Addressing challenges in the hydraulic systems of reverse-pitch drilling rigs—such as the difficulty in early motor fault identification, weak signal characteristics, and complex operating conditions—this paper proposes a Multi-Attention Multi-Modal Convolutional Neural Network (MAMCNN) motor fault diagnosis method. This approach centres on monitoring the current signal of the hydraulic pump drive motor. It employs sliding window segmentation to extract raw time-domain waveforms, frequency-domain energy spectra (FFT), and multi-dimensional statistics. Continuous wavelet transform (CWT) is then applied to convert the signal into a two-dimensional time-frequency image, comprehensively reflecting energy distribution and frequency disturbance characteristics during the fault process. The constructed neural network comprises two parallel branches: the one-dimensional branch models the dynamic evolution of the current signal through a convolutional neural network (1D-CNN), bidirectional gated recurrent units (BiGRU), and an attention mechanism; the two-dimensional branch takes the CWT image as input, extracting time-frequency texture information via a two-dimensional convolutional layer and a channel attention module. Following feature fusion, both streams are fed into a fully connected classifier to achieve high-precision identification of common hydraulic pump motor faults (such as rotor bar breakage and eccentricity faults) versus normal operating conditions. Experimental results demonstrate the method's robust interference resistance and generalisation capabilities, rendering it suitable for intelligent monitoring and early fault warning of reverse-circulation drilling rig motors under complex operating conditions.

Mine hoists operate under high loads for extended periods of time. Early fault diagnosis and predictive maintenance of their key rotating components, the bearings, are important for the safe and efficient operation of hoisting systems. Existing diagnostic methods suffer from performance degradation in the hoist's complex working conditions. Therefore, an improved deep transfer learning diagnostic model was proposed, combining the fault response mechanism and signal time-frequency analysis techniques, and a cross-condition bearing fault diagnosis algorithm model CTJ-FDM was established. Firstly, a wavelet gram matrix group representation method was proposed. This constructed a matrix representation structure of the internal multi-scale features of the vibration signal's different frequency bands. Secondly, a multiple residual block parallel deep feature extraction network and a feature weight adaptive allocation mechanism were designed to realize the fusion extraction of deep features in different frequency bands. Finally, a more stable domain-adaptive deep transfer diagnostic network structure was created by combining multilayer maximum mean discrepancy loss and a domain adversarial mechanism to enhance the model's domain adaptation capability. Based on this, an online diagnosis and predictive maintenance system for the hoist's key components was developed. Following experimental validation, the cross-condition average accuracy of the model on two bearing test benches was 99.81% and 99.39% respectively, indicating improved cross-condition diagnostic capability.

In response to the problems of low knowledge utilization rate, single reasoning method and fixed knowledge base in traditional fault diagnosis expert systems, a fusion reasoning expert system based on fuzzy fault tree is proposed. Firstly, a fuzzy fault tree for belt conveyor was established, and the qualitative and quantitative analysis of faults was realized by combining the intuitionistic fuzzy algorithm, forming a standardized knowledge base and case base for the expert system; Secondly, a knowledge case fusion reasoning algorithm was designed as the reasoning engine of the expert system, enabling it to realize real-time monitoring and reasoning of faults by combining the state monitoring module; Finally, a knowledge base dynamic update algorithm based on BERT model was developed, converting case data actively into knowledge rules to improve the operation and maintenance efficiency of the system. The performance of the expert system was tested through experiments, and the experimental results show that: the average diagnostic accuracy of the proposed expert system can reach 94.75%, and the average reasoning delay is 167.31ms; The designed knowledge base dynamic update algorithm significantly improves the operation and maintenance manpower savings in large case library environments; The key performance parameters of the system increase with the increase of the number of cases, and the expert system shows good scalability and growth potential.

