Extending the lifetime and increasing oil recovery of a reservoir requires reliable reservoir simulation models that allow the accurate prediction of production profiles for different IOR strategies. Major ingredients to these simulation models are multi-phase flow parameters, such as relative permeability and capillary pressure, properly adapted to the relevant flow phenomena and the spatial resolution of the simulation model. Traditionally, relatively little work is spent on preparing multi-phase flow parameters for a reservoir simulation model, as far as the adaptation to the relevant flow phenomenon and the resolution of the model is concerned. Usually, SCAL results are compiled and applied either directly or slightly corrected. At best, a so-called ‘dynamic upscaling’ (history match of a simulation model with respect to relative permeability and capillary pressure) is performed. This approach limits, however, the predictive capability of the simulation models in that it requires historic field data. Such an approach is useful after for example a water injection pilot, when the experiences from the pilot project are to be included into the full field simulation model. If, however, a completely new drainage strategy is to be applied, historic data may be of very limited value.
However, given modern tools for single-phase and multi-phase upscaling, a two-stage workflow has been developed, that eventually might provide a rigorous solution to this problem [1,2,3]. Major ingredients to this upscaling technique are i) appropriate rock models that provide process dependent pore-scale flow parameter through advanced network modelling, ii) detailed lithofacies models for all relevant flow units, and iii) a sound geological model. The pore-scale flow parameters (rock curves) are then upscaled to facies-scale through steady-state methods, thereby accounting for small-scale heterogeneities as well as effects of capillarity and buoyancy. This process results in a set of flow parameter for each flow unit in the geological model. Upscaling from the geological model to the simulation model finally yields appropriate field-scale flow parameters.
R. Lerdahl, SPE, A.B. Rustad, SPE, T.G. Theting, SPE, J.Å. Stensen, SPE, P.E. Øren, SPE, S. Bakke, SPE, T. Boassen, SPE, and B. Palatnik, SPE, Statoil
presented at the SPE Europec/EAGE Annual Conference, 13-16 June 2005, Madrid, Spain