Advanced displacement prediction of deep roadway roof with data fusion using machine-learning models: a comparative study

Coal mining has a significant impact on the movement of the overburden, leading to potential safety hazards in the working face. In this paper, a similarity simulation experiment was conducted to investigate the migration of overburden during the mining process of a specific working face in the Liuzhuang Coal Mine located in southern China. Sand and gravel were used to simulate the geological environment of each rock stratum. The deformation of the stratum was monitored using strain gauges, the fracture and displacement changes of the overburden stratum were recorded using cameras, and the characteristics of roof collapse was monitored using infrared thermal imager. The experimental model fully simulated the situation of the working face, and the actual working face size was obtained by enlarging the model by 100 times. The experiment found that during the initial stage of mining, there was no significant subsidence of the roof. In the course of the advancement of the working face, the primary roof intermittently fractured behind the working face, with subsequent propagation of upper cracks in an upward direction. The overburden rock layer above the goaf experienced continuous compaction, leading to the gradual closure of the separation layer. Simultaneously, new cracks constantly emerged in front of the working face, resulting in the progressive stabilization of the height of the crack zone. The stress measurements at each point exhibit a pattern of initial increased, followed by decrease, and ultimately stabilization. By considering the stress variation law of the overburden rock, the stress changes in key layers of the bedrock during mining could be categorized into four zones: the stress stable zone, stress increasing zone, stress reducing zone, and compaction stable zone. During the initial phase of coal seam mining, the presence of rock layers provided support, resulting in minimal subsidence of the overburden rock. However, as the mining operation progressed, the disturbance force and collapse of the overburden rock leaded to further downward subsidence. When the working face reached the stop line, the collapsed overburden rock gradually consolidates, resulting in a deceleration of energy release and the formation of a pressure relief zone. Consequently, the overburden rock above the working face underwent a slight additional subsidence, reaching its maximum level. A short cantilever rock beam structure was formed in the experiment. This study will provide valuable reference for future coal mining and serve as a vital theoretical basis for roof control in deep coal seam mining.

Deep coal mining is unavoidable, and the complex mining environments and the increasing dangers associated with ultrahigh energy accumulation and release from mining disturbances renders it extremely difficult to maintain a safe and stable stope. Solid backfilling technology directly uses coal gangue and other solid wastes in the mining area to fill the gob after mining. Support from the backfill body can inhibit the movement of overlying rock strata and significantly alleviate the influence of mining. In this study, the correlations between the deformation of gangue filling material and the characteristics of energy dissipation were examined under lateral uniaxial compression. The strain energy density distributions of backfilling and caving mining methods were simulated using numerical modeling. The results showed that the strain energy density distribution of backfilling mining was less concentrated, and its peak value was lower than that of caving mining by 51.0%, indicating that backfilling could effectively reduce the amount of energy released from mining rocks. The dense backfill mining area of the No. 9301 face in Tangkou Coal Mine was used as a case study. Measures for controlling the backfill body compaction for reducing the amount of energy released from mining rocks were proposed. These measures include optimizing the support structure and filling material formula, controlling the preroof subsidence, and ensuring an appropriate number of tamping strokes. The monitoring results of the backfilling quality, surface subsidence, and microseismic energy of No. 9301 working face in Tangkou Coal Mine showed that when the backfill body filling ratio control value was 82.28%, the total number of microseisms and the amount of energy released from the mining working face were significantly lower compared to those of the caving method. This study demonstrated that the backfill body could effectively reduce the amount of energy released from mining rocks, thereby realizing management of mine earthquake and sustainable deep coal mining.