The roller of belt conveyor is susceptible to rust, wear and other faults in long-term operation. The early fault acoustic signal is weak and easily submerged by strong noise in the industrial environment. The traditional diagnosis method has the problems of insufficient feature extraction and poor robustness. In this paper, a fault diagnosis method based on CNN-Swin Transformer dual-branch feature fusion network (CSF) is proposed. The time-frequency domain feature matrix is constructed by fusing variational mode decomposition and fast Fourier transform. Combined with the local feature extraction advantages of CNN and the global attention mechanism of Swin Transformer, the SE-CGA attention mechanism is designed to achieve deep feature extraction. Experiments show that the method achieves a test accuracy of 98.11 % on real industrial data sets, which is more than 9 % higher than that of a single CNN model. The recognition accuracy of 65.59 % is still maintained when superimposing-15 dB extreme noise, which verifies its diagnostic robustness and engineering application value in strong noise scenarios.

Due to the longitudinal tear fault detection of the conveyor belt of the underground belt conveyor, it is often impossible to take into account the detection speed and detection accuracy. When the detection accuracy is high, it may lead to tears that are not detected in time and are rolled into the roller or roller, causing secondary safety accidents. When fast detection speeds are met, false alarms are prone to trigger unnecessary downtime and failure to identify true tears, leading to the latent development of faults and ultimately catastrophic damage. Therefore, this paper proposes an AFA-YOLO deep learning algorithm based on the improved YOLOv11. By introducing the Downsampling Module (ADown), the Shared Convolutional Feature Pyramid (FPSC) and the Auxiliary Detection Head (Aux). Experiments on the longitudinal tearing dataset of the conveyor belt simulated underground mine show that the mAP@0.5 value of the AFA-YOLO model reaches 96.30%, which is 3.55% higher than that of YOLOv11, and the computational complexity (GFLOPs) is reduced by 17.19%. The AFA-YOLO model can achieve the optimal balance of accuracy and speed to meet the detection needs of complex downhole environments.

The mine main ventilation fan, as the core of mine safety operations, consists of two cold storage units of the same type of equipment, which often need to be switched and maintained periodically to ensure its efficient operation. For such systems, the characteristics of alternating equipment operation lead to system failures that are affected by the coupling of equipment states, and different repairs correspond to different effects, times and costs and have different impacts on the subsequent operation of the system. Therefore, the failure analysis of the system and the reasonable maintenance strategy are crucial for the system's reliable operation. This research takes the Mine main ventilation fan as the research object. Firstly, based on the analysis of structure and failure mode, the offline switching maintenance strategy is defined, and the concept of maintenance degree is introduced to model the effect, time, cost and frequency of maintenance in a unified way. Secondly, the stochastic failure model of the system under the influence of the offline switching maintenance and the maintenance degree is deduced. Then, a cost rate model is constructed in the finite time domain to optimize the maintenance period and the maintenance degree. Finally, the validity of the model is verified by numerical experiments. The results show that the offline switching maintenance strategy considering the maintenance degree can effectively reduce the operation and maintenance cost of the mine main ventilation system. Meanwhile, the change of the scrap cost of components has a more sensitive impact on the optimal cost rate of the system, and the manager can further improve the system's economic efficiency through the refinement of the management of the scrap cost of the equipment.

In order to reasonably determine the development plan for the Molong mine field, based on the terrain and landform of the mine field, the characteristics of coal seam occurrence, and the external transportation conditions, following the principle of adapting measures to local conditions, combined with the proposed selection of the industrial site location, three development plans were proposed. After technical and economic comparison, it was determined that the Molong mine field would adopt the industrial site in the southeast part of the southern area and inclined shaft development plan. This plan can maximize the exploitation and utilization of resources, facilitate the scientific and balanced production of the Molong mine field, and provide technical support for the efficient production of the mine.

