New method for predicting n-tetradecane/bitumen mixture density: correlation development

Abstract

Nowadays, incredible growth of the energy consumption has changed the global attention to the production and utilization of the heavy crude oils such as bitumen resources around the globe. Amongst the bitumen properties, density is an important parameter which improves bitumen recovery efficiency and transportation quality. For easy production of bitumen, n-alkanes are usually injected into the reservoir to reduce its viscosity and density; however, there are few numbers of models focusing on proper estimation/prediction of diluted bitumen mixture density in literature. In present work, a new method was proposed to accurately prognosticate the bitumen/n-tetradecane mixture density as a function of thermodynamic conditions using Gene Expression Programming (GEP) for the first time as a function of solvent composition, pressure and temperature. Consequently, the proposed model here predicts the mixture density with the average Absolute Relative Deviation (AARD%) of 0.3016% and R-squared (R 2) of 0.9943. Moreover, it is found out the solvent concentration has the highest impact value on mixture density estimation. In conclusion, results of the present study can be so valuable for field engineers and researchers working on solvent-assisted recovery methods from heavy oil reservoirs.