Data-Driven Analyses of Low Salinity Waterflooding in Carbonates

Low salinity water (LSW) injection is a promising Enhanced Oil Recovery (EOR) technique that has the potential to improve oil recovery and been studied by many researchers. LSW flooding in carbonates widely evaluated coreflooding tests prior studies. A closer look at literature on indicates number of gaps shortcomings. It difficult understand exact relationship between different controlling parameters effect carbonates. The active mechanisms involved improvement are still uncertain more analyses required. To predict performance study displacement, data collected from available experimental studies were analyzed using analysis approaches. We used linear regression relationships single incremental factor (RF). Correlations rock, oil, brine properties tertiary RF weak negligible. Subsequently, we oil/brine multivariable regression. Relatively strong correlations found for combination RF. also nonlinear applying machine learning (ML) models, such as artificial neural network (ANN), support vector (SVM), decision tree (DT). These models showed better fitting results compared Among applied ML DT provided best correlation parameters, ANN SVM overfitted testing data. Finally, based changes effluent PDIs concentration, interfacial tension, pH, zeta potential, pressure drop.