In view of the blind design, large amount of filling and high cost of treatment in the room and pillar mining goaf, the partial filling design method of roof-uncontacted filling and pier column filling was put forward. Based on the characteristics of the rectangular coal pillar gradually peeled by weathering process and the accumulation of peeled bodies, the model of the rectangular coal pillar peeling in the room and pillar mining goaf was builded, and the maximum peeling depth of the rectangular coal pillar was calculated. Based on the collaborative bearing of coal pillar and backfill, the theoretical calculation formulas of backfill with roof-uncontacted filling and pier column filling were derived by using safety factor method for coal pillar ultimate strength. These formulas could guide the design of partial filling treatment of room pillar goaf and reduce the cost of goaf treatment greatly. The results showed that the design strength parameters (elastic modulus, uniaxial compressive strength) of the backfill were inversely proportional to the filling rate, so the elastic modulus and uniaxial compressive strength of the filling material could be increased to significantly reduce the filling volume in the goaf and the treatment cost.

To address the parameter design challenges of gob-side entry retaining in gangue filling ultra-long workfaces, comprehensive research methods including theoretical analysis, mechanical calculations, and numerical simulation were adopted. An innovative box-form filling body casting technology for gob-side entry retaining was proposed. The technical and economic advantages and disadvantages of six gob-side entry retaining methods were analyzed, and key technical parameters for gangue filling ultra-long workfaces were optimized. The results show: theoretical calculations determined the reasonable width of the roadside support body as 2.94 m with a strength of 8.08 MPa; simulations revealed that when the roadside support body width exceeds 3 m, the gob-side entry retains over 85% of its original cross-section, effectively controlling deformation and failure, showing high consistency with theoretical results and verifying parameter accuracy; quantitative analysis indicates the roadside support body width has significantly higher influence weight on surrounding rock deformation control than gangue filling ratio; as gangue filling ratio increases, the cross-section shrinkage rate first stabilizes then decreases, with no significant change observed when gangue filling ratio is below 75%.

To address the failure of wear-resistant linings in large silos under high-energy impact, this paper proposes an innovative prestressed metal-microcrystalline ceramic composite liner. Utilizing the lost-foam casting process, a compressive prestress of 10–20 MPa is generated within the ceramic phase through precisely controlled cooling and differential thermal expansion between materials, significantly enhancing impact resistance and wear performance. An innovative "groove-arch" composite structure was designed for the top of the distribution beam, which retains fine particles to form a self-buffering layer and optimizes impact force transmission. Engineering applications demonstrate excellent performance of the liner under complex conditions including high-intensity impact, corrosion, and high-humidity environments, effectively resolving technical issues such as impact-induced fragmentation, corrosion spalling, and connector shear failure, providing an effective solution for silo wear protection.

Aiming at the problem of poor sealing performance in the treatment process of Ordovician aquifer due to improper selection of grouting materials, The differential influence of drilling fluid on the rheological/thixotropic behavior of cement slurry and clay slurry was analyzed from both physical and chemical perspectives through testing methods such as rheology/thixotropy analyzer, particle size distribution analyzer, Zeta potential analyzer, organic carbon adsorption analyzer, and isothermal calorimeter. The sealing effect of drilling fluid on different grouting materials was verified through the application engineering of Mengjin Coal Mine in Xin'an Coalfield. Results show that: (1) Drilling fluid significantly increases the rheological properties (dynamic yield stress, plastic viscosity) and thixotropy of cement slurry and clay slurry, and exhibits a dose-dependent effect. At the same dosage, the rheological parameters of clay slurry are more than 35% higher than those of cement slurry, and the thixotropy is more than 2.5 times higher than that of cement slurry. (2) By analyzing zeta potential and adsorption capacity, cement slurry with positive zeta potential (+1.39 mV) and drilling fluid with negative zeta potential (-35.8 mV) achieved efficient adsorption through electrostatic attraction. The saturated adsorption capacity (330 ppm) and adsorption efficiency (100%) of cement slurry on drilling fluid are 2.4 times and 1.25 times that of clay slurry (-13.4 mV) on drilling fluid, respectively. the decisive role of electrical matching in adsorption behavior was elucidated. (3) Based on electrical matching mechanism, the technology for early adsorption blocking of cement slurry is proposed. The application project in Mengjin coal mine (leakage rate >30 m3/h) of Xin'an coal field shows that the rise rate of hole pressure (0.045 MPa/h) after cement slurry adsorption of drilling fluid is only 5% of that of clay slurry system (0.86 MPa/h), and the plugging response efficiency is improved by 20 times. This study reveals the interaction mechanism between the grouting material and drilling fluid from a multi-scale perspective, and provides a theoretical basis for the design and engineering application of highly efficient plugging materials for the coal mine Ordovician aquifer.

