Title: Prediction of vapour-liquid equilibrium ratios for the CH₄-CO₂-H₂S systems using artificial neural networks

Authors: Alireza Shams

Addresses: La Prairie, Quebec J5R 0H5, Canada

Abstract: Acid gas removal is an important part of natural gas processing which is designed based on the type and concentration of impurities and the final specifications for the outlet gas. The typical process for removing acid gas involves their absorption from natural gas into a regenerative solvent. The efficiency of this process is affected by the mass transfer process between gas and liquid phases; so, it is necessary to study the phase equilibrium encountered in acid gas mixtures at different pressures and temperatures. In this work, an artificial neural network (ANN) has been developed to predict the vapour-liquid equilibrium ratio (K_LV) for the CH₄-CO₂-H₂S binary and ternary systems. Results show that a neural-network-based method provides an accurate estimation (93-98%) of K_LV while considering all affecting parameters, compared to limited reported correlations. This work can be considered as an introduction to using ANNs for predicting the physical properties of mixtures. [Received: March 7, 2020; Accepted: March 10, 2021]

Keywords: natural gas processing; equilibrium; artificial neural network; ANN; prediction.

DOI: 10.1504/IJOGCT.2022.121051

International Journal of Oil, Gas and Coal Technology, 2022 Vol.29 No.3, pp.226 - 240

Received: 07 Mar 2020
Accepted: 10 Mar 2021
Published online: 23 Feb 2022 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article