Prediction of vapor liquid equilibrium data of CO2 in aqueous solution of methyldiethanol amine using thermodynamic and artificial intelegent models

The reliable design and optimization of the separation equipment such as natural gas treating equipment, at first requires the knowledge of the equilibrium properties in particular vapor-liquid equilibrium (VLE) data. In this work an Artificial Neural Network (ANN) model has been developed for modeling of MDEA-CO2-H2O system. Also a thermodynamic model has been used for describing the vapor-liquid equilibrium data (which applies Pitzer’s molality-scale-based equation for describing the Gibbs excess energy of the aqueous phase) of the system.The performance of ANN model to predict equilibrium data of the system were evaluated by comparing their results with the predictions of the thermodynamic model. The results indicate that ANN model can predict the equilibrium data more accurate than thermodynamic model.