Multi-method investigation of overlying strata stability and hazard mitigation in ultra-shallow-buried isolated mining faces with burned-out zones

Abstract

This study addressed complex engineering challenges of the 30,101 isolated island working face in Hejiata Coal Mine, Shaanxi Province, characterized by three interrelated geological-mining conditions: ultra-shallow burial (78 m), well-developed surface water (dominated by the Wulanmulun River), and porous thermally altered rocks from adjacent coal spontaneous combustion. Contrary to conventional assumptions, the 78 m shallow 3-1 coal seam had high burst risk, confirmed by uniaxial compression tests showing high elastic modulus and abrupt post-peak brittle failure. To evaluate its stability and hazards, a multi-method approach was used: FLAC3D for stress/displacement/plastic zone simulation, UDEC for discontinuous fracture analysis, and field bolt load monitoring for real-time data. Results showed overlying strata formed a typical “S”-shaped structure during mining, with stress concentration factor rising from 1.63 (20 m advance) to a stable ∼3.02 after 180 m. UDEC simulations revealed water-conducting fractures connected surface water, burned-out zone pores and the mine, posing water inrush risk. The 15 m coal pillar retained a ∼3 m elastic core at 220 m advance. Field monitoring showed 30,101 CBE had 40%–50% higher roof displacement and peak bolt loads (192 kN) than RAE, with greater stress on the coal pillar side. Targeted measures (optimized CBE support, comprehensive water prevention-drainage) were proposed. This study provides technical guidance for safe mining of similar ultra-shallow, high-risk isolated island working faces.

1 Introduction

North China coalfield is a core energy base for China, hosting abundant high-quality coal reserves (Zhang L. J. et al., 2022). Most coal seams here are characterized by shallow burial and large thickness (Zhu et al., 2022), facilitating high-efficiency mining but also posing prominent challenges (Xie et al., 2019), such as intense mine pressure induced by shallow overburden, frequent water inrush risks from surface water infiltration (Zhu et al., 2022; Chen et al., 2019; Chen et al., 2025), and rockburst threats in isolated island layouts (Dai et al., 2024; Wang H. W. et al., 2024; Wang Z. S. et al., 2024). Additionally, balancing mining productivity with surface subsidence control and groundwater protection has become a key constraint for sustainable development, as conventional mining methods often struggle to address these interconnected geological and environmental issues. Furthermore, the long-term environmental impact of shallow-buried mining extends beyond the immediate extraction period. As the overlying strata collapse and the goaf compacts over time, surface deformation often evolves into distinct geological hazards, most notably ground fissures and sinkholes. These surface collapses not only damage local ecosystems and infrastructure but also create direct conduits for surface water to infiltrate the mine, thereby creating a cyclical hazard of water inrush and further strata instability.

Existing research on overlying strata caving of shallow-buried working faces (burial depth <100 m) has highlighted that the single-key stratum in such conditions tends to fracture prematurely due to limited overburden thickness, leading to distinct “caving-filling” cycles different from deep-buried faces (Wu et al., 2024; Qi et al., 2025; Wen et al., 2026). Moreover, near-surface strata can generate porous, thermally altered rocks because of coal spontaneous combustion (Yang et al., 2025; Hallman, 2022; Yu and Liu, 2024), then the mechanical strength is significantly reduced. These porous rocks not only exacerbate overlying strata instability but also act as preferential channels for groundwater migration, creating dual threats of water inrush and roadway collapse.

From another perspective, rockburst risk has been assumed to increase proportionally with burial depth (Pytel et al., 2026), as deeper strata accumulate higher in-situ stress (Dubinski et al., 2023). However, recent studies have challenged this simplification (Du et al., 2024; Bo et al., 2023; Rao et al., 2024), revealing that shallow-buried coal seams can exhibit high burst tendency under specific conditions (Xie et al., 2025): intense abutment pressure from adjacent gobs in isolated island layouts, high coal brittleness (indicated by steep stress-strain curves and abrupt post-peak failure), and rapid release of elastic energy. In ultra-shallow strata (<80 m), the limited overburden even fails to “dissipate” stress effectively, leading to concentrated stress near coal pillars or working faces—making shallow-buried faces equally prone to rockbursts, contrary to conventional perceptions. Furthermore, water-sand inrush disasters in shallow-buried coal mine are more likely to occur (Liu et al., 2025).

Existing research on shallow-buried coal seams focuses on single factors (e.g., water inrush (Zhu et al., 2022) or rockbursts (Dai et al., 2024)), but few address the coupling of four critical factors: ultra-shallow burial (78 m), thermally altered rock zones, surface water systems, and isolated island layout. Take the thermally altered rocks as an example, Yang et al. (2025) studied spontaneous combustion but not mining-induced fracture propagation; Take the isolated working faces as an example, then Du et al. (2024) analyzed coal pillar stability but ignored water inrush risks (Feng et al., 2024). If continuing this topic from the perspective of mining simulations, it is important to note that longwall mining simulations require rigorous model calibration and accurate representation of gob behavior (Esterhuizen et al., 2018; Zhang et al., 2019; Tuncay et al., 2021; Zhang et al., 2022a), and a standardized framework for gob material property assignment (e.g., residual strength degradation with compaction) is required (Zhang et al., 2018). For example, Zhang et al. highlighted that stress redistribution in shallow longwall faces was dominated by abutment pressure superposition from adjacent gobs, a phenomenon validated by NIOSH’s longwall monitoring data (Zhang et al., 2022b). Nevertheless, thermally altered rock zones are rarely incorporated in existing longwall mining models.

To fill these research gaps, this study targets the 30,101 isolated island working face in Hejiata Coal Mine, which integrates four critical challenges: ultra-shallow burial (78 m), well-developed surface water, porous burned-out rocks, and confirmed high burst tendency. A combination of FLAC3D (for stress/displacement evolution), UDEC (for discontinuous fracture analysis), and field monitoring (bolt load, displacement) is employed to investigate overlying strata fracture patterns, coal pillar stability, and water inrush risk. Its novelty lies in coupling these four factors—rarely studied collectively—to provide scientific support for hazard control, offering a reference for similar shallow-buried, high-risk working faces.

2 Engineering background and challenges

In this section, the engineering background of the 30,101 working face (Section 2.1), key mining challenges (e.g., mine pressure, spontaneous combustion, water inrush; Section 2.2), and the coal seam’s burst tendency evaluation (Section 2.3) will be outline. Below is the detailed engineering background.

2.1 Engineering background of Hejiata Coal Mine

In Figure 1, the location and the surface water systems of the engineering site were presented. Hejiata Coal Mine is situated approximately 47 km north of the urban area of Shenmu City, Shaanxi Province (Figures 1a,b). The production capacity of the upper coal seam mine in Hejiata Coal Mine is 0.3 million tons per annum (Mt/a). It adopts the inclined shaft development method. This study focused on the mining activity of the 3-1 coal seam, and the longwall fully mechanized coal mining technology is employed. The working face uses the retreating mining method, and the full caving method is used for roof management. The surface water system in the mining area is well-developed, with the Wulanmulun River being the largest watercourse traversing the east and west flanks of the Hejiata mine field (Figure 1c).

FIGURE 1

The thickness of the No. 3-1 coal seam ranges from 2.20 to 3.60 m, with an average of 2.95 m. The false roof of the coal seam is carbonaceous mudstone with a thickness of 0.3 m; the immediate roof is siltstone with a thickness of 6.3 m, and the main roof is medium-grained sandstone with a thickness of 8.22 m. The immediate floor is fine-grained sandstone with a thickness of 10.75 m, and the main floor is medium-grained sandstone with a thickness of 8.22 m.

The elevation of the working face 30,101 ranges from +1,118.0 to +1,142.5 m, while the ground elevation ranges from +1,186.3 to +1,232.4 m. The advancing length and the length of the working face is 1,277 and 175 m, respectively. The total area of the working face is approximately 223,475 square meters, the designed recoverable reserves amount to 750,900 tons. The service life of the working face is 30 months. In the area where the working face 30,101 is located, the thickness of the coal seam ranges from 2.5 to 2.9 m, with an average of 2.7 m. The coal seam is mainly composed of bright coal and semi-bright coal, and the average burial depth of the coal seam is 78 m.