The width of the plastic zone in the vicinity of rock around a roadway is a crucial factor for determining the support design parameters. This study investigates the deformation and failure patterns of the rock surrounding the roadway by examining the effects of varying lateral pressure coefficients, rock types, and depths of roadway burial through theoretical analysis, numerical simulation, and field testing. The findings reveal several key points : (1) With lateral pressure coefficient as the variable, the plastic zone's irregularity around the roadway increases as the coefficient decreases. The shape of the plastic zone evolves from circular to elliptical, then to a butterfly pattern, and finally stabilizes in a butterfly form. The plastic zone at the roof and floor narrows while that at the sides widens, with the roof and floor changes being more pronounced. At a lateral pressure coefficient of 0.8, the plastic zone's distribution and width at the roof, floor, and sides of the roadway are similar. (2) When considering rock type, the plastic zone width in fine-grained sandstone is significantly less than in coal seams. The stronger the overall rock, the smaller the plastic zone's failure width. Shear failure is the primary mode of plastic zone failure, with greater deformation at the sides compared to the roof and floor. Different rock types influence the plastic zone's width but not its shape. (3) Based on the plastic zone's growth pattern in the rock surrounding the roadway, the use of high pre-stressed bolts (or cables) is recommended, with extended lengths and increased diameters to enhance the elongation and strength of the bolts (or cables). Field tests indicate that the optimized support system effectively manages the separation and deformation failure of the roadway. The rock's deformation is kept within 300 mm, significantly enhancing its internal stability and reducing roadway deformation.

In order to explore the stress deflection and stability analysis of roof under the support of gangue in large dip angle working face, the fracture evolution characteristics of gangue support area and roof rock beam under unbalanced filling effect are studied based on theoretical analysis. Combined with numerical calculation, the asymmetric space-time deformation and stress deflection response law of coal rock in stope are analyzed, and the influence mechanism of gangue on the mining dynamic behavior of surrounding rock in large dip angle stope is revealed. The research results show that : affected by the gravity inclination effect, the gangue support filling area changes with the coal seam inclination angle and coal thickness. The length of the gangue support area in the working face is 35.44 m, and the roof is prone to breakage and instability at 75 m. The unloading of coal seam mining leads to the formation of stress concentration at the upper and lower ends and the gangue filling area. The direction of the principal stress vector of the overburden rock is asymmetrically deflected. The principal stress deflection angle gradually decreases from the middle and upper regions to the two sides of the boundary. The peak value of the principal stress deflection angle is 70.05 °. The change rate of the principal stress difference and the shear stress of the rock stratum decrease with the increase of the principal stress deflection angle. The caving gangue bears part of the overburden load, and the stress transfer path changes. There are two asymmetric stress deflection boundaries in the rock layer. The advance abutment pressure of the working face decreases as a whole, and the decrease in the lower part of the tendency is the largest. The peak abutment pressure is transferred to the lower part of the non-filling area.

