Effect of neglecting interfacial coupling when using vertical flow experiments to determine relative permeability

Abstract

The conventional equations used to describe vertical, two-phase flow through porous media are based on the assumption that interfacial coupling may be neglected. This study demonstrates that failure to include interfacial coupling in the mathematical description of such flow can introduce significant amounts of model error into the equations used to describe vertical, two-phase flow through porous media. This is done by utilizing modified forms of the generalized transport equations to construct equations that can be used to remove model error from conventionally determined relative permeability curves, and to calculate the amount of model error introduced by failing to account properly for interfacial coupling. Using these equations, it was found that the effect of removing the model error from the experimentally determined curves was to decrease the magnitude of the wetting-phase curves, and to increase the magnitude of the nonwetting-phase curves. Moreover, it was determined that the model error incurred by failing to account properly for interfacial coupling was approximately an order of magnitude larger for the wetting-phase curves than it was for the nonwetting-phase curves. In addition, it was found that the error incurred by failing to account for capillary coupling was larger for countercurrent flow than it was for cocurrent flow. Finally, it was shown that the corrected relative permeability curves appeared to apply to a more water-wet porous medium than did the experimentally determined curves. The results of this study should be of interest to people who use vertical flow experiments to measure relative permeability, and to people who use such information in reservoir simulation.