No abstract

Abstract. Mounting evidence suggests that subcritical crack growth is an important mechanism for the development of natural fractures. Numerical simulations of fracture patterns are sensitive to the subcritical crack growth index, the exponent used to describe the power law dependence of crack velocity on stress intensity.

No abstract

In conventional reservoir simulations, gridblock permeabilities are frequently assigned values larger than those observed in core measurements to obtain reasonable history matches. Even then, accuracy with regard to some aspects of the performance such as water or gas cuts, breakthrough times, and sweep efficiencies may be inadequate. In some cases, this could be caused by the presence of substantial flow through natural fractures unaccounted for in the simulation. In this paper, we present a numerical investigation into the effects of coupled fracture-matrix fluid flow on equivalent permeability.A fracture-mechanics-based crack-growth simulator, rather than a purely stochastic method, was used to generate fracture networks with realistic clustering, spacing, and fracture lengths dependent on Young's modulus, the subcritical crack index, the bed thickness, and the tectonic strain. Coupled fracture-matrix fluid-flow simulations of the resulting fracture patterns were performed with a finite-difference simulator to obtain equivalent permeabilities that can be used in a coarse-scale flow simulation. The effects of diagenetic cements completely filling smaller aperture fractures and partially filling larger aperture fractures were also studied.Fractures were represented in finite-difference simulations both explicitly as grid cells and implicitly using nonneighbor connections (NNCs) between grid cells. The results indicate that even though fracture permeability is highly sensitive to fracture aperture, the computed equivalent permeabilities are more sensitive to fracture patterns and connectivity.

In conventional reservoir simulations, grid block permeabilities are frequently assigned values larger than those observed in core measurements to obtain reasonable history matches. Even then, accuracy with regard to some aspects of the performance such as water or gas cuts, breakthrough times, and sweep efficiencies may be inadequate. In some cases this could be due to the presence of substantial flow through natural fractures unaccounted for in the simulation. In this paper we present a numerical investigation into the effects of coupled fracture-matrix fluid flow on equivalent permeability. A fracture-mechanics-based crack growth simulator, rather than a purely stochastic method, was used to generate fracture networks with realistic clustering, spacing and fracture lengths dependent on the Young's modulus, the subcritical crack index, the bed thickness, and tectonic strain. Coupled fracture-matrix fluid flow simulations of the resulting fracture patterns were performed using a finite-difference simulator to obtain equivalent permeabilities that can be used in a coarser scale flow simulation. Fractures were represented in finite-difference simulators both explicitly as grid cells and implicitly using nonneighbor connections between grid cells. The results indicate that even though fracture permeability is highly sensitive to fracture aperture, the computed equivalent permeabilities are more sensitive to fracture patterns and connectivity. The effects of diagenetic cements completely filling smaller aperture fractures and partially filling larger aperture fractures were also studied. Introduction High-permeability fracture networks in a matrix system can create high-conductivity channels for the flow of fluids through a reservoir, producing larger flow rates and therefore larger apparent permeabilities. The presence of fractures can also cause earlier breakthrough of the displacing fluid and lead to poorer sweep efficiencies in displacement processes. A better understanding of reservoir performance in such cases may be obtained by including the details of the fluid flow in fractures in a coupled fracture-matrix reservoir flow model. It is impossible to accurately quantify interwell fracture network geometries in sufficient detail to directly model their effect on reservoir behavior. Thus, most modeling approaches have been statistical, using data from outcrop and well-bore observations to determine distributions for fracture attributes such as fracture length, spacing, and aperture to randomly populate a field. In this paper we present a geomechanics-based approach where a model of the fracturing process is used in predicting fracture characteristics. Opening-mode (mode I) fractures propagate in a plane perpendicular to the least compressive stress.1 Such fractures may propagate when the fluid pressure inside them exceeds the local least compressive stress or when the local stress becomes tensile. Although this is a necessary condition for fracture propagation, it is not sufficient. Fracture propagation from an existing flaw also requires a sufficiently large stress intensity factor, KI, to exceed the rock's fracture strength. The mode I stress intensity factor, KI, for a uniformly loaded isolated crack of length 2c is defined asEquation 1 andEquation 2 where ?sI is the driving stress, Pp is the pore pressure in the rock, and smin is the in situ least compressive stress.