TX 75083-3836, U.S.A., fax 01-972-952-9435. AbstractThis paper describes the reservoir flow simulation study of the Namorado Sandstone, Albacora Field (offshore Brazil). This study is the end point activity − to which the geological description, geostatistical modeling and upscaling results converge − of the project PRAVAP-2, a main project designed to develop technology on reservoir characterization at Petrobras. This paper details work executed by a multidisciplinary team on a flow simulation study performed in a stochastic modeling scenario. We present aspects related to the model construction and the results obtained from the flow simulation study. The fluid flow model considers a deterministic framework built with high-resolution stratigraphy and detailed structural analysis. Twelve stratigraphic units were grouped in six simulation layers. The heterogeneities inside each layer were modeled using a geostatistical approach: we used the Markov-Bayes simulation method to integrate seismic and well data.A multi-step upscaling technique was used to generate the effective petrophysical properties for the fluid flow simulation model. The idea is based on (1) a hierarchical description of the reservoir heterogeneities and on (2) the computation of the effective properties bottom-up through this hierarchy. Three main work scales were defined: "porous systems" (microscopic scale); "composite facies" (log and core scale); and "seismic facies" (interwell to field scale). Effective permeabilities and two-phase pseudo-functions were numerically computed for each facies at each scale.The simulation grid has a corner point geometry, comprising a total of 46,200 gridblocks with areal dimensions of 100m x 100m. The interconnection of layers was modeled by changing the vertical transmissibility distribution.The history matching was performed by the adjustment of the average pressure for the six layers and of the reservoir production history. The results demonstrated that our sophisticated model allowed a very fast history matching. Extrapolations investigated the adequacy of the development plan, in special the type of the well, horizontal or hydraulic fractured vertical, the number and locations of these development wells to be drilled.

TX 75083-3836, U.S.A., fax 01-972-952-9435. AbstractThis paper describes the reservoir flow simulation study of the Namorado Sandstone, Albacora Field (offshore Brazil). This study is the end point activity − to which the geological description, geostatistical modeling and upscaling results converge − of the project PRAVAP-2, a main project designed to develop technology on reservoir characterization at Petrobras. This paper details work executed by a multidisciplinary team on a flow simulation study performed in a stochastic modeling scenario. We present aspects related to the model construction and the results obtained from the flow simulation study. The fluid flow model considers a deterministic framework built with high-resolution stratigraphy and detailed structural analysis. Twelve stratigraphic units were grouped in six simulation layers. The heterogeneities inside each layer were modeled using a geostatistical approach: we used the Markov-Bayes simulation method to integrate seismic and well data.A multi-step upscaling technique was used to generate the effective petrophysical properties for the fluid flow simulation model. The idea is based on (1) a hierarchical description of the reservoir heterogeneities and on (2) the computation of the effective properties bottom-up through this hierarchy. Three main work scales were defined: "porous systems" (microscopic scale); "composite facies" (log and core scale); and "seismic facies" (interwell to field scale). Effective permeabilities and two-phase pseudo-functions were numerically computed for each facies at each scale.The simulation grid has a corner point geometry, comprising a total of 46,200 gridblocks with areal dimensions of 100m x 100m. The interconnection of layers was modeled by changing the vertical transmissibility distribution.The history matching was performed by the adjustment of the average pressure for the six layers and of the reservoir production history. The results demonstrated that our sophisticated model allowed a very fast history matching. Extrapolations investigated the adequacy of the development plan, in special the type of the well, horizontal or hydraulic fractured vertical, the number and locations of these development wells to be drilled.