Carbon Dioxide Adsorption Study on Rice Husk Activated Carbons by Artificial Neural Network (ANN)

Abstract

In this study, the effects of artificial neural networks on CO₂ adsorption on several types of rice husk activated carbon samples are investigated. Using conventional approach, the eight activated carbon samples are examined for carbon dioxide adsorption at 298 K and up to 1 bar pressure. The influence of altered training/validating ratios, various data initiation points, various training algorithms and number of neurons necessary for an artificial neural network model were investigated using ANN modelling. The work can give useful information on the effects of each of the investigated factors, which are crucial in ANN modelling and training techniques. The results may be used to create an optimum activated carbon, improved applications of gas and oil purification that plan to use artificial intelligence modelling in their evaluations.