Figure 2 presents the overall layout of the working face 30,101. On the eastern side of the 30,101 working face lies a protective coal pillar within the burned-out zone; to the west are the main district roadways; to the south is the goaf of the 30,102 working face; and to the north is the goaf of Dahaeze Coal Mine, separated by a 40-m-wide protective coal pillar at the mine boundary. The 30,101 working face has now been fully connected, forming an isolated island working face with two free faces (Feng et al., 2024). The overall layout is illustrated in Figure 1.

FIGURE 2

Notably, the 30,101 working face’s 78 m shallow burial depth, coupled with the adjacent burned-out zone and well-developed surface water systems (Figure 1c), exacerbates water hazard risks. Shallow burial in 30,101 working face facilitates the penetration of mining-induced fractures to the surface, while the burned-out zone shown in Figure 2 is characterized by fragmented, porous thermally altered rocks. Therefore, preferential pathways for water migration can be formed. Moreover, atmospheric precipitation and runoff from the Wulanmulun River may infiltrate through these interconnected fractures and the burned-out zone, potentially seeping into the working face or gob areas, posing a persistent threat to safe mining.

2.2 Challenges of the work face 30,101

The lithological firmness of the coal seam roof and floor in the working face 30,101 is predominantly characterized by medium-strength rocks. The coal seam floor mainly consists of siltstone and mudstone, which serve as a relatively water - resistant layer.

The main roof of the No. 3-1 coal seam primarily consists of medium-to fine-grained sandstone, occurring in thick to extremely thick layers ranging from 2.0 to 16.90 m in thickness. In shallow-buried mining conditions, this stratum exhibits the structural characteristics of a single key stratum, which tends to induce intense mine pressure phenomena and dynamic loading effects. Isolated island working faces are typically characterized by a high degree of roof fragmentation, shortened periodic weighting intervals, increased weighting intensity, significantly elevated abutment pressure ahead of the coal wall, and heightened susceptibility to dynamic impacts, thereby posing a considerable rockburst risk.

Furthermore, the No. 3-1 coal seam is shallowly buried and contains a high proportion of fusain, making it highly prone to spontaneous combustion. Spontaneous combustion has occurred extensively in the valley areas of the minefield (see Figure 2), resulting in widespread burning of both the roof and floor strata. These burned zones have largely transformed into sintered rocks, while those near the combustion boundary are predominantly molten-burned rocks. The maximum lateral extent of the burned-out zone reaches nearly 2 km. Shallow burial renders coal seams particularly vulnerable to oxidation and weathering-induced spontaneous combustion. Under prolonged high-temperature exposure and melting conditions, the surrounding rock undergoes significant alterations in structure and composition, forming thermally altered (thermally altered) rocks. As a distinct lithological type, thermally altered rocks exhibit substantially reduced mechanical strength due to the development of joints, fissures, pore networks, and a generally loose texture, presenting major challenges for roadway stability control in burned-out areas.

Moreover, the spontaneous combustion of the coal seam has caused varying degrees of damage to the overlying strata. Additional voids, solution cavities, and fractures have developed upon pre-existing diagenetic and tectonic fissures, creating favorable pathways for groundwater occurrence and migration. Within the burned-out zone, the water-conducting fracture zone generated during the extraction of the No. 3-1 coal seam is highly likely to extend into underlying or adjacent aquifers. This increases the risk of inflow of water from thermally altered rocks into the mine workings, making it a direct source of mine water inrush and posing a significant threat to safe mining operations.

2.3 Burst tendency evaluation of coal seam

In order to evaluate the burst tendency of coal seam, mechanical loading tests were conducted. Coal samples with a diameter of 50 mm and a height of 100 mm were prepared, and uniaxial compression tests were conducted to evaluate their coal burst tendency. The test results are presented in Figure 3, where Figure 3a shows the prepared cylindrical coal specimens (50 mm diameter × 100 mm height).

FIGURE 3

The stress-strain curves in Figures 3b–f exhibit distinct elastic, peak, and post-peak stages. The elastic stage is characterized by a relatively high slope, which is an indicator of substantial elastic modulus. A higher elastic modulus signifies that the coal can store large amounts of elastic energy during loading, and which is an essential condition for coal burst. In the post-peak stage, all curves display abrupt stress drops (brittle failure). For instance, Curve (b) shows a near-vertical stress decline after the peak; Curves (c), (d), and (f) exhibit stepwise stress drops; and Curve (e) presents fluctuating post-peak behavior. Such brittle failure patterns imply that the elastic energy accumulated during the elastic stage is released rapidly and intensely at failure, and this is an intrinsic feature of coal burst phenomena. Furthermore, the lack of significant residual strength in the post-peak region further confirms the coal’s brittle nature (Chen et al., 2024). Therefore, dynamic and violent failure potential should be a prevalent phenomenon in the Hejiata Coal Mine.

In Figure 4, the curves in (a)–(d) reflect the elastic energy storage and dissipation behaviors of four coal specimens. The peak stress of 10 MPa was selected based on the UCS values in Figure 3 (approx. 62% of the 16 MPa). Almost all curves exhibit steep elastic segments, which is an essential prerequisite for coal bursts. Curves (a) and (b) have wider loops, reflecting more plastic energy dissipation; in contrast, curves (c) and (d) have narrow loops, meaning most input energy is stored elastically. The morphology of stress-strain loops and the slope of elastic segments indicate differences in elastic energy accumulation and plastic energy dissipation among the samples. Higher elastic modulus and narrower hysteresis loops suggest stronger elastic energy storage capacity, which is closely related to the Elastic Energy Index (W_ET) and coal burst tendency (Das et al., 2023).

FIGURE 4

To quantitatively assess the risk, W_ET was calculated. W_ET is defined as the ratio of the elastic strain energy accumulated to the plastic strain energy dissipated during a loading-unloading cycle, as expressed in Equation 1, and samples with superior elastic energy storage potential exhibit higher W_ET, such as the case in Figure 4d. Figure 4d shows the strongest elastic energy storage potential (steepest slope + narrowest loop), corresponding to the highest W_ET value, and higher W_ET correlates with elevated burst risk.Where: W_ET is the elastic energy index; E_e is the elastic strain energy stored (represented by the area under the unloading curve); E_p is the plastic strain energy dissipated (represented by the area of the hysteresis loop).

Based on the calculated values, the coal burst tendency is classified into three categories as shown in Table 1. The average W_ET of the 3-1 coal seam samples was found to be 6.4, which falls into the “Strong burst tendency” category. Overall, the test here again emphasizes the great coal burst risk in Hejiata Coal Mine. For mitigation, drilling boreholes (Tai et al., 2024), would be effective countermeasures.

2.4 Borehole logging data analysis

A total of six boreholes were drilled in the roof along the 30,101 return air entry (with dimensions of 5.2 m in width and 3.0 m in height), and a borehole imaging instrument was employed to assess the integrity of the borehole walls. The distances between these boreholes and the 30,101 working face were 30, 60, 90, 120, 150, and 180 m, respectively. The procedure involved drilling the boreholes (diameter of 40 mm), inserting the camera probe, controlling the advancement speed of the probe (slowing down at fractured zones) and processing the data to generate unwrapped borehole images. The processed images are presented in Figure 5, showing intervals of 0–1.0, 2.0–3.0, and 4.0–5.0 m. The observation depth was limited to 5.0 m because the surrounding rock beyond this depth remained intact, showing no visible fractures or significant deformation.

