Rockburst disasters have become increasingly prevalent as distinct forms of subsurface geotechnical engineering advanced to the deep earth. Confronted with such a threatening subsurface geopressure disaster that poses a risk to personnel and equipment safety, the microseismic monitoring technology has been employed to track signals generated from rock fracture and collapse in the field. To guide the prevention and control of the hazard, the investigation conducted an effective microseismic data mining method. Through deep feature engineering and interpretable intelligence, a practical and available short-term prediction approach for the rockburst intensity class was developed. On the basis of rockburst case database collected from various underground geotechnical engineering, the neural network-based feature extraction method was conducted in the process of model training. The optimized model was obtained by combining the K-fold cross-validation approach with the structural parameter search methodology. The evaluation among the considered artificial intelligence models on the testing dataset was conducted and compared. Through analyses, the interpretable coupling intelligent model combining convolutional and recurrent neural networks for rockburst prediction were demonstrated with the most robust performance by evaluation metrics. Among them, the proposed adaptive feature extraction method leads the benchmark method by 6% for both accuracy and precision; meanwhile, the proposed metric generalization loss rate (GLR) for accuracy and precision in the validation–testing process reached 1.5% and 0.2%. Furthermore, the Shapley additive explanations (SHAP) approach was employed to verify the model interpretability by deciphering the model prediction from the perspective of the fined impact of input features. Therefore, the investigation demonstrates that the proposed method can predict rockburst intensity with robust generalization and feature extraction capabilities, which possess substantial engineering significance and academic worth.

Understanding the evolution of coal pore–fracture structures under coupled stress paths and creep deformation is critical for enhancing coalbed methane extraction and preventing coal and gas outbursts. In this study, coal samples from the Ningtiaota Mine were investigated using online Nuclear Magnetic Resonance (NMR) technology combined with triaxial loading–creep coupled experiments. The dynamic evolution of pore–fracture structures (PFSs) under different deviatoric stress levels was characterized and visualized in real time and across multiple scales. The results reveal a pronounced stress-dependent pore evolution during creep. Under low-stress conditions, seepage pores were compressed and gradually transformed into adsorption pores, whereas under high-stress conditions, seepage pores expanded and interconnected, dominating deformation and failure. Fractal theory was employed to quantify pore structure complexity, and repeated experiments demonstrated a significant positive correlation between the fractal dimension and the fractional order. Based on these findings, a fractal-dimension-based fractional creep model was developed by introducing a Riemann–Liouville fractional dashpot. The proposed model accurately captures the nonlinear creep behavior of coal and provides a microstructural interpretation of the fractional order. This study provides theoretical and experimental support for long-term stability assessment of deep coal–rock masses and prediction of coalbed methane migration.

Most of the mine dumps have very prominent slope stability problems. In this paper, by analyzing the gradation characteristics and spatial distribution characteristics of the discharged particles, the functional relationship between the characteristic particle size parameters and the relative height of the dump was obtained. Based on Dagushan environmental and climate characteristics, the indoor triaxial shear test under freeze–thaw cycle conditions was used to study the strength characteristics of discharged particles. The experimental results showed that the particle size of the dump presented a horizontal stratification phenomenon of “small at the top and large at the bottom,” with the content of fine particles (<20 mm) increasing from 13.83% at the bottom to 59.06% at the top. With the increase of the number of freeze–thaw cycles, the damage strength decreased gradually. Under the confining pressure of 50 kPa, the failure strength decreased from 335.8 kPa (0 cycles) to 160.5 kPa (24 cycles), and when the freeze–thaw cycle reached 24 times, the damage strength became stable. The increase of confining pressure in a certain range (50–150 kPa) can effectively inhibit the deterioration effect of freeze–thaw cycle on the damage strength of waste: the failure strength reduction rate at 150 kPa was 26.89% (24 cycles) compared with 52.20% at 50 kPa.

Clarifying the evolution of pore-fracture structure (PFS) and the associated seepage mechanisms of deep coal under coupled seepage and mining-induced stress is important for the safe and efficient in situ fluidized mining of deep coal resources. In this study, a nuclear magnetic resonance (NMR) online triaxial testing system was used to conduct coupled seepage–mining-induced stress tests under different seepage pressures, following a mining-induced stress path characterized by increasing axial stress and decreasing confining pressure. Transverse relaxation time (T₂) spectra and nuclear magnetic resonance imaging (NMRI) were combined to characterize the dynamic evolution of PFS in terms of its spatial distribution, pore volume, mean pore size, pore compressibility, and fractal characteristics. The results show that, based on the NMRI characteristics, the deformation and failure process of coal can be divided into three stages: compaction and elastic deformation, PFS propagation, and post-peak failure. Across the tested specimens, higher seepage pressure was associated with an earlier onset of PFS propagation, a lower PFS damage threshold, and enhanced PFS propagation and connectivity. The volumes of adsorption pores (APs), seepage pores and fractures (SPFs), and total pores (TPs) generally increase initially and then decrease during the compaction and elastic deformation stage, increase slowly or remain relatively stable during the PFS propagation stage, and increase sharply at the peak-strength point, with the magnitude of the increase differing among the three tested specimens. The mean pore size, represented by T_2g, initially increases and then gradually stabilizes with increasing strain, followed by a rapid increase at the peak-strength point. For specimen M2, SPF exhibited a substantially stronger compressibility response than AP and TP at the peak-strength point. The fractal dimension of SPF remains relatively stable before the peak strain but decreases sharply at the peak strain, indicating reduced structural complexity and enhanced connectivity of SPF. These observations suggest that, in the tested specimens, SPF expansion and coalescence under coupled seepage and mining-induced stress were associated with the reorganization of the internal seepage pathways of coal. These findings provide an experimental basis for evaluating stress-dependent permeability, gas transport pathways, and seepage-related failure risks in deep coal under mining-induced stress.