Taking Xinjiang Zhundong No.2 Mine as the engineering background, the typical weakly cemented overburden 1101 fully mechanized caving face is selected as the research object. Through numerical simulation and field measurement, the advancing speed effect of the damage characteristics of the overlying strata under high-intensity mining is studied. The results show that: (1) The numerical simulation results show that the development height of the plastic zone of the overlying rock on the 1101 working face is 181.53m ; the damage degree of overburden rock and the development height of plastic zone decrease with the increase of working face advancing speed. In the process of advancing speed from 3 m / d to 10 m / d, the damage degree of overburden rock decreases from 21.66 % to 12.07 %, which decreases by 9.59 % as a whole. When the advancing speed is 7 m / d, the change rate of overburden rock damage degree is the fastest, and it is considered that the advancing speed of working face is 7 m / d. (2) The relationship between the mine pressure behavior and the advancing speed of the working face is obtained by the field measurement. Within a certain range, the greater the advancing speed, the greater the periodic weighting step distance of the working face. The average advancing speed is 2.5 m / d and 3.5 m / d, and the corresponding average periodic weighting step distance is 18 m and 23.4 m, respectively. The advancing speed to ensure the best state of the support resistance is 5 m / d. (3) Microseismic data indicate that the maximum height of the water-conducting fracture zone in the overlying rock layer is 191 meters, and the development trend is basically consistent with the numerical simulation results. The damage development of the overlying rock has gone through three stages: "low-speed development stage - rapid development stage - stable development stage". (4) Based on the results of numerical simulation, mine pressure behavior and microseismic response, it is considered that the increase of working face advancing speed can effectively reduce the development of overburden damage, strong mine pressure behavior and microseismic energy response characteristics after the 1101 working face of Zhundong No.2 Coal Mine is mined and squared. It is advisable to control the advancing speed of the working face at 3.5 ~ 5 m / d.

In response to the disaster risks faced by deep, extra-thick coal seams along development roadways, such as the superimposition of dynamic and static loads and complex overlying rock structures, this paper takes the 1012006 working face of Yuanzigou Coal Mine in Linbei, Shaanxi, as the engineering background. A combination of numerical simulation, theoretical analysis, and field measurement methods was used to systematically study the disaster mechanism and source-specific prevention and control techniques for rock bursts along development roadways. The results show that the lateral gob 'F'-type overlying rock structure and the wide coal pillar 'elastic core' jointly affect the development roadway, creating a stress environment with superimposed high static and dynamic loads, which easily induces bursts. A multi-level prevention and control system of 'optimized design–source-specific pressure relief–reinforced support' was proposed, including key technologies such as optimization of the cut positions in the working face, zoned pressure relief with large-diameter boreholes, and coordinated anchor-net and cable support. Field applications indicate that this system can significantly reduce microseismic energy release, alleviate stress concentration, effectively control roadway deformation, and achieve proactive rock burst risk prevention. This study provides theoretical basis and practical reference for the prevention and control of rock bursts under similar geological and mining conditions.

In order to accurately predict the height of the water-flowing fractured zone of fully mechanized mining face, based on the analysis of the evolution of overlying rock structure, the mine pressure, water inflow and microseismic events, the concept of the main control roof of the water-flowing fractured zone of overlying rock is presented. Then the composite structure model with horizontal three-zone supporting body of the thick and hard overlying rock is es-tablished, and the stiffness calculation formula of the "three-zone composite supporting body" is given. The calculating formula of the critical periodic caving span of overlying rock is ob-tained, and the method to predict the height of the water-flowing fractured zone based on the main control roof is presented. The method has been successfully used in 31101 fully mecha-nized mining face in Nalinheer mine. The results show that the main control roof of the wa-ter-flowing fractured zone is the first unbroken thick hard rock from bottom to top in the overburden rock, and the main control roof is supported by the composite support body com-posed of the in-situ rock zone, the hydraulic support supporting zone and the supporting zone of the caving collapse roof. Due to swelling and evolution of relative compaction coefficient of the collapse roof, the supporting stiffness of the caving collapse roof zone takes on "slow growth" and "fast growth" two-stage characteristic. The critical periodic caving span is the square span of the working face. The measured height of the water-flowing fractured zone of 31101 working face is 17.21. The two empirical formula in “Code for the Design of Coal Pillar and the Extraction of Coal for Buildings, Water Bodies, Railways, and Main Tunnels” are em-ployed, and predicted the height of the water-flowing fractured zone are 8.51 and 9.83, respec-tively. The predicting result of the height of the water-flowing fractured zone by the method in this paper is 19.58. The predicting result from the method in this paper is closer to the measured result. The method of predicting height of the water-flowing fractured zone in this paper takes into account the influences of overburden structure, distribution of thick and hard rock, mining height, swelling characteristics of collapse rock and other factors on the height of the wa-ter-flowing fractured zone, and it provides a new way to predict the height of the water-flowing fractured zone of fully mechanized mining face.

