Investigation of Rate and Method Dependency in Relative Permeability

Researchers at Imperial College have identified that spatial heterogeneity in local capillary pressure is a major contributor to rate dependency. It is postulated that when flow rates are reduced, a pressure gradient threshold will be reached below which the flow properties will become rate dependent. This represents the emergence of capillary forces that are significant relative to the viscous driven flow forces. The threshold itself will be specific to the rock; particularly the strength of the heterogeneity (e.g. distribution of pore entry pressures), the spatial structure (i.e. layers, vugs, etc.) and orientation of the heterogeneity with respect to flow (parallel, perpendicular, oblique, random). Saturation variability from X-ray imaging will provide information about capillary heterogeneity which could then be parameterised digitally. Together with petrography of the sample, mercury injection capillary pressure measurements and 2D simulation, a prediction of rate dependency can be made for drainage processes, akin to the hydrocarbon trap filling process. This work will be expanded to imbibition processes to investigate water displacing oil mechanisms. To investigate the discrepancy between plug scale data (cm) and simulation grid block scale data (~m's) it is suggested that the boundaries imposed on the rock by the laboratory setup (inlet, outlet and circumference) effect the measurements themselves. The severity of the effect may depend on the level and type of heterogeneity seen in the rock. It has been observed that for a layered rock where dominant flow paths pinch out on the confinement sleeve, relative permeability decreases with decreasing flow rate [1, 2]. The approach proposed will be to create a numerical model of the rock core and use the model to remove the effects that the experimental boundaries have on the result, and thus extract a more appropriate set of flow properties free from laboratory artefacts.