Developing coupled fluid flow and geomechanics simulators to model fracture deformation
This dissertation intends to advance fundamental understanding of two areas of interest in the petroleum industry: complex stimulated fracture network during hydraulic fracturing treatments and induced seismicity during wastewater disposal operations. Successful completion of hydraulic fractures in unconventional formations has been the primary source of increased oil and gas production in the US. However, field observations suggest that the hydraulic fracture networks are much more complex and different from the classical description of bi-wing planar fractures. Thus, the attempts to optimize this stimulation technique are hindered by the uncertainties in predicting the complex fracture network. A by-product of massive improvement in oil and gas production is a significant amount of water being co-produced from these formations. The common practice in the industry is to recycle wastewater for hydraulic fracturing purposes or reinject it into the reservoir through disposal wells. In certain regions of the US, this wastewater injection has led to historically high seismicity rates and earthquakes of Magnitude 5 and above which caused the public to be concerned. To maintain the social license to continue such operations, these concerns need to be addressed, and the physics behind such induced events need to be understood. Two novel hydraulic fracturing and induced seismicity simulators are developed that implicitly couple fluid flow with the stresses induced by fracture deformation in large, complex, three-dimensional discrete fracture networks. The simulators can describe the propagation of hydraulic fractures and opening and shear stimulation of natural fractures. Fracture elements can open or slide, depending on their stress state, fluid pressure, and mechanical properties. Fracture sliding occurs in the direction of maximum resolved shear stress. Nonlinear empirical relations are used to relate normal stress, fracture opening, and fracture sliding to fracture aperture and transmissivity. Field-scale hydraulic fracturing simulations were performed in a dense naturally fractured formation. Height containment of propagating hydraulic fractures between bedding layers is modeled with a vertically heterogeneous stress field or by explicitly imposing hydraulic fracture height containment as a model assumption. The propagating hydraulic fractures can cross natural fractures or terminate against them depending on the natural fracture orientation and stress anisotropy. The simulations demonstrate how interaction with natural fractures in the formation can help explain the high net pressures, relatively short hydraulic fracture lengths, and broad regions of microseismicity that are often observed in the field during stimulation in low permeability formations, some of which were not predicted by classical hydraulic fracturing models. Depending on input parameters, our simulations predicted a variety of stimulation behaviors, from long hydraulic fractures with minimal leakoff into surrounding fractures to broad regions of dense fracturing with a branching network of many natural and newly formed fractures. Induced seismicity simulator was developed to investigate the effects of multiple operational, hydraulic, and geophysical parameters on the magnitude of induced earthquakes. The rate-and-state framework is implemented to include the effect of fault nonlinear friction evolution and to model unstable earthquake rupture. The Embedded Discrete Fracture Model (EDFM) technique is used to model the fluid flow between the matrix and fractures efficiently. The results show that high-rate injections are more likely to induce a more significant earthquake, confirming the statistical correlation attributing induced events to high-rate injection wells. To understand the seismic occurrence outside of the injection zone, the effect of fault permeability structure on seismicity is studied by assigning non-uniform permeabilities as an input parameter. The model shows that the fault rupture is dominantly controlled by initial pressure and stress heterogeneity which ultimately affect the magnitude of an induced earthquake event.