Introduction

Foreword

The overlying roof strata of coal seams are characterized by significant thickness and hardness, exhibiting high strength, strong integrity, and considerable caving intervals. These strata have a propensity for elastic energy accumulation and frequently act as key seismic source layers that induce roof-type rock bursts1,2,3,4,5. According to statistical data6,7,8,9, over the past decade, more than 700 rock bursts triggered by hard roof conditions have been recorded. Moreover, the continuous increase in underground mining intensity, extraction rates, and mining scope has resulted in a marked rise in both the frequency and severity of roof-type rock bursts. Consequently, understanding the behavior of hard roof strata is essential for developing reasonable and effective prevention and control technologies for rock bursts. This is a critical issue that must be addressed urgently to ensure the safe and efficient operation of mines10,11,12,13,14.

In response to the issue of thick and rigid roofs under high-intensity mining conditions being prone to dynamic impact disasters, numerous researchers have conducted extensive studies15,16,17. Wu Yiyi et al.18. investigated the mechanism of rock-burst disasters induced by the fracture of medium-to-low key hard roof strata in fully mechanized caving mining, validating the effectiveness of multi-point segmented hydraulic fracturing for impact prevention. Feng Cui et al.19. classified rock fracture modes into three types based on rock thickness, boundary conditions, and mining conditions: tensile fracture, tensile-shear fracture, and shear fracture. Xinyuan Tian20 studied the movement and fracture patterns of overlying strata under the interaction between supports and roof strata. Hu He et al.21. established a shallow-buried gradient composite plate model for thin bedrock strata and provided analytical solutions. Cai-Ping Lu et al.22. investigated the ground pressure characteristics of high and low hard roof strata, revealing the disaster-inducing mechanisms of articulated and cooperative fracture patterns in composite hard roof blocks. Xu Xuhui et al.23,24. uncovered the evolution patterns of bending moment, shear force, deflection, and strain energy during the initial crack initiation stage of hard roof strata under periodic weighting, providing criteria for assessing the state of roof strata in stopes. Wei Li et al.25. explored the fracture, energy accumulation, and release mechanisms of roof strata under mining-induced conditions and derived mechanical conditions for rock-bursts triggered by main roof fracture. Guo WeiYao et al.26,27,28,29. studied the energy accumulation and release mechanisms of coal seam roof strata during the transition from slow to rapid mining, examining the correlations among mining speed, elastic energy release, and rock-burst occurrence.

The aforementioned studies have provided valuable insights into the influence of face mining on strata pressure, energy accumulation, fracture behavior, and the stability of roof strata, thus offering guidance for safe mining practices on site. However, the rate of face advance is closely related to the intensity of surrounding rock pressure manifestations30,31,32,33. Excessive advance rates can lead to rapid energy accumulation and the fracturing of thick, hard roof strata within a short time frame, thereby increasing the likelihood of inducing rock bursts. Conversely, excessively slow advance rates may reduce mining efficiency, increase mining costs, and significantly hinder production progress. Therefore, based on the engineering context of Panel 311,304 in the Bayangaole Coal Mine, this paper investigates the evolution patterns of surrounding rock stress, energy, and microseismic events in thick coal seams with hard roof strata under varying advance rates34,35. The aim is to optimize a reasonable mining rate to mitigate or eliminate the potential for microseismic events to trigger rock bursts, thereby providing guidance for ensuring safe and efficient mine extraction.

Working face overview

Panel 311,304 in the Bayangaole Coal Mine represents the fifth extraction face within Panel 3113. It has a panel width of 300 m and a mining advance length of 2,460 m, specifically targeting the 3 − 1 coal seam. This seam is located at a depth of 672.6 m, with an average thickness of 5.67 m and an average dip angle of 1.5°. The immediate roof is composed of sandy mudstone with a thickness of 4.9 m, while the main roof consists of fine-grained sandstone, which has a thickness of 12.65 m. This sandstone is characterized by a dense lithology and is classified as a semi-hard to hard rock mass. The nearest sub-key stratum to the coal seam is a medium-grained sandstone layer, measuring 17.3 m in thickness. During mining operations, the roof demonstrates poor caving characteristics and forms long cantilevers, frequently leading to extensive roof hanging. This condition results in the accumulation of substantial elastic energy and the formation of stress concentration zones, which are highly susceptible to sudden coal-rock fracturing under both near- and far-field loading, potentially triggering rock bursts. The strata thicknesses of the roof and floor in Panel 311,304 are illustrated in Fig. 1.