To prevent coal mine roof water damage, the water generally needs to be evacuated in advance. It can be mined with the water inrush risk assessed as safe. However, a single index is often employed in the water safety evaluation after the roof drainage, which causes a large gap between the evaluation results and the actual situation. Therefore, the evaluation cannot be effectively used to guide the safety mining in the working face. In this paper, based on the hydrogeological data of the Liangshuijing coal mine, a multifactor water inrush risk assessment model (IAHP-EWM) and multifactor index system are established for assessing the water inrush risk before and after the roof drainage. The improved AHP method and the entropy weight method are adopted in the model to determine the index weight. This combined way avoids the excessive subjectivity and objectivity of the index weight. A″ Fold undulation degree (Fud)″ is innovatively proposed to quantify the impact of the spatial relief of folds on water inrush in the multifactor index system. The IAHP-EWM model is applied to evaluate the risk of roof water inrush in the 42205 working face of the Liangshuijing coal mine. The evaluation results show that the water inrush risk is ″high″ when the water is not dredged, and the water inrush risk is ″low″ after the water is dredged, which are consistent with the actual water inflow data and evaluation results, which verifies the accuracy of the model. The application results of the IAHP-EWM model in the 42202, 42203, and 42204 working faces verify its universal applicability in the Liangshuijing mining area. It can provide a reference for the evaluation of the roof water damage control effect during coal seam mining.

Due to the concealment of the fire source in the gob, the fire prevention and extinguishing work in the gob is facing great difficulties. This study is made in order to realize the real-time monitoring of gob temperature and the accurate positioning of high temperature area, so that the fire prevention and extinguishing work in gob can be targeted. In this study, the previously developed COMBUSS-3D software was used to predict the high temperature area in the gob of 85001 working face of Yangmeiwu Coal Mine and II830 working face of Zhuxianzhuang Coal Mine in China, and the continuous monitoring system of gob temperature was independently developed to realize the real-time monitoring of gob temperature, achieving the purpose of accurate positioning of high temperature area in gob. The results show that the high temperature area of the gob of 85001 working face of Yangmeiwu Coal Mine was in the range of approximate circle centered on the point (36.6, 30), and the maximum temperature was 31.7°C. The high temperature area of the gob of II830 working face in Zhuxianzhuang Coal Mine presented an approximate ellipse centered on (28.2, 25), and the long axis was parallel to the working face, and the maximum temperature was 43.9°C. The research results are expected to provide reference for the early prediction of spontaneous combustion in gob.

Based on the production conditions of the 10103 excavation working face of the Baozigou coal mine, this paper analyzes the potential sources of H2S and the expected emission concentrations of H2S in the working face. Considering the previous engineering practice for controlling H2S disasters in coal mine working faces, numerical simulations were conducted to investigate air flow and H2S migration and diffusion in the tunnel in the excavation working face. The migration and distribution of H2S in the coal seam mining face were studied, and the effects of outlet wind speed, duct location, and duct diameter on the H2S concentration distribution were explored. The higher the outlet wind speed, the more conducive to the emission of H2S gas, but too high a wind speed will be detrimental to the concentrated extraction and purification absorption of H2S; the closer the outlet position of the air duct is to the end of the working surface, the lower the H2S concentration in the vortex area at the corner; the air duct If the diameter is too small, the harmful gases released from hard-to-break coal cannot be entrained and taken away. When the diameter of the air duct is too large, the entrainment volume during the jet process will be expanded. To verify the field distribution of H2S concentration at the bottom, middle, and top of the boring machine, a CD4-type portable H2S instrument was used to analyze the distribution of H2S near the excavation working face.

