Merging Outcrop Data and Geomechanical Information in Stochastic Models ofFractured Reservoirs

摘要

According to recent estimates, the potential for oil productionrnfrom fractured reservoirs in North America is of the order ofrntens of billion of barrels. With domestic production dependingrnmore and more on mature fields, better technology forrncharacterizing fracture flow paths, especially in deep, nonconventionalrnplays and in carbonate rocks is key to producingrnhydrocarbons economically. Fracture transmissivity (orrnpermeability) can enhance oil production, or on the otherrnhand, result in early water breakthrough and consequentlyrnearly well abandonment. However, spatial characteristics ofrnthe fracture system cannot be known deterministically in thernsubsurface reservoir. Instead, stochastic characterisation ofrnfracture systems is usually attempted. The development of arnstochastic modeling approach that yields realistic field-scalernmodel of fracture networks consistent with patterns observedrnon an outcrop and adhere to a mechanical basis for fracturernpropagation is presented in this paper.