Abstract
Understanding the mechanism of stick-slip instability of faults is of great significance for earthquake forecast and seismic fortification. In this paper, combined with the 3D reconstruction technology and discrete element method (DEM), the stick-slip shear failure process is visualized and the evolution mechanism of each stage of stick-slip process is studied. First, the 3D morphology scanning tests are carried out to define the 3D topography of the sample interface, which is imported into PFC3D to generate particle assembly, and micro-parameters of the particle assembly are calibrated by the experimental results. Second, the loading conditions are applied to the particle assembly, the stick-slip instability of the simulated fault occurs spontaneously, and the evolution of fault displacement and velocity distribution during the whole stick-slip period are studied. Third, the effects of normal stress, shear loading rate and interface roughness on stick-slip instability are investigated. Finally, nine parameters related to the steady-state friction coefficient, namely normal stress, natural density, Young's modulus, Poisson's ratio, maximum static friction coefficient, reference slip velocity, reference friction coefficient, a-b and slip velocity, are selected to establish the steady-state friction coefficient forecast model based on back propagation neural network (BPNN), and the relative importance analysis of relevant parameters is also implemented. Our research is helpful to reveal the seismogenic mechanism.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call