Aiming at the problems such as long ventilation distance of high-temperature and heat-hazardous mining area in deep mining of mines, long distance of underground cooling and cold transmission by ground industrial plaza refrigeration, and insufficient cooling and heat dissipation source of underground refrigeration, we put forward the technical program of travelling through strata cold transmission by ground level refrigeration and underground cooling.Through the structural construction and heat transfer analysis of the multi-layer composite heat preservation cold pipe, and experimental study on mechanical properties of filled insulation cement paste, and engineering implementation applications of“drilling first,finding after”construction process for traversing through strata, the technical program of travelling through strata cold transmission by ground level refrigeration and underground cooling was validated as Feasible.The cooling system is safe, stable and reliable, The cooling system is safe, stable and reliable, and the effect of underground centralized cooling system is remarkable, which can effectively meet the cooling demand of the working face, and effectively solves the problem of Heat dissipation challenges of cooling and refrigeration in the well, shortens the cooling distance, ensures the cooling efficiency.It also avoids the problem of coal safety and explosion-proof of the refrigeration unit in the well, and the refrigeration system is easy to maintain which has gained good economic and social benefits.

In order to further study the overburden movement law, mine pressure manifestation law, and deformation and failure mechanism of self-formed roadway during the whole process of mining without coal pillar, a three-dimensional physical model test system of self-formed roadway was developed. The test system is com-posed of reaction main frame system, hydraulic servo loading system, coordinated control and monitoring system, etc. Through the test system and the self-developed unit plate, unit strip and unit block structure, the fast excava-tion and mining of the roadway are realized. Through the pre-installed internal displacement, stress, strain moni-toring equipment, can be real-time monitoring of the whole working face of the whole process of cutting and un-loading pressure into the lane. Based on the engineering background of Anju Mine, the first coal mine over a kilometer deep mine in China, a large-scale physical model test of 110 mining method was carried out, which re-vealed the mining pressure development rule of 110 mining method, and clarified the deformation control mech-anism of surrounding rock of self-formed roadway. The model test results are basically consistent with the field monitoring data, which verifies the rationality and effectiveness of the development and application of the model test system.

Backfilling mining can achieve the coordinated development of green mining in coal mines and solid waste treatment. For the cemented backfilling mining process, the conveying performance of cemented backfilling slurry is a key indicator for evaluating the conveying of slurry from the surface backfilling station to the goaf. Taking the content of cement in fine materials (cement + fly ash) X1 (28%~40%), fine-aggregate materials ratio (the ratio of fine material to aggregate gangue) X2 (0.44~0.76)and mass fraction X3 (78%~82%) as the analysis factors, the response laws of the conveying performance of cemented gangue backfilling slurry under the influence of the three factors and their interactions were analyzed by using the response surface method (RSM). The results showed that the rheological parameter (yield stress and viscosity coefficient) of slurry increased with the increase of the above three factors, while the change trend of expansion and bleeding rate were contrary to the above properties. Overall, mass fraction X3 is the most significant factor influencing the performance of slurry. The response surfaces of each performance to the interaction terms (X1X2, X1X3, X2X3) were established. In the influence of the interaction terms, the increase of X1, X2 and X3 promoted the influence of another factor on the expansion and rheological parameter, while weakened the influence of another factor on bleeding rate. The expressions of the unit along-the-way resistance loss and each factor were established respectively, the unit along-the-way resistance loss also increased with the increase of the three factors.

