Machine learning models for predicting characteristics of PVAm membranes for post-combustion CO۲ capture application

Facilitated transport membranes fabricated by Polyvinylamine show great potential to competewith convenient carbon capture technologies in a post-combustion CO۲/N۲ separation, with lowerenergy consumption and zero toxicity. Precise mathematical models are needed to predict membranecharacteristics for designing suitable membrane equipment and optimizing process configuration. Twomain features of a membrane are the Permeance of CO۲ gas and CO۲/N۲ Selectivity that shows theoverall performance of each membrane by considering the solution-diffusion model. Two knownmachine learning algorithms were employed to predict the Permeance and Selectivity of a recentlydeveloped membrane based on its four major parameters. Both MLP-ANN and SVM functions had greatpotential to fit experimental data, while the MLP-ANN method works better for Permeance (R۲ equal to۰.۹۷۵) and the SVM method fits Selectivity better (R۲ equal to ۰.۹۴۸).

کلیدواژه ها:

CO۲ Capture ، Polyvinylamine Membrane ، Machine Learning ، Support vector machine ، Artificial neural network