Capillary number correlations for two-phase flow in porous media

Understanding fluid flow in porous media is essential for predicting reservoir performance. Prior research has shown that capillary number correlations are commonly used in reservoir simulations to estimate enhanced water flood performance. However, conventional correlations have limitations when applied to diverse rock types and fluid pairs. This gap motivated the need for more robust correlations that can account for rock and fluid variability. Existing studies typically use simplified assumptions that may not capture the complexity of real reservoir conditions. No prior work had resolved how to systematically order correlations based on rock and fluid properties. The limitations of current methods include inconsistencies in large-scale reservoir applications. This paper addresses these issues by introducing a new approach to capillary number correlations. The study builds on established principles of multiphase flow while introducing novel analytical tools.

The primary aim of this research is to develop and analyze capillary number correlations that are more accurate and broadly applicable. The study focuses on addressing inconsistencies in current correlations when used at reservoir scales. The motivation stems from the need to improve the reliability of reservoir simulations for water flooding. The paper seeks to distinguish between rock types using macroscopic correlations. It also aims to introduce a systematic ordering of experimental data. The study's goal is to eliminate limitations of conventional approaches. The researchers propose a three-parameter fit function to enhance correlation accuracy. The work is driven by the need for better predictive models in enhanced oil recovery.

The study analyzes experimental data from seven different reservoir rocks and 21 fluid pairs. The researchers use a three-parameter fit function to model capillary number correlations. They introduce a novel fluid pair-based figure of merit to order the data. The approach involves comparing generalized local macroscopic correlations with conventional methods. The analysis includes evaluating the applicability of the generalized Darcy law. The study uses systematic ordering to distinguish rock types. The methodology emphasizes eliminating shortcomings of traditional correlations. The results are validated against reservoir-scale limitations.

The results show that local macroscopic capillary number correlations can differentiate rock types effectively. The three-parameter fit function improves the ordering of experimental data. The novel figure of merit provides a systematic way to classify fluid pairs. The study finds that conventional correlations may violate the generalized Darcy law at reservoir scales. The analysis reveals inconsistencies in applying correlations to large systems. The fit function reduces variability in correlation predictions. Experimental data from 21 fluid pairs supports the new approach. The results suggest that the new method enhances predictive accuracy.

The authors propose that generalized local macroscopic capillary number correlations improve predictive accuracy. They suggest that the new approach distinguishes rock types better than conventional methods. The study indicates that the three-parameter fit function enhances data ordering. The researchers propose that the novel figure of merit is useful for fluid pair classification. The findings suggest that conventional correlations may fail at reservoir scales. The authors suggest that the generalized Darcy law has limits in large-scale applications. They propose that the new method addresses these limitations. The study concludes that the new approach enhances reservoir simulation reliability.