La Pointe, Paul¹ 
(1) Golder Associates Inc, Redmond, WA

ABSTRACT: A Workflow for Building and Validating Static Reservoir Models for Fractured Reservoirs for Direct Implementation into Dynamic Simulators

In most fractured reservoirs, the most important parameter to characterize is fracture network connectivity. In progressing from static to dynamic models, it is important to determine the scale of fracture heterogeneity in terms of the scale of the flow processes in the dynamic simulation, determine the geological controls on fracture heterogeneity at this scale, and to develop a quantitative model of the natural fracturing that can be directly validated against transient well test, tracer or other types of dynamic data prior to incorporation in a reservoir simulation. Since direct fracture data, like image logs or core, and well test data do not capture the flow properties nor the heterogeneity of the fracture system at the scale of interest for dynamic modeling, it is necessary to build a model that behaves properly at the dynamic scale from data that represents properties on a much smaller scale. Discrete Fracture Network (DFN) models, which represent fractures as discrete objects with fluid flow and mechanical properties, provide this tool. The permeability, storage properties and transport properties of these discrete fractures can be calibrated through numerical history matching of transient well tests, tracer tests or other types of well test information. Once calibrated, structured or unstructured reservoir simulation grids can be superimposed on the calibrated DFN model. Simulation parameter values such as fracture porosity, directional fracture permeability and sigma factor can be directly extracted from the discrete fractures in each simulation grid cell and used as direct input to the dynamic simulation. A series of case histories in sandstone, carbonate and crystalline fractured reservoirs will be used to illustrate this workflow, focusing on determining what scale of fracture network heterogeneity is important to incorporate into the dynamic model, and showing methods that have proven useful for quantifying and capturing the heterogeneous flow response of the fracture network once the appropriate scale has been identified.

AAPG Search and Discovery Article #90026©2004 AAPG Annual Meeting, Dallas, Texas, April 18-21, 2004.