5:15 PM - 7:15 PM
[SSS10-P32] Investigating Fracture Mechanisms in Hydraulic Fracturing by Dynamic Event Localization
Keywords:Hydraulic-Fracturing, Event-Localization, Mechanism
This paper presents a study investigating the fracture mechanisms through laboratory-scale hydraulic fracturing experiments. During the experiment, we monitored the acoustic emission (AE) signals of the fracturing process using a 24-sensor array, with four sensors placed on each surface of the sample cube-shaped shale specimen, containing a natural pre-existing fracture. A triaxial pressure system was used to apply confining stress, simulating real underground stress conditions. Fluid was injected through a central borehole, generating a fracture perpendicular to the pre-existing natural fracture.
By analyzing the recorded AE signals recorded by the sensor array during the experiment, we identified approximately 320 event signals. Each event was localized using NonLinLoc, an open-source nonlinear Bayesian localization software. Clustering the localized events allowed us to infer the evolution of fracture geometry and reconstruct the 3D velocity model revolution. We then performed the moment tensor inversion, incorporating velocity model variation, to analyze the source mechanisms of the events. Our results reveal a transition from dilatancy-dominant to shear-dominant mechanisms, offering new insights into the fluid-induced seismicity. These findings underscore the potential for dilatancy-based precursors in monitoring and early-warning systems for hydraulic fracturing-induced seismic events.
By analyzing the recorded AE signals recorded by the sensor array during the experiment, we identified approximately 320 event signals. Each event was localized using NonLinLoc, an open-source nonlinear Bayesian localization software. Clustering the localized events allowed us to infer the evolution of fracture geometry and reconstruct the 3D velocity model revolution. We then performed the moment tensor inversion, incorporating velocity model variation, to analyze the source mechanisms of the events. Our results reveal a transition from dilatancy-dominant to shear-dominant mechanisms, offering new insights into the fluid-induced seismicity. These findings underscore the potential for dilatancy-based precursors in monitoring and early-warning systems for hydraulic fracturing-induced seismic events.