Effective local and extended local single-phase upscaling Conference Paper uri icon

abstract

  • Effective strategies for upscaling of flow in porous media rely upon (1) the characterization of flow in the reservoir (intercell transmissibility), (2) flow near wells (effective permeability) and (3) an a priori analysis of error. Transmissibility upscaling represents approximately constant gradient Darcy flow and, compared to permeability upscaling, maximizes the spatial resolution of intercell fluid flow available within a simulator. Near well permeability upscaling replies upon the line source approximation and the calculation of well productivity for radial flow near wells. It provides a characterization of the local reservoir quality. It can be performed without knowledge of the location or rates of physical wells. The error analysis has three components. It relies upon a connectivity analysis, an analysis of variance of velocity or slowness, and an estimate of off-diagonal flow terms in the permeability tensor. All three of these error components can be estimated using local flow information. This a priori error analysis allows a practicing engineer or a spatially adaptive algorithm to adjust the resolution of the upscaled model in order to minimize the most egregious upscaling errors. All three of these calculations can be performed using local or extended-local boundary conditions. This is in contrast to global upscaling approaches which have been previously used in the literature but which rely upon knowledge of a global flow field. Although global upscaling approaches can be quite accurate, their utility is limited as specific well locations and flow rates are not available at the point in the life of a project when the geologic model is typically upscaled in preparation for reservoir flow simulation. The utility of this combined upscaling and error analysis approach is demonstrated using the SPE10 test model, which is simple structurally but which has a fairly complex and heterogeneous permeability field.

published proceedings

  • ECMOR 2010 - 12th European Conference on the Mathematics of Oil Recovery

author list (cited authors)

  • Du, S., Hosseini, S. A., Zhou, Y., & King, M. J.

complete list of authors

  • Du, S||Hosseini, SA||Zhou, Y||King, MJ

publication date

  • January 2010