--> Abstract: Integrating Detailed Cross-Fault Fluid Flow Behaviour into Existing Reservoir Simulation Models, by Stephen R. Freeman, Simon D. Harris, and Robert Knipe; #90082 (2008)

Datapages, Inc.Print this page

Integrating Detailed Cross-Fault Fluid Flow Behaviour into Existing Reservoir Simulation Models

Stephen R. Freeman, Simon D. Harris, and Robert Knipe
Rock Deformation Research, Leeds, United Kingdom

Reservoir simulation models are necessary simplifications of more complex permeability distributions. Across faults, combinations of rapid geometric and property changes can induce changes in flow behaviour that simplified models may not appropriately capture. A combination of fine-scale permeability variations and geometric uncertainty within fault zones makes a unique approximation of a fault zone unrealistic. Even if such a fine scaled grid could be produced, computation times required for a reservoir simulation would make the solution inefficient. Rather than defining all the required connections physically within the grid, cross-fault transmissibilities appropriate for various detailed internal fault geometries are computed. These transmissibilities are then applied as non-physical connections into the pre-existing reservoir grid that override the existing physical connections. The result is a coarse-scale simulation model that behaves like a model with detailed fault architectures.
Specific geometric and property styles of fault zone architectures depend on the specific mechanical stratigraphy and geohistory. A stochastic set of architectures is generated via a Monte Carlo simulation, and used to determine the variability in cross-fault transmissibility. This approach allows a range of geological fault solutions to be applied to a single simulation grid without the need to regenerate the physical grid.

The technique developed allows detailed fault permeability and mapping uncertainty to be incorporated effectively into existing reservoir simulation models and avoids remodelling of the geological grid. The method has little impact on simulation run times, and provides a fast and effective method for more accurately determining and integrating the impact of faults into simulation models.

AAPG International Conference and Exhibition, Cape Town, South Africa 2008 © AAPG Search and Discovery