FIGURE 5

As shown in Figure 5, borehole #1 exhibits severe fracturing from 0 to 3.0 m, while borehole wall integrity is maintained beyond 4.0 m, despite the presence of a visible longitudinal crack. In borehole #2, laminated strata are clearly observed between 2.0–2.5 and 4.0–5.0 m, although no significant fractures are apparent. Boreholes #3, #4, and #6 generally maintain good structural integrity. Borehole #5 contains a brown interlayer between 2.5 and 3.0 m, which may be prone to delamination under mining-induced stress. Based on the spatial distribution relative to the working face, it can be inferred that borehole #1 is located within a zone significantly influenced by mining-induced pressure. Therefore, enhanced support measures should be implemented in the vicinity of this borehole. The borehole logging results provide critical input for numerical modeling, enabling estimation of the extent of the stress-affected zone and facilitating calibration of subsequent simulation results.

3 Numerical model setting

FLAC3D (Fast Lagrangian Analysis of Continua in 3 Dimensions) is employed to analyze the evolution laws of the stress field (Unver and Yasitli, 2006), displacement field, and plastic zone during the mining process of this working face, and to assess the stability of the sectional coal pillar. For the numerical model, the average burial depth of this working face is 78 m. On the basis of considering multiple factors such as computational efficiency, full mining of surrounding rock, and rockburst proneness of the working face, in this numerical calculation, the length of the 30,102 working face is set to 180 m, and the advancing distance of the working face is 322 m; the length of the 30,101 working face is 175 m, and the advancing distance of the working face is 300 m. The numerical model is generated by the Generate command in FLAC3D. The model size is width X × thickness Y × height Z = 387.7 m × 340.6 m × 138 m, and the advancing direction of the working face is along the Y - axis. As can be seen in Figure 6. Longwall mining is simulated with a stepwise advance of 5 m per iteration (consistent with field mining practice). Gob areas were simulated using FLAC3D’s FISH function to replicate strain-hardening behavior based on Tulu et al.’s gob compaction model (Tuncay et al., 2021). The mechanical behavior of the goaf was simulated using a strain-hardening model implemented via FLAC3D’s FISH language. The vertical stress (σ_v) in the goaf is related to the vertical strain (ε_v) by the hyperbolic equation (Equation 2).where a represents the stiffness coefficient of the caved rock mass, and b represents the maximum potential volumetric strain. Based on the lithology of the caved zone (predominantly medium-grained sandstone), the parameters listed in Table 2 were adopted to simulate the compaction and hardening characteristics of the goaf.

FIGURE 6

The coal seam is excavated sequentially, with the 30,102 working face mined first (advance distance 322 m) to simulate residual stress, followed by the 30,101 face (advance distance 300 m).

The protective coal pillar between the conveyor belt entry of the 30,102 working face (30,102 CBE) and that of the 30,101 working face is 15 m, this value was determined based on the successful mining experience of adjacent panels at Hejiata Mine. The width of the coal pillar along the return airway of the 30,101 (30,101 RAE) working face at the mine boundary is 40 m. Meanwhile, to avoid the influence of boundary effects on the numerical calculation results as much as possible, the size of the boundary coal pillar is set to 60 m.

The Mohr-Coulomb constitutive model is used to simulate the strength of coal and rock strata, and the brick element is used to simulate the coal and rock strata. The displacements of the front, back, left, right, and lower surfaces of the numerical model are fixed. To balance accuracy and efficiency, the model uses variable zone sizes tailored to each geological layer: Loess (0.5 m × 0.5 m × 0.5 m), Clay (0.8 m × 0.8 m × 0.8 m), Sandy mudstone (1.0 m × 1.0 m × 1.0 m), Medium-grained sandstone (1.2 m × 1.2 m × 1.2 m), Fine-grained sandstone (1.0 m × 1.0 m × 1.0 m), 3-1 Coal (0.6 m × 0.6 m × 0.6 m), Mudstone (0.9 m × 0.9 m × 0.9 m), Siltstone (0.8 m × 0.8 m × 0.8 m). The total number of elements is 2,580,970, with finer discretization for the coal seam and thermally altered rock zones to capture stress concentration. The mechanical parameters of all strata are listed in Table 3. All parameters are derived from lab tests on core samples collected from Hejiata Mine (three replicates per stratum): 1) Unit weight: Measured via volumetric method; 2) Elastic modulus/compressive strength: Uniaxial compression tests; (3) Tensile strength: Brazilian splitting tests; (4) Internal friction angle/Poisson’s ratio: Direct shear tests. The siltstone internal friction angle (8.3°) reflects weathering-induced weakening in the near-surface zone, confirmed by field observations.

Interfaces between adjacent strata are modeled using FLAC3D’s built-in contact logic with shear strength parameters derived from direct shear tests, such as friction angle and cohesion. Then a realistic shear-slip behavior at interfaces can be achieved, which is critical for simulating strata deformation and failure propagation. The boundary conditions are defined as follows: (1) Vertical boundaries (X- and Y-directions): horizontal displacement is fixed; (2) Bottom boundary (Z-direction): vertical displacement is constrained; (3) Top boundary: an overburden pressure of γH = 17.86 kN/m³ × 78 m = 1.4 MPa is applied to simulate the burial depth. The horizontal in-situ stresses are applied according to field measurements from Hejiata Mine, with a maximum horizontal stress (σ_hmax) of 4.2 MPa and a minimum horizontal stress (σ_hmin) of 2.8 MPa. The direction of σ_hmax is oriented at 30° relative to the working face advance direction (Y-axis).

To evaluate the reliability of the numerical model and the influence of key parameters on simulation results, a sensitivity analysis was conducted for three critical mechanical parameters (derived from Table 3) that dominate strata stress and displacement responses: 1) elastic modulus of the 3-1 coal (baseline: 25 GPa); 2) internal friction angle of the main roof (medium-grained sandstone, baseline: 29°); and 3) cohesion of thermally altered siltstone (derived from Mohr-Coulomb criterion using Table 3 values, baseline: 1.8 MPa). Each parameter was independently varied by ±10%, and the stress concentration factor (CF) at the 30,101 working face and maximum roof vertical displacement were monitored.

Results showed that the internal friction angle of the main roof was the most sensitive parameter: a ±10% change (26.1°–31.9°) led to a ±6.1% variation in CF (from 2.84 to 3.20) and ±5.3% variation in roof displacement (from 232 to 258 mm). The elastic modulus of coal was moderately sensitive: ±10% change (22.5–27.5 GPa) caused ±4.2% CF variation (2.89–3.15) and ±3.8% displacement variation (236–254 mm). The cohesion of thermally altered siltstone was least sensitive: ±10% change (1.62–1.98 MPa) resulted in only ±2.9% CF variation (2.94–3.10) and ±2.5% displacement variation (240–250 mm). These results ensure that the model’s core outputs will be stable within the reasonable parameter variation range, and the main roof’s friction angle should be prioritized for shallow-buried strata movement.

In accordance with the mining regulations of the working faces, the 30,102 working face and the 30,101 working face shall be mined sequentially. Subsequently, the numerical calculation model was saved, and post - processing was carried out.

4 Numerical results of stability of working face 30,101

4.1 Stress and displacement of surrounding rock after mining at the 30,102 working face

The 30,101 working face is adjacent to the goaf of the previously mined 30,102 working face. The 30,101 conveyor belt entry (30,101 CBE) is driven along the boundary of the 30,102 goaf, with a 15-m coal pillar left intact for support. The 30,101 working face is subjected to high abutment pressure from both the adjacent goaf and the isolated pillar, resulting in significant stress concentration. Following the extraction of the 30,102 working face, the vertical stress distribution and deformation characteristics of the surrounding rock have become critical external factors influencing the excavation and safe operation of the 30,101 working face. This section presents a detailed analysis of these geomechanical conditions.

The vertical stress distribution during the advancement of the 30,102 working face is illustrated in Figures 7a,b. During the mining process, the overlying strata load is transferred to the adjacent coal ribs surrounding the goaf. In the inclination direction of the working face (or along vertical “width X” section in Figure 6b), the overburden load is transmitted to the protective coal pillar, where the maximum support pressure reaches 6.06 MPa. Along the strike direction of the working face (or along horizontal section), a stress concentration zone develops behind the goaf, with a peak abutment pressure of 6.32 MPa occurring on the coal rib.

