Phase equilibrium modeling of clathrate hydrates of methane, carbon dioxide, nitrogen, and hydrogen + water soluble organic promoters using Support Vector Machine algorithm

In this work, the Least Squares Support Vector Machine (LSSVM) algorithm is employed to present several numerical models for calculation/estimation of the clathrate hydrate dissociation conditions of methane, carbon dioxide, nitrogen, and hydrogen in the presence of 3 most-widely used water “soluble” organic promoters including tetrahydrofuran (THF), 1,4-dioxane, and acetone. The Hybrid Genetic Algorithm (H-GA) is applied to obtain the optimal models parameters. The final results indicate the accuracy and reliability of the proposed models with the absolute average relative deviations (AARD) of generally less than 10% and squared correlation coefficients of generally greater than 0.99. It was found that the models contribute to high deviations for the representations/predictions of the N2 + THF, CH4 + 1,4-dioxane, and H2 + acetone clathrate hydrates phase equilibria.

► The Support Vector Machine (SVM) algorithm is used to present several hydrate numerical models. ► The objective is to determine the hydrate dissociation conditions in presence of promoters. ► The investigated organic promoters include tetrahydrofuran (THF), 1,4-dioxane, and acetone. ► The studied hydrate formers are CH4, CO2, N2, and H2. ► The AARD of the models are generally less than 10% and R2 values are generally greater than 0.99.