To investigate the failure mechanisms of coal samples with impact tendency under combined static and dynamic loading, split Hopkinson pressure bar (SHPB) tests were conducted under varying loading conditions. High-speed camera monitoring and macroscopic fragment size analysis were employed to examine the dynamic response characteristics of the coal samples. The results demonstrate that the dynamic strength of the coal samples exhibits a positive correlation with impact air pressure but a negative correlation with axial pressure. Notably, the dynamic strength initially increases and then decreases with rising axial pressure. Under constant static axial load, higher impact air pressure leads to an increased proportion of small-sized fragments and a rising fractal dimension, indicating greater fragmentation. Conversely, under fixed impact loading, sample damage first decreases and then increases with static load, while the fractal dimension follows a decreasing-then-increasing trend, suggesting that the coal’s resistance to dynamic loading initially strengthens before weakening as axial pressure increases. When the static load exceeds 50% of the uniaxial compressive strength, the samples become significantly more susceptible to dynamic failure. These findings provide valuable theoretical insights for the prevention and control of dynamic disasters in deep coal and rock masses.

In response to the current coal mine gas disaster warning mostly focuses on disaster classification warning, disaster-causing factors and related laws and regulations, preventive measures failed to be associated in time, the disaster warning information content is complicated, the information fragmentation is serious, the presentation form is single, and it can't form the disaster warning prevention and control system well. The author proposes a research on coal mine gas disaster prediction and early warning technology based on knowledge mapping, utilizing the Interpretive Structural Model (ISM) for the construction of coal mine gas disaster indicator system, and then applying the Bayesian Network Model (BN) for the early warning of coal mine gas disaster; constructing the coal mine gas disaster mapping through the extraction of the knowledge entities of the gas disaster indicator system, gas laws and regulations, rules and regulations, etc.; and combining the Bayesian Network Model and the gas disaster warning information with the gas disaster indicator system, and the knowledge entities of the coal mine disaster prediction and early warning information. The combination of Bayesian network model and gas disaster mapping realizes the mapping construction of coal mine gas warning and prevention system, which facilitates the timely sending of gas warning information and the rapid response of prevention and control measures.

The water inrush risk assessment method is the theoretical basis for the design and optimization of mine production and drainage system. In order to construct a more scientific and rigorous risk assessment system for coal seam roof water inflow, the AHP (analytic hierarchy process)-entropy weight method was used to optimize the comprehensive weight of the influencing factors of water inflow, and the main controlling factors of coal seam roof water inflow were determined. Based on the sample data of six indexes, including permeability coefficient, aquifer thickness, pore water pressure, temperature and cohesion, a coal seam roof water inrush evaluation model was trained with the logical framework of "data input-model output-field verification-performance improvement-simulation prediction". The results show that the thickness of the aquifer predicted by the KNN model has the greatest impact on the water inflow of the roof of the Galutu coal seam, followed by the permeability coefficient. The accuracy of the KNN prediction model is 0.9598 and F1-score: 0.9569, which verifies the rationality of the selection of the evaluation index of coal seam water inrush risk and the feasibility of the model prediction. It is predicted that the auxiliary transportation of 2105 working face will be in the high-risk area in the range of 0~330m and 1620~2210m from the cutting hole of the working face, and the research results have expanded the risk assessment method of underground coal seam water inflow.

Discrimination of surface movement duration and residual movement deformation in multiple coal seams under repeated mining conditions is complicated. At present, there is no function equation of surface movement and deformation time for the whole life cycle under large mining depth and repeated mining conditions, and there is no zoning criterion for comprehensive treatment of subsidence area. Based on the high-precision large-deformation monitoring characteristics of surface movement observation station in the short to medium term, combined with the small deformation, long period and slow deformation characteristics monitored by InSAR, the paper applies the monitoring data of "conventional surface movement observation station +InSAR residual deformation" to fit surface movement deformation-time function equation of the mine life cycle. Surface movement cycle is divided into four stages. Based on the relevant rules and regulations, the regional criteria for stability of vegetation restoration and treatment in coal mining subsidence area are established. Based on the calculation results of surface movement and deformation, the suitable areas of surface vegetation restoration and treatment in each coal mining subsidence area are defined, and the corresponding comprehensive treatment time is proposed according to the characteristics of surface movement and deformation. Results show that the surface movement and deformation of Fengshuigou Coal Mine last for 56 days at the beginning stage, 162 days at the active stage, 167 days at the decay stage, and the cumulative surface movement time is 385 days. The combination of surface movement observation station and InSAR monitoring data can realize the high-precision and life-cycle surface movement deformation-time function equation of mine under complex conditions. The whole process of surface movement and deformation can be divided into four stages: initial stage, active stage, decline stage and residual deformation stage. Based on the moving deformation-time function equation of mine life cycle, coal mining subsidence zoning can be divided into three types: unsuitable vegetation restoration and management area, selective suitable area and suitable area.

