An End-to End-Approach to Naturally Fractured Reservoir Modeling: Workflow and Implementation

Fracture network models provide both qualitative and quantitative information about the reservoir. They can serve various, sometimes contradictory, purposes: understanding early water breakthrough, optimizing productivity or maximizing recovery. In order to optimize its predictive value, the model must account for all available data: seismic, well logs, well tests and production history. In this paper, we present a general workflow and implementation for building an easily updatable fractured reservoir model. It consists in (1) building a statistical representation of the fracture network, taking into account the spatial variability of the parameters that describe the model (fracture density, orientation, etc.), (2) generating stochastically a geologically consistent discrete fracture model from this representation and (3) computing equivalent fracture permeability, porosity and shape factor to feed the flow simulation model. This process is tightly integrated in the reservoir modeling workflow, so that model updating and uncertainty analysis can be automated.