model for estimating the CO2 absorbed rate in various amine blends. The model was used an artificial neural network (ANN) with three inputs, which was trained on a dataset of 159 experimental data points associated with CO2 absorption by amine blends. Sterically hindered amines, primary amines, tertiary amines, and a few promoters along with physical solvents were among the amine blends that were taken into consideration. Three crucial input variables that were included in the study: Amine temperature (T), Blended amine concentration (AmH) and the Partial pressure of CO2(PCO2 ), and. RCO2 (unit as, kmol.m−2.s−1), or the rate of CO2 absorption, was the model’s target value. 15 neurons were found to be the ideal count of neural network models, and the dataset’s mean square error (MSE) was found to be 21.18. These findings demonstrated how effectively the created ANN model predicted CO2 absorption rates, representing a crucial step in the processes of carbon capture and sequestration.