|
|
| |
Micromechanics of Creep Failure in Rocks Under Fluid–Rock Interactions
Omid MORADIAN, Renzo SOLIS VEGA
[New Mexico institute of Mining and Technology, USA]
Understanding how fluid–rock interactions influence time-dependent deformation is essential for evaluating the long-term stability and permeability evolution of reservoirs in Enhanced Geothermal Systems (EGS). Previous studies have extensively examined the creep behavior of rocks and investigated the influence of fluids on deformation processes; however, the micromechanical processes governing creep in the presence of fluids remain poorly constrained. This study investigates the comparative micromechanics of creep failure in rocks under two controlled laboratory conditions, dry and fully saturated with prescribed pore pressure, to isolate the effects of fluid–rock coupling. Cylindrical rock cores were subjected to constant-stress triaxial creep tests while high-frequency acoustic emission (AE) signals were continuously recorded to capture microcrack activity throughout the deformation process. To interpret the complex, high-dimensional AE waveforms, an unsupervised deep learning framework based on deep clustering was implemented. The AE signals were first converted into time–frequency spectrograms, which were then processed using convolutional autoencoders to extract latent feature representations. These latent vectors were subsequently clustered using k-means to identify emergent patterns and transitions in the temporal evolution of AE activity. The findings of this study can provide new insight into the time-dependent weakening of geothermal reservoir rocks and demonstrate the potential of deep-learning-based AE analysis as a foundation for early detection of creep-induced instability in long-term EGS monitoring.
Topic: Geophysics