Résultat de la recherche
Hydromechanical insight of fracture opening and closure during in-situ hydraulic fracturing in crystalline rock
Six hydraulic fracturing (HF) experiments were conducted in situ at the Grimsel Test Site (GTS), Switzerland, using two boreholes drilled in sparsely fractured crystalline rock. High spatial and temporal resolution monitoring of fracture fluid pressure and strain improve our understanding of fracturing dynamics during and directly following high-pressure fluid injection. In three out of the six experiments, a shear-thinning fluid with an initial static viscosity approximately 30 times higher than water was used to understand the importance of fracture leak-off better. Diagnostic analyses of the shut-in phases were used to determine the minimum principal stress magnitude for the fracture closure cycles, yielding an estimate of the effective instantaneous shut-in pressure (effective ISIP) 4.49±0.22 MPa. The jacking pressure of the hydraulic fracture was measured during the pressurecontrolled step-test. A new method was developed using the uniaxial Fibre-Bragg Grating strain signals to estimate the jacking pressure, which agrees with the traditional flow versus pressure method. The technique has the advantage of observing the behavior of natural fractures next to the injection interval. The experiments can be divided into two groups depending on the injection location (i.e., South or North to a brittle-ductile S3 shear zone). The experiments executed South of this zone have a jacking pressure above the effective ISIP. The proximity to the S3 shear zone and the complex geological structure led to near-wellbore tortuosity and heterogeneous stress effects masking the jacking pressure. In comparison, the experiments North of the S3 shear zone has a jacking pressure below the effective ISIP. This is an effect related to shear dislocation and fracture opening. Both processes can occur almost synchronously and provide new insights into the complicated mixedmode deformation processes triggered by high-pressure injection.
Seismo-hydromechanical interaction during in-situ hydraulic fracturing experiments
Hydraulic fracturing is a common technique used in a variety of fields like civil and mining engineering, oil & gas and geothermal industry. It can be used to enhance the permeability of low permeable rocks, to increase the connectivity of natural fractures, to modify the rock mass strength, or to measure the Earth’s stress field. In the context of deep geothermal energy exploitation, a heat exchanger needs to be created at depth with characteristics favorable for heat extraction i.e. sufficient permeability and heat exchanger area. The creation of the heat exchanger for geothermal heat extraction remains a critical element with high associated risks including poor reservoir performance and induced seismicity. Hence the need for a better understanding of the coupled seismic-hydromechanical processes during stimulation operations. The execution of experiments on the intermediate-scale has the advantage of a better control on the processes associated with induced seismicity and reservoir performance compared to full-scale and allow to use comprehensive real time monitoring of pore pressure, rock mass deformation and seismicity. This scale is closer to the full-scale stimulation than laboratory scale, where seismo-hydromechanical interactions are generally focused on single fractures. The decameter-scale In-situ Stimulation and Circulation (ISC) project took place between 2015 and 2018 at the Grimsel Test Site (GTS), Switzerland. The GTS is located in the Central Swiss Alps, beneath the mountains of the Grimsel Pass. Overall, the moderately crystalline fractured rock mass shows a pervasive foliation and was intersected by six major sub-vertical shear zones. For each of the two assumed stimulation endmembers, hydraulic shearing and hydraulic fracturing, six experiments were conducted. Prior to the experiments, the test volume was characterized in great detail with respect to geology, geophysics, hydrogeology and in-situ stress field. This doctoral thesis aims at better understanding tensile fracture growth. It includes study of fracture toughness and fracture process zone on laboratory scale and the investigation of the seismo-hydromechanical coupled processes during in-situ hydraulic fracturing experiments. The tested intact Grimsel Granodiorite samples indicate that the resistance against material failure is significantly higher across the foliation plane than along it. The results from Digital Image Correlation (DIC) confirm the development of a semi-elliptical fracture process zone (FPZ) with an average length to width ratio of about two for both principal directions. This agrees well with the available results in the literature. The experimental results of the length of the FPZ give supporting evidence to the fact that a nonlinear cohesion stress distribution provides an accurate cohesive model that agrees well with the experimental results. Additionally, the conformity of the