Coal mining in areas with deep confined water is very dangerous; to ensure safety, it is necessary to clarify the damage characteristics of the working face floor. To directly reflect the failure characteristics of the working face floor under the coupled effects of mining stress and confined water pressure, this study considers the mining above confined water in the deep coal seam of the II633 working face of the Hengyuan coal mine in the Huaibei mining area as the engineering background. With the use of the similar material simulation experimental method and a self-designed monitoring system for confined water diversion and a confined water loading system and a confined water lifting system that can directly reproduce the floor confined water lifting characteristics affected by floor failure during coal mining, the mining stress distribution patterns, deformation and failure characteristics of the overburden, and diversion characteristics of the confined water in the working face floor are investigated. The results show that the floor undergoes three stages of deformation in the horizontal direction: premining stress concentration-related compression (10–15 m ahead of the working face), postmining floor pressure relief-related expansion, and roof collapse-related stress recovery (the distance from the lagging working face is 15–20 m). In the vertical direction, a soft rock layer blocks the continuous transfer of mining stress to deeper layers and produces an important cushioning effect. In the process of coal mining, shear cracks easily develop in the coal wall in front of and behind the working face. After the coal seam is excavated, the length of the fractures that develop in the physical model is 27 cm. The confined water loading system can visually reproduce the hydraulic characteristics of the confined water during the mining process; that is, the confined water easily bursts at the front and back ends of the coal wall in the goaf. The error, as determined by comparison between the field measurement results and the theoretical calculation results, is only 0.617 m, verifying the reliability of the similar simulation method.

In order to study the prevention and control technology for hard roof type coal burst in the isolated working face of the Chenjiashan coal mine, the 418 isolated working face was selected as the engineering case study. Based on the different impact danger zones and mining areas, a roof breaking blasting pressure relief technical scheme was proposed. The anti-impact effect was verified through hole peeping and infrared radiation data. The research shows: (1) According to the geological conditions of the 418 working face of the Chenjiashan coal mine, the working face is divided into weak impact hazard areas and moderate impact hazard areas. Targeted roof blasting schemes were proposed for the initial square area of the working face, moderate danger area, weak danger area, initial mining and initial caving area, and the strike area of the working face. (2) On-site borehole data show that after blasting, a large number of fractures and delaminations were formed in the roof, and some fractures further developed into delaminations, with local areas showing crushed zones. This proves the formation of a “buffer zone” in the roof and floor, achieving pre-cracking of the thick and hard roof, full development of fractures, significant reduction in stress concentration, and the roof blasting can achieve good pressure relief effect. (3) The temperature monitoring near the blasting point and the infrared radiation temperature shows that within an hour after the implementation of the roof blasting, the coal mass at the breaking position experienced a process of heating up and then cooling down, with the temperature at the monitoring point rising by 0.5–0.7 °C. After the implementation of the roof blasting, the key layer above the working face was destroyed, and the stress was released and transmitted to the corresponding area of the coal mass, the stress of the coal increased, and the infrared radiation temperature increased, proving that the blasting pressure relief achieved the expected effect.

Filling mining is one of the feasible methods to prevent rock burst in coal mining, which is a research significance topic in the field. In this paper, we first described the filling mining effect based on “equivalent mining height” and analyzed the evolution of overlying rock strata. It was found that time‐space hysteresis occurred during the movement of overlying rock strata in filled working face. On this basis, the mechanical relationship of dynamic transformation among “filling rate–strata movement–abutment stress” was obtained by analyzing the coal stress characteristics. Then, an advanced abutment stress estimation model of filled working face in deep coal mine was established. Finally, the C5301 working face of Yunhe coal mine was taken as the engineering background, and example calculation and field monitoring were carried out. The results showed that the influence range of the advanced abutment stress is 91–97 m, and the peak value is 41.1 MPa, which is 21–50 m away from the working face. In addition, large pressure steps and long interval time were the characteristics of time‐space hysteresis in the filled working face. This study could provide guidance and reference for deep filling mining under the same or similar conditions.

