The modelling of fractured reservoirs: constraints and potential for fracture network geometry and hydraulics analysis

Abstract

AbstractThe interconnectedness of a subsurface fracture system depends on five fracture and fracture set geometry characteristics: density, orientation, dimensions (height and length), connectivity and aperture. Orientation and density can be determined with some degree of accuracy but the uncertainty ranges associated with the other characteristics will be large. Calibration with dynamic indicators will bring out the hydraulic properties of the network and can help to reduce uncertainty in the geometric characteristics. The fracture network geometry in the inter-well space can be described through geostatistical analysis but more commonly it is derived from seismic data, through edge detection or the analysis and calibration of seismic velocity anisotropy attributes or curvature analysis. Geomechanical analysis is necessary to evaluate the impact on network conductivity of the changing stress state in the reservoir. Using percolation theory, a qualitative assessment can be made of how the conductivity and interconnectedness of the fracture network is developed in terms of network clusters. This integrated analysis of static and dynamic fracture data leads to the formulation of conceptual models and a set of modelling rules. A Digital (Discrete) Fracture Network model, based on these concept(s), rules and inter-well interpolation data, can then be analysed in terms of expected network cluster size, distribution and hydraulics. This provides a control on the expectations embedded in the concepts prior to dynamic modelling. The uncertainties are normally quite large and very seldom will they allow construction of the definitive fracture model. The systematic and integrated analysis of fracture network geometry, constrained with dynamic indicators and subsequent network cluster analysis, are necessary preparatory steps for construction and analysis of dynamic models.