ratio of the FPZ length in two principal directions with the theoretical predictions gives supporting evidence to the proportionality of the FPZ length with respect to the square of fracture toughness to tensile strength. At the decametric scale during the in-situ experiment, the hydromechanical coupled responses of the rock mass and its fractures were captured by a comprehensive monitoring system installed along the tunnels and within dedicated boreholes. At the borehole scale, these processes involved newly created tensile fractures intersecting the injection interval while at the cross-hole scale, the natural network of fractures dominated the propagation process. The six HF experiments can be divided into two groups based on their injection location (i.e., south or north to a brittle ductile shear zone), their similarity of injection pressures and their response to deformation and pressure propagation. The experiments executed north of the shear zone, show smaller injection pressures and larger backflow during bleedoff phases. In addition, we observe re-orientation of the seismic cloud as the fracture propagated away from the wellbore. The re-orientation during propagation is interpreted to be related to a strong stress heterogeneity and the intersection of natural fractures striking different from the propagating hydraulic fracture. This leads in the details to complex geometry departing from theoretical mode I fracture geometries. The seismic activity was limited to about 10 m radial distance from the injection point. In contrast, strain and pressure signals reach further into the rock mass indicating that the process zone around the injection point is larger than the zone illuminated by seismic signals. Furthermore, strain signals indicate not just single fracture openings but also the propagation of multiple fractures. Various methods to estimate the fracture opening and fracture contact pressure were applied and compared from single injection borehole observations with the strain gauge in distance from the injection point. The results show, that the fracture opening pressure was also observed at the strain gauge, associated with a strong increase of fracture transmissivity. The combination of injection pressure and strain observation allows to define an aperture-stress relationship with a general trend toward decreasing normal fracture stiffness during fracture opening. The fracture contact pressure can be estimated, but hydromechanical superposition of pump shut-in and corresponding pressure loss and interaction of the connected surrounding fractures make this task very challenging and error-prone. The pore pressure data set differentiate two distinct responses based on lag time and amplitudes. This allow to distinguish a near- and far-field response. The near-field response is due to pressure diffusion and the far-field response is due to stress perturbation. The far-field pore pressure response is consistent for all experiments, indicating the dominant failure mechanism. This change in the far-field are very sensitive and can be used as a complementary method to seismic monitoring during hydraulic stimulations. The exceptional hydromechanical dataset allow to test numerical stimulations and can help to improve injection strategies, the monitoring design and the numerical modelling.
Stress Measurements for an In Situ Stimulation Experiment in Crystalline Rock: Integration of Induced Seismicity, Stress Relief and Hydraulic Methods
An extensive campaign to characterize rock stresses on the decameter scale was carried out in three 18–24 m long boreholes drilled from a tunnel in foliated granite at the Grimsel Test Site, Switzerland. The survey combined stress relief methods with hydrofracturing (HF) tests and concomitant monitoring of induced seismicity. Hydrofracture traces at the borehole wall were visualized with impression packer tests. The microseismic clouds indicate sub-vertical south-dipping HFs. Initial inversion of the overcoring strains with an isotropic rock model yielded stress tensors that disagreed with the HF and microseismic results. The discrepancy was eliminated using a transversely isotropic rock model, parametrized by a novel method that used numerical modelling of the in situ biaxial cell data to determine the requisite five independent elastic parameters. The results show that stress is reasonably uniform in the rock volume that lies to the south of a shear zone that cuts the NNW of the study volume. Stress in this volume is considered to be unperturbed by structures, and has principal stress magnitudes of 13.1–14.4 MPa for σ1, 9.2–10.2 MPa for σ2, and 8.6–9.7 MPa for σ3 with σ1 plunging to the east at 30–40°. To the NNW of the uniform stress regime, the minimum principal stress declines and the principal axes rotate as the shear zone is approached. The stress perturbation is clearly associated with the shear zone, and may reflect the presence of more fragmented rock acting as a compliant inclusion, or remnant stresses arising from slip on the shear zone in the past.