FIGURE 7

The displacement contour maps of the 30,102 working face during advancement are presented in Figures 7c,d. These figures indicate that as the working face advances, the roof strata above the goaf lose structural support, leading to significant deformation zones in both the roof and floor. Specifically, the floor of the goaf exhibits upward displacement, while the roof undergoes downward movement, resulting in roof-to-floor convergence in the goaf region.

The distribution of the plastic zone during the advancement of the 30,102 working face is illustrated in Figure 8. As the mining distance increases, distinct variations in the plastic zones of the surrounding rock’s roof and floor are observed. Along the inclination direction of the working face, the roof strata exhibit a composite plastic failure mode predominantly characterized by tensile failure. The height of plastic failure in the roof adjacent to the coal ribs on both sides of the working face exceeds that at the center of the goaf (Jayanthu et al., 2004).

FIGURE 8

4.2 Stress evolution characteristics of working face 30,101

As the advancing distance of the 30,101 working face increases, the deformation extent and vertical displacement of the roof and floor strata progressively increase. Throughout different stages of mining, significant changes occur in both the peak magnitude and location of vertical displacement within the overlying strata above the 30,101 working face. The variation pattern of roof strata vertical displacement along the strike direction during the advancement of the working face is illustrated in Figure 9.

FIGURE 9

4.2.1 Evolution of vertical displacement

When the working face advances to 20 m, under the influence of overburden pressure in the goaf, the roof strata begin to exhibit measurable deformation. The maximum vertical displacement in the goaf reaches 0.26 m, located 7.8 m from the open-off cut. At an advancement of 80 m, with the intensification of mining-induced disturbances and the increased exposure length of the roof strata, both the deformation zone and vertical displacement continue to expand. The maximum vertical displacement increases to 3.04 m, occurring at a distance of 21.5 m from the open-off cut. Upon reaching 140 m, further growth in vertical displacement of the roof strata above the goaf is observed. The maximum displacement rises to 3.06 m, now situated 70.0 m from the open-off cut. At 200 m of advancement, as the working face continues to progress, the maximum vertical displacement of the roof strata above the goaf reaches 3.09 m, with its location shifting to 107.8 m from the open-off cut. Finally, when the working face advances to 220 m, the vertical displacement of the roof strata slightly increases, achieving a peak value of 3.10 m at a position 110.0 m from the open-off cut.

4.2.2 Evolution of plastic zone

Figure 9 also illustrates the distribution characteristics of the plastic zone in the surrounding rock under varying advancing distances of the 30,101 working face. As shown in the figure, with the continuous advancement of the 30,101 working face, the extent of plastic deformation and damage in the surrounding rock progressively intensifies. The plastic damage zone in the roof strata above the goaf expands upward, while the depth of plastic damage in the floor increases correspondingly.

At an advancing distance of 20 m, due to the relatively limited mining length, the maximum damage height reaches 10.0 m. When the working face advances to 80 m, the plastic damage range in the roof expands significantly, resulting in a damage height of 28.0 m. Upon reaching 140 m, extensive plastic damage occurs in the surrounding rock, with further expansion of the damage height in both the coal wall roof and the area behind the goaf. Shear failure is predominantly observed along the sides of the goaf roof, whereas tensile failure is mainly concentrated at the central region. The maximum plastic damage height of the goaf roof reaches 28.5 m at this stage.

When the working face advances to 180 m, the plastic damage heights in both the roof and floor gradually stabilize. The roof exhibits a combined failure mode involving both tensile and shear damage, while the floor is primarily affected by shear damage. The maximum plastic damage height in the roof is observed at the center of the goaf, which remains consistent throughout the subsequent mining stages.

At an advancing distance of 220 m, as the mining length reaches its maximum, the height and width of the plastic zones in the surrounding rock of both the roof and floor attain their peak values, with the maximum roof damage height reaching 29.0 m.

In summary, during the advancement of the working face, the damage to the overlying strata in the goaf develops progressively from bottom to top. In terms of damage types, tensile damage is primarily distributed in the lower-level roof at the center of the goaf, whereas shear damage is more prevalent in the upper-level roof strata.

4.2.3 Evolution of abutment pressure

During the advancement of the 30,101 working face, a monitoring line was established along the strike direction at the mid-section of the numerical model on the upper surface of the coal seam to track the evolution of abutment pressure (He et al., 2022). As illustrated in Figure 10 and summarized in Table 4.

FIGURE 10

As can be seen in Figure 10, the distribution characteristics of abutment pressure exhibit notable variations in the peak abutment pressure, the location of the peak, and the stress concentration coefficient. When the working face had advanced 20 m, the peak abutment pressure and the stress concentration coefficient were recorded as 3.31 MPa and 1.63, respectively. With increasing mining distance, both the peak abutment pressure and the stress concentration coefficient show a continuous upward trend, although the rate of increase gradually diminishes. At an advancing distance of 220 m, the rate of increase in the stress concentration coefficient reaches its minimum, indicating a trend toward stabilization of stress redistribution. At this stage, the abutment pressure and the stress concentration coefficient stabilize at 6.13 MPa and 3.02, respectively.

Table 4 shows that the stress concentration factor (CF) rises from 1.63 at 20 m advance to a stable value of ∼3.02 at 180–220 m, with the growth rate decelerating sharply after 80 m as strata stress gradually rebalances. This stabilized CF reflects the cumulative effect of active mining disturbance and adjacent gob-induced stress superposition.

4.3 Interpretation of roof collapse patterns

During coal extraction in the working face, mine pressure behavior is primarily attributed to energy release induced by the movement of the coal seam roof (Feng et al., 2019). As the working face advances, lower-level strata cave in and gradually fill the gob. Once the gob is completely filled by strata within the caved zone, higher-level strata lack sufficient space for immediate fracturing and subsidence. Consequently, these higher-level strata undergo gradual bed separation and fracturing over an extended period when the load they bear exceeds their own strength.

Based on the variation in mining layouts, the spatial structural movement of the overlying strata can be categorized into two typical patterns: the “S”-shaped pattern and the “C”-shaped pattern (as illustrated in Figure 11).

FIGURE 11

4.3.1 Dynamic stress of the “S”-shaped overlying strata spatial structure

For working faces with one side mined out and the other side remaining as intact coal (Case 1 in Figure 11), an “S”-shaped overlying strata spatial structure is formed during extraction. As the working face advances continuously, similar to the periodic fracturing of the main roof, high-level rock beams also fracture periodically once they reach a certain span. Correspondingly, the “S”-shaped overlying strata spatial structure expands progressively with the advance of the working face. The dynamic pressure generated by the fracturing of high-level strata is a key factor inducing rockbursts. Notably, in the static “S”-shaped overlying strata spatial structure, the location of the high-stress peak (i.e., the gob-side roadway and the upper part of the working face) is particularly vulnerable—dynamic pressure in this area can generate stress several times higher than static stress, which is one of the critical factors leading to rockbursts.

4.3.2 Dynamic stress of the “C”-shaped overlying strata spatial structure

For isolated island working faces with both sides mined out (Case 2 in Figure 11), a “C”-shaped overlying strata spatial structure is developed during mining.

Affected by this “C”-shaped structure, the gob of the current working face interacts synergistically with the gobs of adjacent working faces. This synergy increases both the fracture height of the overlying strata and the influence range of the overlying strata spatial structure. Under the disturbance of mining activities, the overlying strata on the working face and its two sides undergo large-scale re-fracturing. The dynamic pressure caused by the large-scale movement of the overlying strata spatial structure inflicts severe damage on the roadways on both sides of the working face, with a wide range of influence. The large-scale movement of the “C”-shaped overlying strata spatial structure is the primary reason for the increased advance abutment pressure in the isolated fully mechanized caving face and the intensified mine pressure behavior. Importantly, this pattern of overlying strata movement causes far more severe damage to working face roadways than other overlying strata movement patterns.