Fig. 1
Fig. 1The alternative text for this image may have been generated using AI.
Full size image

Borehole histogram of panel 311,304.

To monitor microseismic activity during the mining of the working face, a 32-channel ARAMIS M/E microseismic monitoring system has been deployed. This system forms a joint monitoring network through the installation of microseismic probes at the gateways of the working face and within the mining-affected area, thereby enabling comprehensive real-time monitoring of the face. The layout of the monitoring points is depicted in Fig. 2.

Fig. 2
Fig. 2The alternative text for this image may have been generated using AI.
Full size image

Layout plan of microseismic measurement points in No. 311,304 working face.

Evolution of surrounding rock stress and energy in stopes under different mining rates

Using panel 311,304 as the numerical simulation object, the FLAC3D software was utilized to create a numerical model with dimensions of length × width × height = 304 m × 665 m × 120 m. Fixed constraints were applied to the four sides and the bottom boundary of the model, while the top boundary was designated as a free boundary, subjected to a self-weight load (q) corresponding to the thickness of the overlying strata. Here, q is defined as \(q=\gamma H\), where γ represents the unit weight (25 kN/m³) and H denotes the thickness of the overlying strata (653 m), resulting in q = 16.325 MPa. The mining panel was oriented along the Y-axis and advanced in the negative X-axis direction. The Mohr-Coulomb criterion was employed as the strength criterion. The established numerical model and the schematic diagram of panel mining are illustrated in Figs. 3 and 4, respectively, with the objective of investigating the stress and energy evolution laws of the surrounding rock under varying mining rates.

Fig. 3
Fig. 3The alternative text for this image may have been generated using AI.
Full size image

Numerical model of working face.

Fig. 4
Fig. 4The alternative text for this image may have been generated using AI.
Full size image

Mining simulation schematic diagram of working face.

To accurately characterize the influence of mining rate effects on the stress transfer process and the stability of the surrounding rock at the working face, a method that controls the stress release rate of the surrounding rock is employed to simulate the time-dependent behavior of the working face advance. This approach facilitates an intuitive back-analysis of the relationship between the stress-strain of the surrounding rock and time. The specific simulation scheme is detailed in Table 1.

Table 1 Numerical simulation scheme of working face.
Full size table

Evolution laws of surrounding rock stress in the working face stope

Under varying mining rates at the working face, the stress release rate of the surrounding rock in the stope exhibits significant variation, resulting in differing degrees of alteration in the stress state of the surrounding rock. According to the simulation plan, vertical stress nephograms of the hard roof at different mining rates were simulated and are presented in Fig. 5.

Fig. 5
Fig. 5The alternative text for this image may have been generated using AI.
Full size image

Vertical stress distribution cloud map of roof under different mining rates. (a) Rapid mining, (b) Relatively rapid mining, (c) Low-speed mining, (d) Slow mining.

Analysis of Fig. 5 indicates that as the mining rate of Panel 311,304 decreases, the stress adjustment range of the stope roof expands correspondingly. Within the goaf area, the vertical stress of the roof continues to diminish, exhibiting a relatively significant degree of stress adjustment. Furthermore, the levels of stress concentration and the affected area within the coal-rock mass ahead of the working face, near the goaf of Panel 311,305, progressively increase. Consequently, the likelihood of rock burst hazards occurring in this region also escalates.

Based on the stress monitoring results from Survey Line 1 and Survey Line 2 during the excavation process, the distribution curves of vertical stress in the hard roof of the stope and the abutment pressure ahead of the coal wall under varying mining rates have been obtained, as illustrated in Fig. 6.

Fig. 6
Fig. 6The alternative text for this image may have been generated using AI.
Full size image

Vertical stress variation curves of the stope roof under different mining rates. (a) Rapid mining, (b) Relatively rapid mining, (c) Low-speed mining, (d) Slow mining.

