Prediction of CO2-Brine interfacial tension using a rigorous approach

Geologic sequestration of CO2 in deep saline aquifers is becoming increasingly important as a method with the greatest potential to economically sequester large volumes of anthropogenic CO2. The interfacial tension (IFT) between the formation brine in the aquifer and the injected CO2 phase has a significant influence on the displacement, and its precise determination is essential for accurate modeling and evaluation of such a process. This paper presents two new mathematical models to calculate the brine/CO2 IFT. The two models differ in input parameters; pressure, temperature, and salinity for the first model, and pressure, temperature, and brine composition for the second one. The proposed intelligent models were developed using the least square support vector machine (LSSVM) modeling optimized with coupled simulated annealing (CSA) optimization approach. The models were developed and tested employing a total set of 1019 experimentally measured IFT values covering wide ranges of pressures, temperatures, salinities and salt types. The results show that the proposed models significantly outperform all the existing correlations. Moreover, the relative impact of input parameters on IFT was determined through three different sensitivity analysis. Finally, an outlier diagnosis was performed on the whole data set to detect the probable outliers from experimental data.