4.3.3 Roof collapse pattern of the 30,101 working face

In the case of the 30,101 working face, one side is separated from the gob of the 30,102 working face by a 15 m-wide coal pillar, while the other side is bounded by a 40 m-wide coal pillar (adjacent to the intact coal mass, see Figure 6b). From a mining layout perspective, this means one side (30,101 CBE) of the 30,101 working face is adjacent to a gob (30,102 working face side), and the other side (30,101 RAE) can be considered as intact coal. After the working face advances a certain distance, a typical “S”-shaped overlying strata spatial structure is formed in this area. This structure continues to move as the working face advances further.

The formed “S”-shaped overlying strata spatial structure exerts dynamic pressure on the 30,101 working face and its conveyor belt entry. For practical engineering application, the dynamic stress generated by this structure is superimposed with the static stress. The superimposed stress is approximately 3.0 times the static stress.

The simulation results explicitly link the face advance distance to the roof collapse and stress evolution (Figure 10 and Table 4). The roof behavior proceeds through three distinct stages: 1) Initial Stage (0–20 m): At 20 m advance, the immediate roof has not yet fully caved, and the strata have not formed a stable arch. Consequently, the stress concentration factor (CF) remains low at 1.63. 2) Caving and Bending Stage (20–80 m): As the face advances to 80 m, the immediate roof collapses, and the main roof begins to bend significantly. This movement triggers the superposition of abutment pressure from the adjacent 30,102 gob and the active 30,101 face, causing the CF to rise sharply to 2.87. 3) Stabilization Stage (80–180 m): By 180 m advance, the “S”-shaped overlying structure (shown in Figure 11) fully forms and stabilizes. At this distance, the main roof completes its fracture and rotation, forming a load-bearing arch. The CF plateaus at approximately 3.02, indicating that the stress redistribution between the coal pillar, gob, and overlying strata has reached equilibrium. This aligns with Zhu et al. (2022) observation that main roof fracture in shallow-buried thin bedrock drives the redistribution and stabilization of abutment pressure.

4.4 Prediction of the weighting distance

4.4.1 Prediction of the first weighting

The first weighting manifestation in the isolated island working face is significantly influenced by the surrounding goaf conditions on both sides (Zhen et al., 2023), resulting in intensified stress redistribution within the overlying strata and a pronounced superposition of lateral abutment pressures (Le and Bui, 2020). This leads to greater challenges in surrounding rock control and more intense ground pressure behavior. In this study, the working face is located at a burial depth of 78 m with a mining height of 2.7 m, classifying it as a shallow-buried isolated island panel. In ultra-shallow isolated mining, this first weighting poses heightened risks due to reduced overburden confinement, which leads to amplified dynamic loading and limited warning time. Consequently, accurately identifying this phase and its transition to periodic weighting is critical for timely hydraulic support adjustment and effective hazard management. Structurally, the immediate roof consists of 6.3 m of siltstone, while the main roof is composed of 8.22 m of dense, strong, and highly intact medium-grained sandstone, providing favorable geological conditions for the formation of a large-scale suspended roof structure. The floor is primarily made up of fine sandstone and siltstone, exhibiting moderate mechanical strength and adequate load-bearing capacity, which facilitates periodic weighting and contributes to a distinct initial pressure event.

Historical data indicate that the step distance of the first weighting in previously mined panels was approximately 45 m, suggesting that under similar geological conditions, the main roof tends to fracture relatively early, resulting in a short pressure step distance. Numerical simulation results show the roof and floor displacements at 40 and 60 m of face advance, as illustrated in Figure 12. When the working face advances to 40 m, significant subsidence of the roof and floor is observed, indicating complete caving of the immediate roof and the onset of load transfer to the main roof, which begins to undergo elastic bending deformation—marking the preliminary stage preceding strata pressure. Upon advancing to 60 m, direct contact between the roof and floor occurs, with vertical stress in the abutment zone reaching 1.2 MPa. This signifies the fracture and collapse of the main roof, instability of overlying strata, and the formation of a compacted arch structure. The first mining-induced pressure event occurs prior to this point. Integrating the pronounced subsidence observed at 40 m with historical field measurements, it is reasonable to conclude that the first weighting is likely to occur between 40 and 45 m of face advance.

FIGURE 12

4.4.2 Prediction of the periodic weighting

During the mining process of an isolated island working face, following initial roof weighting, continued advancement induces periodic movement in the main roof strata behind the goaf, characterized by a cycle of “fracture–rotation–instability–rebalancing” (Li et al., 2023). This cyclic behavior leads to the periodic manifestation of mine pressure, known as periodic weighting (Bai et al., 2025). The main roof in this case consists of medium-grained sandstone, which exhibits high strength and good structural integrity, resulting in a relatively consistent fracture interval. Due to the elevated lateral support pressure typical of isolated island panels, the energy accumulation required for roof failure occurs more rapidly, potentially increasing the frequency of periodic weighting events. Additionally, the immediate roof, composed of siltstone with a moderate thickness of 6.3 m, facilitates the formation of a stable caved rock structure. This provides effective rear abutment support for the “arch-beam” structure of the main roof, thereby promoting a regular and predictable pattern of periodic pressure manifestation.

After the initial weighting, as the working face advances further, the main roof strata form an “arch-beam” structure, supported at the front by the coal wall and at the rear by the caved gangue in the goaf. The periodic weighting interval typically ranges from 40% to 60% of the initial weighting interval. Given the measured initial weighting distance of 40–45 m, the periodic weighting distance can be preliminarily estimated at approximately 16–27 m. Considering additional field data from previously operated working faces, where periodic weighting occurred at intervals of 18–20 m, it is comprehensively concluded that the periodic weighting interval for the 30,101 working face is expected to be 16–20 m.

4.5 Displacement of entries of 30,101 working face

The 30,101 working face had two entries: the 30,101 conveyor belt entry (30,101 CBE) and the 30,101 return airway (30,101 RAE). This section analyzes the stability of these two entries under varying advancing distances of the 30,101 working face. Simulation results are presented in Figure 13.

FIGURE 13

Figure 13 presents contour maps of vertical displacement in the surrounding rock of the roadways, with the cross-section taken along the coal seam inclination, and this orientation allows both the 30,102 and 30,101 working faces to be visualized, where the 30,102 face has been fully mined and its gob stabilized, while the 30,101 face is in active extraction (with the mining direction perpendicular to the plane of the figure). Beyond the qualitative observation of greater deformation in the 30,101 CBE, quantitative analysis of Figure 13 reveals distinct vertical displacement gradients across different regions of the cross-section, which directly reflect the influence of mining history and structural boundaries on surrounding rock stability.

In the region above the fully mined 30,102 gob, the vertical displacement of the roof strata is relatively stable. This moderate and uniform displacement indicates that the overlying strata above the 30,102 gob have completed post-mining compaction and subsidence, and the gob has reached a state of mechanical equilibrium. By contrast, the vertical displacement in the 30,101 active mining region exhibits significant spatial variation: along the 30,101 CBE (adjacent to the 30,102 gob via the 15 m-wide coal pillar), the roof vertical displacement reaches 20–24 mm (5 m to mining face), with a maximum of 245 mm observed directly above the 30,101 CBE roadway; in comparison, the roof displacement along the 30,101 RAE (farther from the 30,102 gob) is notably lower, ranging from 13 to 17 mm (5 m to mining face). This quantifiable difference—where the 30,101 CBE roof displacement is approximately 40%–50% higher than that of the 30,101 RAE—stems from the dual disturbance of the 30,101 active mining and the residual stress from the 30,102 gob, which causes the 15 m coal pillar to bear concentrated loads and transmit additional deformation to the adjacent 30,101 CBE. Additionally, the floor vertical displacement in the 30,101 mining area follows a similar trend: the 30,101 CBE floor exhibits upward displacement of 8–10 mm (due to gob-induced floor heave), while the 30,101 RAE floor displacement is only 5–7 mm (100 m to mining face). Therefore, it further confirms the amplified deformation effect of the adjacent 30,102 gob on the 30,101 CBE.