Analysis of Fig. 6 indicates that as the mining rate of the working face transitions from rapid to slow, the stress adjustment of the surrounding rock in the excavated area becomes more adequate. This adjustment allows the surrounding rock stress to gradually transfer and release, resulting in a transition in the evolution of vertical roof stress from uneven distribution towards uniformity. Concurrently, the abutment pressure ahead of the coal wall increases from 24.2 MPa to 31.4 MPa, with the distance between its peak and the coal wall expanding from 13.4 m to 18.9 m. In other words, a higher mining rate of the working face results in a lower abutment pressure ahead of the coal wall and a smaller distance between the pressure peak and the coal wall. However, if the distance between the abutment pressure peak and the coal wall is excessively small, it may induce safety hazards such as coal wall spalling or even rock bursts. Therefore, for thick coal seams with hard roofs, determining a reasonable mining rate for the working face is crucial to ensuring safe and efficient mine production.

Energy evolution patterns of hard roof in the working face

The accumulation of substantial elastic energy within the coal-rock mass serves as the primary energy source for inducing rock bursts. According to the principle of energy conservation, and assuming no energy exchange occurs between the coal-rock mass and the external environment, the expression for the stored elastic strain energy (Ue) in the coal-rock mass, influenced by mining excavation, can be derived from the generalized Hooke’s law as follows:

$${U_{\text{e}}}{\text{=}}\frac{1}{{2E}}\left[ {\sigma _{1}^{2}+\sigma _{2}^{2}+\sigma _{3}^{2} - 2\mu \left( {{\sigma _1}{\sigma _2}+{\sigma _1}{\sigma _3}+{\sigma _2}{\sigma _3}} \right)} \right]$$
(1)

In the formula: σ₁, σ₂, σ₃ represent the maximum principal stress, intermediate principal stress, and minimum principal stress respectively; E is the elastic modulus of the coal-rock mass; µ is the Poisson’s ratio.

Based on Eq. (1) and the stress-energy data of the coal-rock mass obtained from numerical simulations utilizing the Mohr-Coulomb criterion, a calculation command was formulated in Fish language to determine the elastic strain energy stored within the coal-rock mass. This approach produced nephograms of elastic strain energy density distribution and distribution curves for the hard roof in the stope, reflecting the effects of varying mining rates, as illustrated in Figs. 7 and 8.

Fig. 7
Fig. 7The alternative text for this image may have been generated using AI.
Full size image

Elastic strain energy distribution of roof under different mining rates. (a)Rapid mining, (b) Relatively rapid mining, (c) Low-speed mining, (d) Slow mining.

Fig. 8
Fig. 8The alternative text for this image may have been generated using AI.
Full size image

Energy distribution curves of hard roof under different mining rates.

Analysis of Figs. 7 and 8 indicates that varying mining rates at the working face result in different degrees of energy accumulation in the hard roof. Specifically, a faster mining rate leads to a higher energy accumulation in the hard roof. In contrast, a slower mining rate permits most of the energy in the hard roof to be released during the mining process, resulting in a lower degree of energy accumulation. Furthermore, the goaf significantly influences the energy accumulation in the roof. When the mining rate is constant, the energy accumulation in the roof decreases with increasing distance from the goaf. Conversely, when the distance from the goaf is held constant, the energy accumulated in the roof increases significantly in an approximately exponential manner as the mining rate increases, thereby making it more susceptible to inducing high-energy microseismic events.

Patterns of microseismic activity in the stope under the influence of mining rate effects

Relationship between mining rate and energy/frequency of microseismic events

Based on the microseismic monitoring network system implemented during the mining period of Panel 311,304, a statistical analysis was conducted on the microseismic monitoring data collected from April 29 to July 7, 2022. Data deemed invalid due to factors such as production halts for maintenance were excluded from the analysis. The distribution of daily advance in relation to the frequency of microseismic events and the corresponding microseismic energy was plotted, as illustrated in Fig. 9.

Fig. 9
Fig. 9The alternative text for this image may have been generated using AI.
Full size image

Microseismic events statistics of mining face during the mining.

Analysis of Fig. 9 indicates that when the daily advance of the working face is less than 4.8 m, both the microseismic frequency and energy increase significantly. However, when the daily advance exceeds 4.8 m, both parameters exhibit a declining trend while remaining at relatively high levels, particularly regarding the maximum daily microseismic event energy. This phenomenon can be attributed to the increased disturbance to the coal-rock mass in the stope due to the greater daily advance. In thick coal seams with hard roofs, the average daily energy accumulation in the surrounding rock of the stope rises, resulting in a relatively higher frequency and energy of microseismic events. Furthermore, as the daily advance continues to increase, the frequency of low-energy microseismic events also rises, alongside an increased likelihood of inducing high-energy microseismic events.

