Modeling of Hydraulic Fracture Network Propagation in a Naturally Fractured Formation

Abstract

Hydraulic fracturing in shale gas reservoirs has often resulted in complex fracture network growth, as evidenced by microseismic monitoring. The nature and degree of fracture complexity must be clearly understood to optimize stimulation design and completion strategy. Unfortunately, the existing single planar fracture models used in the industry today are not able to simulate complex fracture networks. A new hydraulic fracture model is developed to simulate complex fracture network propagation in a formation with preexisting natural fractures. The model solves a system of equations governing fracture deformation, height growth, fluid flow, and proppant transport in a complex fracture network with multiple propagating fracture tips. The interaction between a hydraulic fracture and pre-existing natural fractures is taken into account by using an analytical crossing model and is validated against experimental data. The model is able to predict whether a hydraulic fracture front crosses or is arrested by a natural fracture it encounters, which leads to complexity. It also considers the mechanical interaction among the adjacent fractures (i.e., the "stress shadow" effect). An efficient numerical scheme is used in the model so it can simulate the complex problem in a relatively short computation time to allow for day-to-day engineering design use. Simulation results from the new complex fracture model show that stress anisotropy, natural fractures, and interfacial friction play critical roles in creating fracture network complexity. Decreasing stress anisotropy or interfacial friction can change the induced fracture geometry from a bi-wing fracture to a complex fracture network for the same initial natural fractures. The results presented illustrate the importance of rock fabrics and stresses on fracture complexity in unconventional reservoirs. They have major implications on matching microseismic observations and improving fracture stimulation design.

Numerical methods for hydraulic fracture propagation: A review of recent trends

“…(22), but with different physical constants determining the coefficient C c . However, because these generalizations also lead to v L ∝ (t − t o (x)) −1/2 , the effects can be lumped into a single, composite fluid loss constant C L , which is typically determined from calibration experiments such as diagnostic fracture injection tests (Nolte, 1979;Castillo, 1987;Barree and Mukherjee, 1996) and/or matching model predictions to fracture geometries inferred from microseismic monitoring (Weng et al, 2011b). The resulting form of Carter's leakoff equation is thus given by…”

“…Weng et al, 2011a;Meyer and Bazan, 2011). When the fracture height growth is involved during the fracture network formation, the stacked cell model consisting of many rows of elements in the vertical direction was proposed by Cohen et al (2015).…”

Development of an Efficient Embedded Discrete Fracture Model for 3D Compositional Reservoir Simulation in Fractured Reservoirs

“…Maxwell et al (2002) and Fisher et al (2005) interpreted microseismic data in the Barnett Shale and observed significant branching of hydraulic fractures, likely the result of opening pre-existing natural fractures. Also, Weng et al (2011) showed that low in-situ stress anisotropy and pre-existing natural fractures play important roles in creating fracture network complexity.…”

A Review of Numerical Simulation Strategies for Hydraulic Fracturing, Natural Fracture Reactivation and Induced Microseismicity Prediction

“…In a fracture/fault system, the presence of discontinuities or weakness planes affect the way the induced hydraulic fractures propagate. They can force the hydraulic fractures to deviate from the general propagation direction (direction of least resistance in case of tensile failure) and lead to extensive branching [3,4], or they can arrest the hydraulic fracture at the point of interaction. Extensive technical literature [5 -9] experimental studies and simulation analysis is available on the effects of pre-existing fractures on hydraulic fracturing evolution.…”

“…The modelling of the induced fracture propagation in a naturally fractured reservoir was addressed via the BEM approach by Koshelev and Ghassemi [87] and via the PKN approach by Potluri [88], while Weng et al [4] combined a discrete fracture network with a P3D fracture propagation simulation.…”