Figure 14, which extracts the roof, floor, and rib deformation curves from Figure 13, illustrates the variation of roadway surrounding rock deformation with distance from the 30,101 working face. As the distance to the working face decreases (i.e., closer to the active mining front), the deformation of all four surrounding rock zones (roof, floor, pillar-side rib, working face-side rib) gradually increases. Among the four zones, the roof deformation is consistently the largest (peaking at ∼24 mm at 0–20 m from the working face). The roof-deformation hierarchy of the 30,101 RAE and 30,101 CBE further validates the earlier observation that the 30,101 CBE is more vulnerable to stability issues.

FIGURE 14

Regarding the supporting pattern in real engineering site, the roadway employs a combined bolt-mesh-cable support system. The roof is supported by ∅20 mm × 2,100 mm HRB335 bolts spaced at 800 mm × 800 mm, while the ribs utilize ∅20 mm × 1,500 mm bolts with identical spacing. To reinforce the roof strata, ∅17.8 mm × 6,300 mm prestressed steel strand cables (200 kN pre-tension) are installed at intervals of 2,400 mm × 2,400 mm. The surface is retained by a ∅6 mm welded steel mesh with a 100 mm × 100 mm grid. Figure 15 presents representative deformation patterns observed in the 30,101 working face. Figure 15a illustrates the deformation control performance at the 30,101 RAE, where the roadway exhibited minimal deformation, indicating effective support. Figure 15b depicts the condition at the 30,101 CBE; due to the adjacent coal pillar, significant deformation, particularly at the roof, was observed. Figure 15c shows the open-off cut of the 30,101 working face, where bending of steel straps is visible, but the overall control performance remained acceptable owing to the use of hydraulic jacks.

FIGURE 15

Field monitoring was conducted from January to March 2024, the stress monitoring utilized high-precision pressure gauges equipped with automatic data loggers, enabling continuous, unmanned acquisition of support resistance and borehole stress data. The bolt load was measured every 2 h using vibrating wire sensors (accuracy ±3 kN). The diameter and the length of bolt were 22 and 2,200 mm, respectively; the borehole diameter was 30 mm and two 500 mm resin cartridges were filled to meet an end-encapsulated bonding pattern. Monitoring result at engineering field are shown in Figure 16, Figure 16a presents the loading changes of bolts at roof and ribs of both entries (Feng et al., 2022), Figures 16b,c shows the photos of monitoring site in entry. As can be seen, the bolt load increases over time due to mining-induced stress redistribution in the surrounding rock. 30,101 CBE exhibits higher bolt loads than 30,101 RAE, as the former is subjected to dual disturbances from the 30,102 gob and a 15 m narrow coal pillar, while the latter is only moderately disturbed by a 40 m wide coal pillar and intact coal. Specifically, in 30,101 CBE, the coal pillar side (15 m) bears greater loads than the mining side, reflecting intensified stress concentration near the narrow coal pillar, whereas the load difference between the two sides of 30,101 RAE is relatively small due to the stable stress environment of the wide coal pillar and intact coal.

FIGURE 16

4.6 Stability analyse of coal pillar between the mining faces

This section analyzes the stability of coal pillar between the 30,101 and 30,102 working faces. Figure 17 illustrates the plastic zone evolution around the 15 m coal pillar (between 30,101 and 30,102 working faces) as the 30,101 face advances to 20, 140, and 220 m. At 20 m advance, the plastic zone is narrowly confined to the 30,101 side of the coal pillar, with a penetration depth of only 2–3 m. This limited distribution reflects minimal new mining disturbance, as the coal pillar primarily bears residual stress from the fully mined 30,102 gob, retaining a large intact elastic core in its center.

FIGURE 17

When advancing to 140 m, the plastic zone expands bidirectionally: it extends 5–6 m inward from the 30,101 side and 3–4 m from the 30,102 gob side. The overlapping of 30,101’s advanced abutment pressure and 30,102’s residual stress intensifies coal pillar damage, reducing the elastic core to ∼5 m. Moreover, the localized shear zone observed in the bottom-right corner is attributed to the stress concentration at the boundary of the adjacent burned-out zone as the working face approaches.

At 220 m (near full extraction), the plastic zone approaches but does not fully penetrate the coal pillar, with a residual elastic core of ∼3 m. This stabilized state arises from balanced stress redistribution, as the coal pillar’s bearing capacity adapts to cumulative mining disturbance. The plastic zone evolution confirms the 15 m coal pillar maintains structural integrity while enduring significant stress concentration, providing a basis for support optimization.

5 Numerical results of water seepage tendency above the gob area

As discussed in the engineering background, coal seam #3 faces significant challenges from both surface and subsurface conditions. At the surface, numerous water bodies are present, particularly the Wulanmulun River, as shown in Figure 1c. Subsurface conditions are complicated by a severe combustion zone located near the gob areas of the mining faces, as illustrated in Figure 2.

The presence of fusain has high susceptibility to spontaneous combustion, also considering the widespread occurrence of sintered rocks resulting from past burning events, it is therefore essential to evaluate the roof cracking characteristics to prevent water seepage or intrusion during mining operations (Du et al., 2025). In this shallow-buried coal seam, the caving zone could frequently connect with overlying aquifers, leading to water seepage into the gob. This water accumulation softens the immediate floor strata, reducing its bearing capacity. To mitigate these risks, the longwall advance rate must be periodically reduced to allow for drainage operations and to ensure the stability of the support system on the softened floor. Therefore, this section employs UDEC to analyze roof caving patterns.

The choice of UDEC is driven by its discrete element method (DEM) framework, which is inherently well-suited for simulating the discontinuous features of fractured rock masses (He et al., 2018), and it is a critical requirement for analyzing roof fracture distribution (Kang et al., 2025), as the development and propagation of fractures in overlying strata involve distinct block interactions rather than uniform continuous deformation. Unlike FLAC3D, which relies on a continuous medium assumption and excels in capturing overall stress and displacement fields of intact or slightly fractured rock, UDEC offers superior precision in modeling the initiation, extension, and inter-block sliding/rotation of discrete rock fragments, which can generate a more realistic reproduction of the actual fracture evolution process in the overlying strata above the gob area. In this case, the model setup is depicted in Figure 18, which includes simulations along both the inclination and strike directions of the coal seam. Note that the inclination direction and the strike direction indicate how the profiles are extracted from the modelling results, with “inclination” and “strike” following the definitions in Figure 6b, denoted by “X” and “Y”, respectively.

FIGURE 18

As shown in Figure 18, the plastic zone distribution maps—generated from the UDEC model constructed along both the inclination and strike directions of the coal seam (Figure 18)—exhibit distinct evolutionary characteristics that align with the geometric boundaries of the two model orientations. For the profiles along the coal seam strike direction (Figures 19a–d), which focus on the 30,101 working face alone (with no visibility of the 30,102 working face, as the section cuts through the middle of the 30,101 face), the plastic zone evolution reflects localized strata damage tied to the advancement of this single working face. At the early advancing stages (20 and 40 m), plastic failure is confined primarily to the immediate roof (siltstone) and floor (fine-grained sandstone) within a narrow range around the 30,101 excavation. As the 30,101 face advances to 80 and 160 m, the plastic zone expands upward into the main roof (medium-grained sandstone) and downward into the deeper floor strata, with the vertical extent of failure increasing by approximately 2–3 times compared to the initial stage—this upward expansion directly correlates with the gradual loss of structural support in the overlying strata.

FIGURE 19

For the profile along the coal seam inclination direction (Figure 19e), which encompasses both the 30,101 and 30,102 working faces (consistent with the inclination direction model in Figure 17), the plastic zone distribution reveals a critical interconnection between the two working faces after full extraction of the 30,101 face. The plastic zones originating from the 30,101 working face and the pre-mined 30,102 gob area exhibit partial overlap in the intermediate coal pillar region, indicating that cumulative mining disturbance has weakened the integrity of this pillar.