Spatiotemporal evolution laws of microseismic events in stope under the effect of mining rate

(1) Temporal distribution patterns of microseismic events.

Based on the microseismic monitoring results from Panel 311,304, the relationship among daily advancement, total microseismic energy, and the frequency of microseismic events from April 29 to July 7, 2022, is illustrated in Fig. 10.

Fig. 10
Fig. 10The alternative text for this image may have been generated using AI.
Full size image

Energy and frequency distribution curves of microseimic events.

Analysis of Fig. 10 indicates that during the mining period, the frequency and energy of microseismic events in the thick coal seam with a hard roof generally exhibit trends of increase or decrease corresponding to variations in the mining rate. As the working face advances, the overall microseismic frequency and energy in the stope demonstrate an upward trend. During periods of uniform mining rates, the microseismic frequency and energy initially decline before rising. Furthermore, the longer the duration of high uniform mining rates, the greater the increase in frequency and energy. This phenomenon can be attributed to the extensive hanging roof span in the goaf area of the hard roof, which is less prone to caving, leading to the accumulation of substantial elastic energy in the surrounding rock of the stope and, consequently, inducing a higher number of high-energy microseismic events. Therefore, it can be concluded that a reasonable mining rate serves as an effective prevention and control measure for managing high-energy microseismic events in stopes with hard roofs.

(2) Spatial distribution patterns of microseismic events.

This study investigates the spatial distribution patterns of microseismic events in the stope, focusing on the effects of mining rate during the extraction period. Microseismic data were collected from two consecutive 10-day mining intervals: the first from May 13 to May 22, 2022, with a mining rate of 7.04 m/d, and the second from June 5 to June 14, 2022, with a mining rate of 5.36 m/d. The spatial distribution of microseismic events in the stope under these varying mining rates is illustrated in Fig. 11.

Fig. 11
Fig. 11The alternative text for this image may have been generated using AI.
Full size image

Spatial distribution of microseismic events in mining areas under different mining rates.

Analysis of Fig. 11 indicates that microseismic events in the stope, influenced by varying mining rates, exhibit distinct spatial distribution characteristics. The mining speed affects the degree of stress adjustment in the surrounding rock of the stope, thereby impacting the spatial distribution of microseismic events. Specifically, as the mining rate decreases, the distribution range of microseismic events expands. Under low-speed mining conditions, microseismic events are primarily located ahead of and behind the working face coal wall, whereas under high-speed mining, they tend to be concentrated behind the coal wall. This pattern explains the occurrence of rock bursts induced by hard roof conditions, which are frequently observed behind the working face coal wall during high-speed mining.

Optimization of working face mining rate

Based on the patterns of microseismic activity observed in thick coal seams with hard roof stopes, the mining rate of the working face is optimized according to various rock-burst hazard zones. This optimization aims to reduce the frequency of high-energy microseismic events occurring in the stope, while also improving the spatial distribution of these events. Ultimately, this approach ensures the safe and efficient production of the working face.

(1) Rock-burst hazard zones in the working face.

The average mining depth of Panel 311,304 is 672.6 m. During the mining process, the roadways will be significantly affected. As the panel traverses the initial weighting and square areas, both energy activities and hazard levels will escalate. Furthermore, in the final mining stage of the panel, roof activities intensify, thereby heightening the likelihood of rock burst incidents. The panel roadways contain multiple intersection zones, which further increases the probability of rock burst occurrences during mining. Additionally, the goaf of Panel 311,305 will exert a certain influence on the mining process of Panel 311,304, consequently elevating its rock burst risk.

By comprehensively considering factors such as mining depth, hard and thick strata, roadway intersections, roof activity, and goaf conditions in Panel 311,304, this study analyzes their cumulative effects on the delineation of rock burst hazard zones and their intensity levels. The rock burst hazard zones during the mining period of the working face have been systematically defined, as illustrated in Fig. 12.

Fig. 12
Fig. 12The alternative text for this image may have been generated using AI.
Full size image

Division map of rock burst hazard zones during mining in the 311,304 working face.

(2) Optimization of mining speed in different rock-burst hazard zones.

