Experimental measurement and prediction of carbon dioxide clathrate hydrate phase boundary in presence of ethanol + sodium chloride, potassium chloride, magnesium chloride and calcium chloride aqueous solution

Abstract

The dissociation condition of clathrate hydrate of carbon dioxide in aqueous solutions of ethanol + NaCl, KCl, MgCl₂ and CaCl₂ are measured in the temperature range of (259.4–274.7) K and pressures up to 3.7 MPa. Isochoric pressure-search method is used for all measurements. In addition for the first time in open literature a machine learning algorithm model based on artificial neural network is developed to predict equilibrium temperatures. Absolute relative deviation (ARD) and correlation coefficient (R²) for trained neural network are 0.0305 and 0.9994, respectively.

Keywords:

• ANN
• CaCl₂
• carbon dioxide hydrate
• ethanol
• KCl
• MgCl₂
• NaCl

Disclosure statement

Neural networks have suitable potential to predict non linear complex functions. To our best knowledge for the first time in open literature, an artificial neural network is developed to predict temperature of hydrate formation.