These profiles shows that the plastic zone in the overlying strata extends laterally toward the burned-out zone (referenced in Section 2.2), which raises key considerations for water inrush risk. Additionally, UDEC simulations were post-processed to quantify fracture connectivity using two metrics: (1) Fracture density (number of fractures per m³): 12.8 fractures/m³ in the main roof; (2) Connectivity index (ratio of interconnected fractures to total fractures): 0.78. These metrics confirm that 78% of fractures in the water-conducting zone penetrate the bedrock-top interface, forming a continuous pathway between surface water (Wulanmulun River), burned-out zone pores (porosity = 0.25, measured via lab tests), and the mine. The estimated water inflow potential is 50 m³/h (calculated using Darcy’s law: Q = kAΔh/μ, where k = 1 × 10⁻⁴ m/s for fractured rock, A = 175 m × 29 m, Δh = 78 m). Therefore, it can be concluded that the water-conducting fracture zone of the 30,101 working face will penetrate the top interface of the bedrock and connect with the unconsolidated overburden on the surface. Atmospheric precipitation may infiltrate the goaf of the No. 3 coal seam through possible collapse fractures, and the accumulated water may also flow or seep into the mine via fractures, posing a hidden danger of water disasters to the mine. To mitigate this risk, the working face needs to be equipped with a water prevention and drainage system, safety technical measures to prevent sand and water inrush, and water spraying prevention measures at the working face.

6 Discussion

6.1 Calibration of numerical results

To calibrate the numerical results presented above, this section selects four indicators to evaluate precision: the stress concentration factor, water-conducting fracture height, elastic core width in the coal pillar, and periodic weighting interval. The results and their validity are summarized in Table 5. As shown, the simulated results generally fall within the reference ranges, with deviations remaining within acceptable limits.

It is worth to mention that a limitation of this calibration is the lack of in-situ monitoring data for direct validation. Follow-up work will conduct field measurements (e.g., borehole imaging for fracture height, fiber optic monitoring for stress) in the 30,101 working face to further refine model parameters. For now, the consistency between simulated results and empirical/peer-validated ranges ensures the model’s reliability for hazard assessment.

6.2 Targeted hazard control measures

Furthermore, the method of goaf management plays a critical role in controlling surface subsidence, particularly in shallow-buried coal seams where the fracture zone often extends directly to the surface. While this study focuses on the caving method (longwall mining with roof collapse), alternative approaches such as backfilling mining can significantly reduce strata movement and surface deformation. As noted by Skrzypkowski (2021), the selection of the goaf elimination method—whether by caving or backfilling—must be carefully balanced against the requirements for surface protection and the geological conditions of the overburden.

7 Conclusion

References

• 1

  BaiW. Y.YangX.HuY.XuQ.WangB.LiT.et al (2025). Study on mine pressure law and support resistance of working face under shallow buried gully terrain. Sci. Rep.15, 16553. 10.1038/s41598-025-01004-3

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 2

  BoH. Z.GuoG.LiH.WangY.JiangQ.HuS.et al (2023). Study on surface subsidence prediction method of shallow coal seam backfill-strip mining under the hard roof. Bull. Eng. Geol. Environ.82, 281. 10.1007/s10064-023-03284-3

  • CrossRef
  • Google Scholar
  • View reference in article

• 3

  ChenK.ZhangJ.XueG.HuangH.ChenW.HaoJ.et al (2019). Feasibility of monitoring hydraulic connections between aquifers using time-lapse tem: a case history in Inner Mongolia, China. J. Environ. Eng. Geophys.24, 361–372. 10.2113/jeeg24.3.361

  • CrossRef
  • Google Scholar
  • View reference in article

• 4

  ChenS.TangJ. P.PanY. S. (2024). Constitutive model of uniaxial compression for rock (Coal) and bursting liability index based on the structure ensemble dynamics theory. Rock Mech. Rock Eng.57, 10831–10848. 10.1007/s00603-024-04112-0

  • CrossRef
  • Google Scholar
  • View reference in article

• 5

  ChenW.GengS.ChenX.LiT.TsangaratosP.IliaI. (2025). Prediction of the height of water-conducting fissure zone for shallow-buried coal seams under fully mechanized caving conditions in Northern Shaanxi Province. Water17, 312. 10.3390/w17030312

  • CrossRef
  • Google Scholar
  • View reference in article

• 6

  DaiL. P.PanY.XiaoY.WangA.WangW.WeiC.et al (2024). Parameter design method for destressing boreholes to mitigate roadway coal bursts: theory and verification. Rock Mech. Rock Eng.57, 9539–9556. 10.1007/s00603-024-04042-x

  • CrossRef
  • Google Scholar
  • View reference in article

• 7

  DasA. J.MandalP. K.GhoshN.SinghA. P.KumarR.TewariS.et al (2023). Evaluation of energy accumulation, strain burst potential and stability of rock mass during underground extraction of a highly stressed coal seam under massive Strata-a field study. Eng. Geol.322, 107178. 10.1016/j.enggeo.2023.107178

  • CrossRef
  • Google Scholar
  • View reference in article

• 8

  DuX. H.XueJ.YuL.LeiW.MaH.CaoC. R.et al (2024). Coal damage and energy characteristics during shallow mining to deep mining. Energy291, 130375. 10.1016/j.energy.2024.130375

  • CrossRef
  • Google Scholar
  • View reference in article

• 9

  DuL.ZhuW.WangL.LiH.JiaoX.QinT. (2025). Using roof borehole electrical resistivity tomography to monitor roof water infiltration in a mine work face. Sci. Rep.15, 2621. 10.1038/s41598-025-86747-9

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 10

  DubinskiJ.StecK.KrupanekJ. (2023). Comprehensive use of the Gutenberg-Richter law and geotomography for improving seismic hazard evaluation in hard coal mines. Int. J. Coal Sci. Technol.10, 3. 10.1007/s40789-022-00561-6

  • CrossRef
  • Google Scholar
  • View reference in article

• 11

  EsterhuizenG. S.GearhartD. F.TuluI. B. (2018). Analysis of monitored ground support and rock mass response in a longwall tailgate entry. Int. J. Min. Sci. Technol.28, 43–51. 10.1016/j.ijmst.2017.12.013

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 12

  FengX. W.ZhangN.XueF.XieZ. Z. (2019). Practices, experience, and lessons learned based on field observations of support failures in some Chinese coal mines. Int. J. Rock Mech. Min. Sci.123, 104097. 10.1016/j.ijrmms.2019.104097

  • CrossRef
  • Google Scholar
  • View reference in article

• 13

  FengX. W.XueF.CarvelliV.ZhaoT.HeF.WangD. (2022). A novel rock bolting system exploiting steel particles. Int. J. Min. Sci. Technol.32, 1045–1058. 10.1016/j.ijmst.2022.08.003

  • CrossRef
  • Google Scholar
  • View reference in article

• 14

  FengD.ZhenhuaL.SongtaoL.XiaoleiL.GuodongL.XuanF.et al (2024). Breaking law of overburden rock and key mining technology for narrow coal pillar working face in isolated island. Sci. Rep.14, 13045. 10.1038/s41598-024-63814-1

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 15

  HallmanD. S. (2022). Foamed backfilling for combatting mine fires. Environ. Geotech.-J.9, 310–317. 10.1680/jenge.19.00079

  • CrossRef
  • Google Scholar
  • View reference in article

• 16

  HeS.LiY.AydinA. (2018). A comparative study of udec simulations of an unsupported rock tunnel. Tunn. Undergr. Space Technol.72, 242–249. 10.1016/j.tust.2017.11.031

  • CrossRef
  • Google Scholar
  • View reference in article

• 17

  HeF. L.LiL.LvK.QinB.XuX.MaQ.et al (2022). Study on evolution of front abutment pressure at working face in repeated mining of close-distance coal seams. Sustainability14, 12399. 10.3390/su141912399

  • CrossRef
  • Google Scholar
  • View reference in article

• 18

  JayanthuS.SinghT. N.SinghD. P. (2004). Stress distribution during extraction of pillars in a thick coal seam. Rock Mech. Rock Eng.37, 171–192. 10.1007/s00603-003-0003-2

  • CrossRef
  • Google Scholar
  • View reference in article

• 19

  KangX. T.MaY.HuJ.XuZ.ZengY.LiC. (2025). Study on the range of influence of faults on overburden fissures in the working face under repetitive mining action. Sci. Rep.15, 33539. 10.1038/s41598-025-18192-7

