Regional analysis techniques for integrating experimental and numerical measurements of transport properties of reservoir rocks

Abstract

Assessing the mechanisms of micro-structural change and their effect on transport properties using digital core analysis requires balancing field of view and resolution. This typically leads to the compromise of working with relatively small samples, where boundary effects can be substantial. A direct comparison with experiment, as e.g. desirable to eliminate unknown parameters and integrate numerical and physical experiments, needs to consider these boundary effects. Here we develop a workflow to define measuring windows within a sample where these boundary effects are minimised allowing the integration of physical and numerical experiment. We consider in particular sleeve leakage and use a radial partitioning of the solutions to various transport equations to derive relevant regional measures, which may be used for the development of cross-correlations between physical properties. Samples of Bentheimer and Castlegate sandstone as well as Mt. Gambier limestone and a sucrosic dolomite are considered. The sample plugs are encased in rubber sleeves and micro-CT images acquired at ambient conditions. Using these high-resolution images we calculate transport properties, namely permeability and electrical conductivity, and analyse the resulting field solutions with regard to flux across different regions of interest. The latter are selected on the basis of distance to the sample sleeve inner surface. Clear bypassing at the sleeve-sample interface in terms of elevated fluxes is observed for all samples, although to different extent. We consider different sleeve boundary conditions to define a measuring window minimising these effects, use the procedure to compare flux averages defined over these measuring windows with conventional choices of simulation domains, and compare resulting physical cross-correlations