Faults in conventional flow simulation models: a consideration of representational assumptions and geological uncertainties

Even when geologically based methods are used to determine fault rock permeabilities and thicknesses for input into flow simulators, a wide range of simplifying assumptions regarding fault structure and content are still present. Many of these assumptions are addressed by defining quantitative and flexible methods for realistic parameterization of fault-related uncertainties, and by defining automated methods for including these effects routinely in full-field flow simulation modelling. The fault effects considered include: the two-phase properties of fault rocks; the spatial distributions of naturally variable or uncertain single-phase fault rock properties and fault throws; and the frequencies and properties of sub-resolution fault system or fault zone complexities, including sub-seismic faults, normal drag and damage zones, paired slip surfaces and fault relay zones. Innovative two-phase or geometrical upscaling approaches implemented in a reservoir simulator preprocessor provide transmissibility solutions incorporating the effect, but represented within the geometrical framework of the full-field flow simulation model. The solutions and flexible workflows are applied and discussed within the context of a sensitivity study carried out on two faulted versions of the same full-field flow simulation model. Significant influence of some of these generally neglected fault-related assumptions and uncertainties is revealed.