Compositional Model for Estimating Asphaltene Precipitation Conditions in Live Reservoir Oil Systems

Asphaltenes form the most polar fractions in crude oil, which decrease considerably the rock permeability and the oil recovery and in total can cause operational problems. Hence, it is important to estimate the asphaltene precipitation as a function of operating conditions, crude oil composition, and characterization. In this article, a reliable and robust model, namely, the least squares support vector machine, is applied to predict the onset pressures of asphaltene precipitation in live oil systems as well as oil saturation conditions. To pursue our objective, we used literature-reported onset and saturation (bubble point) pressures data of various live oils from different regions, but mostly from the Middle East, with different amounts of asphaltenes. The results indicate that the proposed strategy provides reasonably satisfactory predictive results. Additionally, the obtained results demonstrate not only the validation of the proposed method but also pose an interesting alternative to the classic methods of estimating asphaltene precipitation due to the low number of adjustable parameters used in our model. [GRAPHICS]