Mathematical modeling and optimization of gas transport through carbon molecular sieve membrane and determining the model parameters using genetic algorithm

Abstract

Permeation of N2, CH4, O2 and CO2 molecules through a carbon molecular sieve (CMS) was studied over a wide range of pressures using the transport mechanism. For proper utilization of carbon molecular sieve membrane in gas separation processes, prediction of behavior and recognition of proper gas transport mechanism as well as finding effective permeation parameters are necessary. A mathematical model of the gas transfer through a CMS membrane was developed using genetic algorithm (GA). Numerous types of mechanisms have been proposed so far for gas transport through capillaries, namely: Knudsen, slip and viscous flow. Moreover, surface flow usually occurs in parallel with other transport mechanisms such as Knudsen or viscous flow. The experimental data of gas permeation in CMS membranes and an appropriate genetic algorithm-based optimization method were used to establish the transport parameters. A GA, an optimization procedure based on the theory of evolution, was compared with non-linear regression for the ability of these two algorithms to fit the coefficients of Poultry growth models. It was found that GA approach could be more capable to define the parameters of permeation equation than non-linear regression. The model in most cases showed a good agreement between the predicted and measured values of the permeability.