Surface subsidence monitoring of coal mines is of great significance for ensuring mining safety and assessing ecological disturbances. This study, based on unmanned aerial vehicle (UAV) oblique photography and multispectral remote sensing technology, utilized multi-period elevation data to invert dynamic subsidence basins and combined ground control points to verify the accu-racy. It revealed the spatiotemporal distribution characteristics and variation patterns of surface subsidence at the working face scale of a coal mine in Inner Mongolia, and calculated the vegeta-tion coverage change ratio between the subsidence area and non-subsidence area of the working face to clarify the impact of subsidence on vegetation. The results show that: (1) UAV oblique photography can achieve high-precision subsidence monitoring, with average errors of 1.48 cm and 2.60 cm in elevation inversion for the two working faces, and inversion errors of 3.37 cm and 1.97 cm in subsidence monitoring results; (2) Subsidence in the mining area shows signifi-cant spatiotemporal heterogeneity. The subsidence center of working face A moved northward due to the disturbance of the goaf, with a subsidence rate of ≥ 2.281 cm/d during the active period, and the subsidence value reached about 1.5 m within two months after mining and then stabi-lized. Due to the lag effect, the subsidence basin of working face B was not fully subsided when it was first monitored one month after mining, and the subsequent monitoring showed that the subsidence basin was mainly concentrated in the actual mining range of the previous month, with a subsidence value of about 2.6 m and then stabilized. A stepped subsidence was monitored, and the horizontal movement speed of the subsidence basin's centroid was 8.36 m/d, which was high-ly consistent with the actual mining progress; (3) The low vegetation coverage in the subsidence area of working face A significantly increased during the mining period, while the medium, me-dium-high and high vegetation coverage generally showed a pattern of "first decreasing and then increasing", with a certain increasing trend in medium and medium-high vegetation coverage after mining cessation. It indicates that mining activities caused vegetation degradation in high-cover areas and promoted growth in low-cover zones, while moderate and moderately-high cover areas also declined during mining but showed self-recovery ability after mining ceased. The find-ings provide a scientific basis for optimizing and precisely designing ecological restoration and land reclamation schemes.

To investigate the impact of coal mining on soil water movement, study plots were established in both mined-out areas and control areas within the Upper Bay and Burmese Terrace mining districts, using Artemisia ordosica as the indicator species. By employing hydrogen and oxygen isotope tracing techniques, we analyzed soil profile water content and plant water uptake strategies to systematically elucidate the mechanisms of soil water movement. The results indicate that structural damage to the soil in mined-out areas creates preferential flow pathways, leading to significantly higher deep-layer water content compared to control areas. Plant water-use strategies differed markedly: vegetation in control areas primarily absorbed deep, stable water sources, while plants in mined-out areas shifted towards utilizing shallower and mid-layer water sources. Soil water movement is collectively controlled by evaporation, preferential flow, infiltration, and plant water uptake. Isotopically labeled water injected at 170 cm depth was found to evaporate and migrate to the surface. Infiltration was stronger in mined-out areas, resulting in greater migration depths than in control areas. Coal mining alters soil structure and hydrological pathways, causing spatiotemporal redistribution of water movement and forcing vegetation to adapt their water-use strategies. The law of soil water transport in coal mining area was revealed from multi-dimension, which has certain practical guidance significance for promoting ecological restoration and sustainable development of arid mining area in Northwest China.