As indicated by the division of rock-burst hazard zones in the working face, as shown in Fig. 12, the extraction areas classified as low rock-burst hazard zones include those at depths of 20–40 m, 80–250 m, and 350–400 m. The corresponding mining dates for these areas are May 2 to May 5, May 14 to June 18, and June 26 to July 7, 2022, respectively. Conversely, the extraction areas at depths of 0–20 m, 40–80 m, and 250–350 m are classified as moderate rock-burst hazard zones, with corresponding mining dates of April 29 to May 1, May 6 to May 13, and June 9 to June 25, 2022. The high rock-burst hazard zone is located within the 20 m section of the return airway behind the retreat roadway. Based on on-site microseismic monitoring data, the distribution curves of daily advance versus microseismic events in the stope for different rock-burst hazard zones were obtained, as illustrated in Fig. 13.

Fig. 13
Fig. 13The alternative text for this image may have been generated using AI.
Full size image

Distribution of microseismic events with the different rock burst risk areas.

Analysis of Fig. 13 indicates that the total daily energy, maximum energy, and frequency of microseismic events generally increase with the rise in daily advance. However, the evolution patterns between microseismic energy/frequency and daily advance differ across various hazard zones. In the low rock-burst hazard zone, when the daily advance is less than 6.4 m, both microseismic energy and frequency exhibit a gradual upward trend as the daily advance increases. When the daily advance exceeds 6.4 m, both parameters increase significantly, accompanied by a rise in the frequency of high-energy microseismic events. In the moderate rock-burst hazard zone, when the daily advance reaches 4.8 m, microseismic energy and frequency demonstrate a marked increase with further advances. The high rock-burst hazard zone is situated outside the stope area. Considering the geological and mining technical conditions of the working face, it is concluded that the mining rates for the low, moderate, and high rock-burst hazard zones should not exceed 6.4 m/d, 4.8 m/d, and 3.2 m/d, respectively.

(3) Validation of optimized mining rate effectiveness for the working face.

Based on the optimized mining rates established for the working face, the distribution patterns of microseismic events in the hard roof stope across various rock-burst hazard zones were monitored and documented, as illustrated in Fig. 14.

Fig. 14
Fig. 14The alternative text for this image may have been generated using AI.
Full size image

Distribution of microseismic events after optimization of the mining rate. (a) Low rock-burst hazard, (b) Moderate rock-burst hazard.

Analysis of Fig. 14 indicates that in the low rock-burst hazard zone, when the mining rate of the stope was maintained below 6.4 m/d, the maximum total microseismic energy, maximum daily energy, and maximum frequency were recorded at 9.65 kJ, 3.36 kJ, and 28 events, respectively. In the moderate rock-burst hazard zone, when the mining rate was controlled below 4.8 m/d, the corresponding values were 2.13 kJ, 5.84 kJ, and 40 events, all remaining at relatively low levels. This suggests that maintaining mining rates below 6.4 m/d in the low rock-burst hazard zone and below 4.8 m/d in the moderate rock-burst hazard zone is a reasonable approach, ensuring both mining efficiency and safety. To date, mining operations at this working face have been completed without any high-energy microseismic events, leading to significant cost savings and a notable reduction in the labor intensity associated with rock-burst prevention measures.

Conclusion

(1) As the mining rate at the working face increases, the peak position of the abutment pressure in front of the coal wall shifts closer to the goaf. The vertical stress evolution in the roof transitions from a state of uniformity to one of non-uniformity, with the accumulated elastic energy in the roof rising significantly in a quasi-exponential manner. This increase in elastic energy consequently heightens the likelihood of coal wall spalling and rock bursts.

(2) In stopes with hard roofs, both the energy and frequency of microseismic events generally increase with higher daily advance rates, thereby raising the likelihood of inducing high-energy microseismic events. The mining rate at the working face alters the spatiotemporal distribution patterns of microseismic events. Under low-speed mining conditions, these events tend to concentrate ahead of the coal wall, whereas during rapid mining, they predominantly occur behind the coal wall. This observation explains why rock-bursts induced by hard roof conditions often occur behind the coal wall.

(3) Based on the patterns of microseismic activity observed in the working face stope and the identified rock-burst hazard zones during mining operations, it was determined that the mining rates for low and moderate rock-burst hazard zones should be maintained below 6.4 m/d and 4.8 m/d, respectively. Field studies have demonstrated that the recommended mining rates for various hazard zones can ensure both the safety and efficiency of production at the working face.