Coal mining in China has extended to greater depths, making the urgent resolution of associated rock burst issues a critical priority. The retention of coal pillars often leads to the formation of stress concentration zones, increasing the rock burst hazard on lower coal seams and posing challenges to the safe production of coal mines. Using the 8332 working face of Xuzhuang Mine as an example, this paper investigates the factors influencing the rock burst hazard of the working face under the residual coal pillar in the goaf, as well as the evolution of the overburden structure and the stress distribution characteristics during the mining process. Based on the microseismic data collected during mining, the spatio-temporal evolution of microseismic activity in the working face under the residual coal pillar in the goaf is analyzed. The results show that: the ‘F’ and ‘T’ spatial structures are determined as the critical stress concentration zone. A high rock burst hazard in the affected area of the residual coal pillar. During the mining period, microseismic energy and frequency are generally low, exhibiting a trend of first increasing, then decreasing, and finally increasing again. A prevention strategy for rock bursts in the 8332 working face was developed based on its impact characteristics. Its feasibility and effectiveness were verified, providing theoretical support and technical guidance for mining under similar conditions. The findings of this study provide valuable insights for developing effective rock burst prevention strategies in deep coal mines, thereby enhancing mining safety and efficiency.

In order to achieve the accuracy of gas emission prediction for different workplaces in coal mines, three coal mining workings and four intake and return air roadway of working face in Nantun coal mine were selected for the study. A prediction model of gas emission volume based on the grey prediction model GM (1,1) was established. By comparing the predicted and actual values of gas emission rate at different working face locations, the prediction error of the gray prediction model was calculated, and the applicability and accuracy of the gray prediction method in the prediction of gas gushing out from working faces in coal mines were determined. The results show that the maximum error between the predicted and actual measured values of the gray model is 2.41%, and the minimum value is only 0.07%. There is no significant prediction error over a larger time scale; the overall prediction accuracy is high. It achieves the purpose of accurately predicting the amount of gas gushing from the working face within a short period of time. Consequently, the grey prediction model is of great significance in ensuring the safety production of coal mine working face and promote the safety management of coal mine.

Accurately predicting the development height of the water-conducting fracture zone (HW) is imperative for safe mining in coal mines, in addition to the protection of water resources and the environment. At present, there are relatively few fine-scale zoning studies that specifically focus on predicting the HW under high-intensity mining conditions in western China. In view of this, this paper takes the Yushen mining area as an example, studies the relationship between the water-conducting fissure zone and coal seam mining height, coal seam mining depth, hard rock scale factor, and working face slope length, finally proposing a method to determine the development height of the HW based on multiple nonlinear regression models optimized using the entropy weight method (EWM-MNR). To compare the reliability of this model, random forest regression (RFR) and support vector machine regression (SVR) models were constructed for prediction. The findings of this study showed that the results of the EWM-MNR model were in better agreement with the measured values. Finally, the model was used to accurately predict the development height of the hydraulic conductivity fracture zone in the 112201 working face of the Xiaobaodang coal mine. The research results provide a theoretical reference for water damage control and mine ecological protection in the Yushen mine and other similar high-intensity mining areas.

The water inrush of a working face is the main hidden danger to the safe mining of underwater coal seams. It is known that the development of water‐flowing fractured zones in overlying strata is the basic path which causes water inrushes in working faces. In the engineering background of the underwater mining in the Longkou Mining Area, the analysis model and judgment method of crack propagation were created on the basis of the Mohr–Coulomb criterion. Fish language was used to couple the extension model into the FLAC3d software, in order to simulate the mining process of the underwater coal seam, as well as to analyze the initiation evolutionary characteristics and seepage laws of the fractured zones in the overlying strata during the advancing processes of the working face. The results showed that, during the coal seam mining process, the mining fractured zones which had been caused by the compression‐shear and tension‐shear were mainly concentrated in the overlying strata of the working face. Also, the open‐off cut and mining working face were the key sections of the water inrush in the rock mass. The condition of the water disaster was the formation of a water inrush channel. The possible water inrush channels in underwater coal mining are mainly composed of water‐flowing fractured zones which are formed during the excavation processes. The numerical simulation results were validated through the practical engineering of field observations on the height of water‐flowing fractured zone, which displayed a favorable adaptability.