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 20

  LeT. D.BuiX. N. (2020). Effect of key parameters on top coal first caving and roof first weighting in longwall top coal caving: a case study. Int. J. Geomech.20, 04020037. 10.1061/(asce)gm.1943-5622.0001667

  • CrossRef
  • Google Scholar
  • View reference in article

• 21

  LiL.KongD. Z.LiuQ. Z.CaiH. C.ChenL. (2023). Study on law and prediction of surface movement and deformation in Mountain area under repeated mining of shallow coal seam. Bull. Eng. Geol. Environ.82, 76. 10.1007/s10064-023-03105-7

  • CrossRef
  • Google Scholar
  • View reference in article

• 22

  LiuZ.DongS.GuoX.LiX.ChenD. (2025). Research status of roof water and sand bursting disaster in thick bedrock stope of coal seam. Sci. Rep.15, 29490. 10.1038/s41598-025-14746-x

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 23

  PytelW.Palac-WalkoB.ButraJ.MertuszkaP. (2026). Friction stabilizers for rock bursts hazard mitigation in deep mines' environments. Tunn. Undergr. Space Technol.168, 107096. 10.1016/j.tust.2025.107096

  • CrossRef
  • Google Scholar
  • View reference in article

• 24

  QiJ. C.PengZ. J.GaoY. B.HeM. C.LiuK. (2025). Physical experiment and numerical simulation study on balanced mining based on the 110 mining method in shallow-buried thin bedrock with a thick weathered layer. Int. J. Geomech.25, 04025276. 10.1061/ijgnai.gmeng-11943

  • CrossRef
  • Google Scholar
  • View reference in article

• 25

  RaoG. Z.RaoY.WanJ.HuangQ.XieY.LaiQ.et al (2024). Long-term prediction modeling of shallow rockburst with small dataset based on machine learning. Sci. Rep.14, 16131. 10.1038/s41598-024-64107-3

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 26

  SkrzypkowskiK. (2021). Determination of the backfilling time for the zinc and lead ore deposits with application of the backfillcad model. Energies14, 3186. 10.3390/en14113186

  • CrossRef
  • Google Scholar
  • View reference in article

• 27

  TaiL. H.LiC.HuY.YuX.XuZ.ZhangX.et al (2024). Experimental study on the significance of pressure relief effect and crack extension law under uniaxial compression of rock-like materials containing drill holes. Sci. Rep.14, 1104. 10.1038/s41598-024-51490-0

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 28

  TuncayD.TuluI. B.KlemettiT. (2021). Investigating different methods used for approximating pillar loads in longwall coal mines. Int. J. Min. Sci. Technol.31, 23–32. 10.1016/j.ijmst.2020.12.007

  • CrossRef
  • Google Scholar
  • View reference in article

• 29

  UnverB.YasitliN. E. (2006). Modelling of strata movement with a special reference to caving mechanism in thick seam coal mining. Int. J. Coal Geol.66, 227–252. 10.1016/j.coal.2005.05.008

  • CrossRef
  • Google Scholar
  • View reference in article

• 30

  WangH. W.XuL. M.YuH. T.ZhangJ. Z. (2024). Research on prediction of high energy microseismic events in rock burst mines based on Bp neural network. Sci. Rep.14, 29934. 10.1038/s41598-024-81614-5

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 31

  WangZ. S.LiuY. L.SongZ. X.CongL.ZhuY. T. (2024). Rockburst mechanism and prevention technology for wide coal pillar return airway. Energy Sci. Eng.12, 4991–5007. 10.1002/ese3.1922

  • CrossRef
  • Google Scholar
  • View reference in article

• 32

  WenH. C.LiY.WuG.KongD.XuM.LiaoJ. (2026). Study on overlying strata breaking law and dynamic pressure characteristics of shallow buried coal seam mining under gully landform. Eng. Fail. Anal.183, 110276. 10.1016/j.engfailanal.2025.110276

  • CrossRef
  • Google Scholar
  • View reference in article

• 33

  WuZ. Y.SunQ.WangY. B. (2024). Evolution laws of water-flowing fracture zone and mine pressure in mining shallow-buried, hard, and extra-thick coal seams. Appl. Sci.-Basel.14, 2915. 10.3390/app14072915

  • CrossRef
  • Google Scholar
  • View reference in article

• 34

  XieZ. Z.ZhangN.FengX.LiangD.WeiQ.WengM. (2019). Investigation on the evolution and control of surrounding rock fracture under different supporting conditions in deep roadway during excavation period. Int. J. Rock Mech. Min. Sci.123, 104122. 10.1016/j.ijrmms.2019.104122

  • CrossRef
  • Google Scholar
  • View reference in article

• 35

  XieJ. H.DingZ. W.ZhuG. A.HaoX. Q.JiaJ. D. (2025). Research and engineering practice on mechanical reaming for pressure relief and rock burst prevention technology in hard coal seam roadways. Sci. Rep.15, 37368. 10.1038/s41598-025-02002-1

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 36

  YangR. B.WangH.ZhangP.GuoZ.LiM.HuS.et al (2025). Characteristics of coal spontaneous combustion induced by air leakage in the goaf of shallow buried close-distance coalbed. Combust. Sci. Technol.197, 5869–5887. 10.1080/00102202.2024.2412223

  • CrossRef
  • Google Scholar
  • View reference in article

• 37

  YuS. L.LiuX. (2024). Study on multi-indicator quantitative risk evaluation methods for different periods of coal spontaneous combustion in coal mines. Combust. Sci. Technol.196, 1628–1641. 10.1080/00102202.2022.2121608

  • CrossRef
  • Google Scholar
  • View reference in article

• 38

  ZhangP.TuluB.SearsM.TrackemasJ. (2018). Geotechnical considerations for concurrent pillar recovery in close-distance multiple seams. Int. J. Min. Sci. Technol.28, 21–27. 10.1016/j.ijmst.2017.12.012

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 39

  ZhangP.GearhartD.Van DykeM.SuD.EsterhuizenE.TuluB. (2019). Ground response to high horizontal stresses during longwall retreat and its implications for longwall headgate support. Int. J. Min. Sci. Technol.29, 27–33. 10.1016/j.ijmst.2018.11.020

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 40

  ZhangL. J.ZhouX.ZhouY.ZhouJ.GuoJ.ZhaiZ.et al (2022). Surface coal mining impacts on land use change and ecological service value: a case study in shengli coalfield, Inner Mongolia. Int. J. Coal Sci. Technol.9, 65. 10.1007/s40789-022-00518-9

  • CrossRef
  • Google Scholar
  • View reference in article

• 41

  ZhangP.SuD.KimB. H.MidlerE. (2022a). Comparison of measured and modeled casing deformations of a test well in a longwall abutment pillar. In U.S. rock Mechanics/Geomechanics symposium, 2022–2280.

  • Google Scholar
  • View reference in article

• 42

  ZhangP.EsterhuizenG.SearsM.TrackemasJ.MinoskiT.TuluB. (2022b). Roof stability and support strategies associated with longwall-induced horizontal stress changes in belt entries. Min. Metallurgy & Explor.39, 1873–1885. 10.1007/s42461-022-00634-9

  • CrossRef
  • Google Scholar
  • View reference in article

• 43

  ZhenE. Z.DongS.HuangJ.WangY.WangM.ZhangX.et al (2023). Analysis of the efarc non-pillar mining stope: roof failure and overlying pressure in inclined coal seams. Geomech. Geophys. Geo-Energy Geo-Resour.9, 151. 10.1007/s40948-023-00691-4

  • CrossRef
  • Google Scholar
  • View reference in article

• 44

  ZhuL.YaoQ.XiaZ.XuQ.YuL.LiX. (2022). Prevention of water and mud inrush in longwall top coal cave mining under shallow buried thin bedrock. Bull. Eng. Geol. Environ.81, 355. 10.1007/s10064-022-02846-1

  • CrossRef
  • Google Scholar
  • View reference in article