Characterization of C7+ fraction properties of crude oils and gas-condensates using data driven models

摘要

Petroleum fractions, in particular heptane-plus fraction (C7+), are complex mixtures and measuring properties of hydrocarbon-plus fractions is difficult and time-consuming. In this study, the normal boiling point, specific gravity and molecular weight of C7+ are estimated as a function of molecular weight, specific gravity as well as cumulative weight fractions using four soft computing strategies called least squares support vector machine, decision tree, gene expression programing, and artificial neural network. The results obtained in this study demonstrate that the developed models could be applied properly for the characterization and estimation of hydrocarbon-plus properties of crude oils and gas-condensates.