On the prediction of interfacial tension (IFT) for water-hydrocarbon gas system

It is important to develop accurate and efficient modeling approaches for estimation of interfacial tension between water and hydrocarbon gases because of notable importance of this parameter in various areas of chemical and petroleum engineering. This work presents three predictive models for hydrocarbon gas-water IFT prediction namely radial basis function networks optimized by generic algorithm (GA-RBF), least square support vector machine optimized by coupled simulated annealing (CSA-LSSVM) and adaptive neuro-fuzzy inference system optimized by combination of PSO and hybrid methods (CHPSO-ANFIS). An extensive number 011105 experimental data were collected from literature covering wide ranges of operational conditions. The outcomes of the implemented models were compared with four literature correlations. Results indicate that although all the developed computer-based models were effective and accurate in reproducing the target data, the GA-RBF model provides better results compared to CSA-LSSVM and CHPSO-ANFIS models. Moreover, the predictions of the developed models are better than four selected literature correlations. (C) 2016 Published by Elsevier B.V.