A5101794174- Rock Mechanics and Modeling
- Landslides and related hazards
- Dam Engineering and Safety
- Geophysical Methods and Applications
- Geotechnical and Geomechanical Engineering
- Tunneling and Rock Mechanics
- Geotechnical Engineering and Analysis
- Seismic Imaging and Inversion Techniques
- Flood Risk Assessment and Management
- Hydraulic Fracturing and Reservoir Analysis
- Seismology and Earthquake Studies
- Drilling and Well Engineering
- Hydraulic flow and structures
- Advanced Neural Network Applications
- Seismic Waves and Analysis
- Medical Imaging and Analysis
- Hydrological Forecasting Using AI
- Mineral Processing and Grinding
- Soil and Unsaturated Flow
- Fault Detection and Control Systems
- earthquake and tectonic studies
- Erosion and Abrasive Machining
- Smart Agriculture and AI
- Advanced Radiotherapy Techniques
- Ferroelectric and Negative Capacitance Devices
State Key Laboratory of Hydraulics and Mountain River Engineering
2021-2023
Sichuan University
2018-2023
China Coal Technology and Engineering Group Corp (China)
2022
China Coal Research Institute (China)
2022
Northeastern University
2014-2018
The damage or failure of coal rock is accompanied by energy accumulation, dissipation and release. It crucial to study the evolution characteristics for mechanics mining engineering applications. In this paper, specimens sourced from Xinhe mine located in Jining area China were initially subjected uniaxial compression, micro-parameters two-dimensional particle flow code (PFC2D) model calibrated according experimental test results. Then, PFC2D was used subject substantial laws with various...
This paper revisits the implementation of $\textbf{L}$oad-$\textbf{b}$alancing $\textbf{L}$oss (LBL) when training Mixture-of-Experts (MoEs) models. Specifically, LBL for MoEs is defined as $N_E \sum_{i=1}^{N_E} f_i p_i$, where $N_E$ total number experts, $f_i$ represents frequency expert $i$ being selected, and $p_i$ denotes average gating score $i$. Existing MoE frameworks usually employ parallel strategy so that are calculated within a $\textbf{micro-batch}$ then averaged across groups....
The monitoring of data anomaly identification is an important basis for dam safety online and evaluation. In this research, a cluster models was constructed, three-stage method proposed to discriminate outliers. combined detection measured values based on single-point time series simulation, measurement error reduction remote retesting spatio-temporal analysis, environmental response mutation recognition. It brought about efficient accurate classified its inducement. Additionally, problems...
Rock mass is a fractured porous medium usually subjected to complex geostress and fluid pressure simultaneously. Moreover, the properties of rock change in time space due mining-induced fractures. Therefore, it always challenging accurately measure properties. In this study, three-dimensional (3D) microseismic (MS) data-driven damage model for jointed under hydro-mechanical coupling conditions proposed. It 3D finite element that takes seepage, stress field effects into account jointly....
To study the mechanism of surface collapse and crack evolution in a roadway chain failure process pillar recovery Hongling lead zinc ore Inner Mongolia Province, China, microseismic monitoring technology, moment tensor theory, numerical simulation are used for inversion rock mass fracturing, destruction type classification crack, surrounding rock. Research shows following: (1) fracturing is first produced within +955 m level, before extending through hanging wall to ground surface. Then,...
Parameter-efficient tuning has become a trend in transferring large-scale foundation models to downstream applications. Existing methods typically embed some light-weight tuners into the backbone, where both design and learning of are highly dependent on base model. This work offers new paradigm, dubbed Res-Tuning, which intentionally unbinds from backbone. With theoretical empirical evidence, we show that popular approaches have their equivalent counterparts under our unbinding formulation,...
Sudden inelastic deformations in rock are associated with acoustic emission (AE). Therefore, AE monitoring technique can be used to study the fracture processes of rock. In this paper, tests were conducted on granitic gneiss specimens under uniaxial compressive loading conditions. The temporal changes hit parameters and spatial‐temporal evolution events during failure process studied, several characteristic phenomena (i.e., dramatic increase dominant frequency, energy, rate, event a high...
Deformation mechanism in the core rockfill dams with heavy load and high-stress level is difficult to predict control, which one of key problems be solved dam operation safety management control. Aiming at large error obtained by parameter-based functional models (regression model, grey theory etc.) deformation prediction dams, a fractal method its technical process combining variable dimension "metabolism" data are proposed through analyzing adaptability characteristics original monitoring...