This study addresses the problem of excessive damage to flexible formwork concrete pier columns caused by secondary mining in pillarless coal mining, with a focus on the 1315 working face of Zhaozhuang Coal Mine and the 23,107 working face of Xiegou Coal Mine. Through field research, numerical simulation, theoretical analysis, big data machine learning, and field testing, the stress migration patterns and destabilization mechanisms of flexible formwork concrete pier columns under secondary mining conditions were investigated. The findings revealed that stress concentration in the columns during mining could lead to excessive damage, compromising safety. A Gaussian process regression (GPR)-based stress prediction model was developed (optimal kernel: ARD-Rational-Quadratic-Kernel, with MSE = 1.3463, RMSE = 1.1603, MAE = 0.6138, and MAPE = 0.4041), demonstrating significantly higher accuracy than linear regression models (error reduced by 1–2 orders of magnitude) and BP neural networks (MSE = 2.0962). The model further indicated that the damage extent of the columns followed a two-stage pattern with increasing distance from the mining face: initial near-linear growth, followed by a stabilized rate of increase. Field tests confirmed that reinforcing the flexible pier columns with Z6 concrete reinforcing agent ensured safe mining operations, validating the practical applicability of the prediction model.

The impact risk evaluation for the strip filling of working faces has always been a research hotspot and a difficulty in the field of rock bursts. In this paper, the concept of the critical filling rate is first proposed, and the criterion for identifying the impact risk of the strip filling of a working face is established. Then, the membership function of coal body stress and the coal seam elastic energy index to impact risk was established, and the classification index of the impact risk grade was formed. On this basis, the overall and local evaluation method of the rock burst hazard for the strip filling of working faces was proposed. Finally, the C8301 working face of the Yunhe coal mine was taken as the engineering background, and the impact risk evaluation was carried out. It was found that the overall risk of the C8301 working face was determined as a strong impact risk, and there were six local risk areas, which included two weak impact risk areas, two medium impact risk areas, and two strong impact risk areas. This study can provide guidance and a reference for the impact risk evaluation of strip-filling mining under the same or similar conditions.

This paper studies the width of narrow coal pillars, mining‐induced failure characteristics, and surrounding rock control effect of gob‐side entry driving (GED) adjacent to 2‐1208 filling working face with an approximately 900 m depth. Laboratory experiments, numerical simulations, loosening circle tests, and engineering practices are conducted. The mechanical properties of the filling body, the distribution and evolution law of the second invariant deviatoric stress (J2), and the variation in the plastic zone of the surrounding rock in GED are studied. The conditions of various coal pillar widths and the gob backfilled or not of the adjacent working face are also considered. The results of roadway driving along the filling working face in the deep coal mine are that: (1) The evolution law of J2 and the plastic zone features of the surrounding rock in GED have changed significantly. Thus, it is unreasonable to still adopt the customary theory of narrow coal pillar determination and roadway support design method of GED adjacent to nonfilling working face. (2) The plastic zone of GED has typically asymmetric distribution characteristics during driving and retreating, which is mainly concentrated at the upper corner of the virgin coal rib. And the elastic zone changes slightly by the influence of mining‐induced pressure on the working face, indicating that support in this area is easier. (3) During the stable period of roadway driving, there includes a “ring‐shaped” J2 depression zone at the upper corner of the coal pillar rib. While the peak zone distribution of J2 in the virgin coal rib of GED is approximately “crescent‐shaped,” and the peak zone is inclined to the upper corner of the virgin coal rib, which implies that this area is a crucial control region. (4) During the retreating period of panel 2‐1210, the direction of the maximum principal stress of GED gradually deflects from the direction of the gob to that perpendicular to the roof and floor, which represents that the key support area of GED has changed. Therefore, the advanced support of GED needs to be reinforced by single hydraulic props. The width of the coal pillar was determined to be 5 m, and a targeted truss anchor cable support method with bidirectional resistance function is proposed for the roof and both ribs, which realized the stability control of GED during deep filling mining.