Mining-induced earthquakes (MIEs) in underground coal mines have been a common phenomenon that easily triggers rock bursts, but the mechanism is not understood clearly. This research investigates the laws of focal mechanism and source parameters based on focal mechanism and source parameters analysis of MIEs in three frequent rock burst areas. The relative moment tensor inversion (MTI) method was introduced, and the way to construct the inversion matrix was modified. The minimum ray and source number conditions were calculated, and an optimized identification criterion for source rupture type was proposed. Results show that the geological structure, stress environment, and source horizon influence the focal mechanism. The tensile type sources can distribute in the roof and coal seam, while the shear types are primarily located in the coal seam. In the typical fold structure area, the difference in source rupture strength and stress adjustment between tensile and shear types is negligible, while the disturbance scale of tensile types is distinct. The shear types have higher apparent volume and seismic moment in the deep buried fault area but lower source energy. The apparent stress of the tensile types is higher than that of the shear types, representing that the stress concentration still exists in the roof after the MIEs, but the stress near the faults could be effectively released. In the high-stress roadway pillar area, the primary fracture of the coal pillar easily produces a continuous shear rupture along the dominant stress direction under the extrusion of the roof and floor. The source parameters (except apparent stress) of shear types are higher than tensile types and have higher dynamic risk. The results contribute to expanding the understanding of rock burst mechanisms and guide MIEs’ prevention.

With the increase in coal mining depth, engineering geological conditions and the stress environment become more complex. Many rock bursts triggered by two combined faults have been observed in China, but the mechanism is not understood clearly. The focus of this research aims at investigating the influence of two combined faults on rock burst mechanisms. The six types of two combined faults were first introduced, and two cases were utilized to show the effects of two combined faults types on coal mining. The mechanical response of the numerical model with or without combined faults was compared, and a conceptual model was set up to explain the rock burst mechanism triggered by two combined faults. The influence of fault throw, dip, fault pillar width, and mining height on rock burst potential was analyzed. The main control factors of rock burst in six models that combined two faults were identified by an orthogonal experiment. Results show that six combinations of two faults can be identified, including stair-stepping fault, imbricate fault, graben fault, horst fault, back thrust fault, and ramp fault. The particular roof structure near the two combined faults mining preventing longwall face lateral abutment pressure from transferring to deep rock mass leads to stress concentration near the fault areas. Otherwise, a special roof structure causing the lower system stiffness of mining gives rise to the easier gathering of elastic energy in the coal pillars, which makes it easier to trigger a rock burst. There is a nonlinear relationship between fault parameters and static or dynamic load for graben faults mining. The longwall face has the highest rock burst risk when the fault throw is between 6 and 8 m, the fault dip is larger than 65°, the mining height is greater than 6 m, and the coal pillar width is less than 50 m. The stair-stepping, imbricate, horst, and ramp fault compared to the other fault types will produce higher dynamic load stress during longwall retreat. Fault pillar width is the most significant factor for different two combined faults, leading to the rise of static load stress and dynamic proneness.

The increasingly severe state of coal burst disaster has emerged as a critical factor constraining coal mine safety production, and it has become a challenging task to enhance the accuracy of coal burst disaster prediction. To address the issue of insufficient exploration of the spatio-temporal characteristic of microseismic data and the challenging selection of the optimal time window size in spatio-temporal prediction, this paper integrates deep learning methods and theory to propose a novel coal burst spatio-temporal prediction method based on Bidirectional Long Short-Term Memory (Bi-LSTM) network. The method involves three main modules, including microseismic spatio-temporal characteristic indicators construction, temporal prediction model, and spatial prediction model. To validate the effectiveness of the proposed method, engineering application tests are conducted at a high-risk working face in the Ordos mining area of Inner Mongolia, focusing on 13 high-energy microseismic events with energy levels greater than 10⁵ J. In terms of temporal prediction, the analysis indicates that the temporal prediction results consist of 10 strong predictions and 3 medium predictions, and there is no false alarm detected throughout the entire testing period. Moreover, compared to the traditional threshold-based coal burst temporal prediction method, the accuracy of the proposed method is increased by 38.5%. In terms of spatial prediction, the distribution of spatial prediction results for high-energy events comprises 6 strong hazard predictions, 3 medium hazard predictions, and 4 weak hazard predictions.

Coal burst prediction is an important research hotspot in coal mine production safety. This paper presents FDNet, which is a knowledge and data fusion-driven deep neural network for coal burst prediction. The main idea of FDNet is to extract explicit features based on the existing mine seismic physical model and utilize deep learning to automatically extract the implicit features of mine microseismic data. The key innovations of FDNet include an expert knowledge indicator selection method based on a subset search strategy, a mine microseismic data extraction method based on a deep convolutional neural network, and a feature deep fusion method of mine microseismic data based on an attention mechanism. We conducted a set of engineering experiments in Gaojiapu Coal Mine to evaluate the performance of FDNet. The results show that compared with the state-of-the-art data-driven machines and knowledge-driven methods, the prediction accuracy of FDNet is improved by 5% and 16%, respectively.

With the increase of mining depth, rock burst disasters frequently occur in steeply inclined coal seams. Firstly, this paper analyzes the rock burst of 5521-20 working face in Yaojie No. 3 coal mine and summarizes the characteristics of rock burst in horizontal section mining of steeply inclined extra-thick coal seam (SIETCS). Then, the static load distribution characteristics and the influence of dynamic load in the horizontal section mining of SIETCS are systematically studied by combining theoretical analysis with numerical simulation. On this basis, the mechanism of rock burst in horizontal section mining of SIETCS is put forward, verified by actual measurement. The results show that the SIETCS is “clamped” under the combined action of the same change trend of roof and floor. The maximum principal stress peak values on the roof and floor sides reach 22.0 MPa and 20.5 MPa. The maximum shear stress earned 8.7 MPa and 8.4 MPa, which makes the shear stress concentration in the coal body high and tends to “shear dislocation.” Under this “shear-clamping” action, an approximate “trapezoidal” plastic zone and a “rectangular” stress concentration zone are formed under the section. With the increase of mining depth, the “shear-clamping” action of SIETCS becomes more and more intense. When the roof cantilever reaches the ultimate span and breaks, the intense dynamic load increases the shear stress and failure of coal, which is easy to induce rock burst. The superimposed load greatly affects the area from the roof side to the middle of the working face, and the rock burst is intense. The rock burst is weak on the floor side due to the pressure relief of the surrounding plastic zone. The monitoring results show that the supports pressure and MS events activity on the roof side and near the middle part of the working face is considerable, while the floor side is opposite, which verifies the research results.