Brine permeability predictions for sand packs and sandstones using Navier-Stokes equations and three-dimensional micro-tomography images of pore spaces

Abstract

A sand pack and a sandstone were imaged with micro-computed tomography at nominal voxel sizes of approximately (6μm)3. From these images the pore morphologies of the porous media were obtained by segmentation. The segmented images were then used to generate surface and volume meshes of pore spaces for flow analysis. Three-dimensional, steady state, isothermal, incompressible single phase fluid flow fields were obtained by solving the continuity and Navier-Stokes equations. An inlet boundary condition was set by specifying the brine injection velocity while a pressure boundary condition was prescribed at the outlet, which resulted in laminar flow, and which is representative of flow in aquifers or oil reservoirs. From the pressure and velocity vector fields we computed the total pressure drop across the sample and the area-averaged velocity at the inlet with which we then determined brine permeability for each porous medium. The predicted permeabilities were consistent with experimental core-flood data; the presented approach is therefore a rapid and cost-effective method to determine single-phase permeabilities of incompressible fluids in porous media. Only small dry rock fragments are required for the described analysis.