Research on the technology of unloading pressure and preventing punching in extra-thick coal seams

Abstract

This study addresses stress concentration in roadways and elastic energy accumulation in the cantilever beam of hard roofs within the goaf of extra-thick coal seams. Using the 61,607 working face of Longwanggou Mine as a case study, the key roof layer was identified through theoretical calculations, and hydraulic fracturing parameters were optimized via zoning and grading. Stress changes, displacement, and fracture distribution in the downtrack and retreat areas were analyzed using 3DEC software before and after fracturing. Microseismic monitoring evaluated energy and frequency of microseismic events, while borehole imaging observed post-fracturing fracture distribution. Results show that hydraulic fracturing reduces peak compressive stress from 34.8 to 19.2 MPa, representing a 44.8% reduction, and tensile stress in the mined-out void from 12.7 to 5.8 MPa, marking a 54.3% reduction. The high-stress concentration zone shrank from 68 to 20 m, corresponding to a 70.6% reduction, improving stress redistribution and structural stability. Roof subsidence increased by a factor of 1.96 to 2.07, facilitating earlier and more uniform subsidence of the hard rock layer and reducing overhang risks. Microseismic events with energy levels of 9500 J or greater decreased by 79%, confirming effective stress relief. Borehole imaging verified that hydraulic fracturing induced new fractures and expanded existing ones in the hard roof, disrupting its integrity, reducing seismic risks, and improving mining safety.

Introduction

As coal mining extends to greater depths, dynamic disasters associated with strong mining-induced stress under hard roof conditions in extra-thick seams have become more frequent, posing serious risks to operational safety. The hard roof often fails to collapse promptly during extraction, leading to the formation of extensive overhanging areas. Sudden failure of these structures can generate high-impact loads, concentrated stresses, and disturbances caused by roof bending and fracturing, which may trigger dynamic events such as rock bursts and impact ground pressure^1–7.

The overburden fracture caused by coal mining activities is the main source of forming strong formation dynamic behavior^8–15. Zhang¹⁶ believes that impact ground pressure is caused by stress concentration and transient changes in the soft zone. Dou^17,18 suggested that when the system is under a high static load, even a small dynamic disturbance can result in significant dynamic responses. Rong¹⁹ argued that the regional stress field determines the baseline stress level of the mining stress environment, thereby increasing the likelihood of impact ground pressure events. Therefore, effective control of strong rock pressure in large mining spaces requires proactive strategies, such as pre-fracturing of the hard roof, rather than relying solely on passive control measures²⁰.

To mitigate accidents caused by the rupture of thick and hard coal seams, roof collapse can be transformed from sudden, irregular impact failures to periodic, controllable, and segmented movements. Extensive research using physical experiments and numerical simulations has demonstrated the effectiveness of underground hydraulic fracturing in weakening hard roofs. Matsui²¹ reported that hydraulic fracturing outperforms deep-hole blasting. Shen²² and Yu²³ confirmed through field tests that it significantly reduces roof strength and integrity, thereby shortening collapse distances. Kang²⁴ and Huang²⁵ further explored fracture propagation and stress relief mechanisms associated with hydraulic fracturing.Despite these advances, most mines still rely on single-parameter fracturing, which limits pressure relief range and continuity, hindering effective roof control.

This study investigates the 61,607 working face of the Longwanggou Coal Mine. Key roof strata were identified through theoretical analysis, and fracturing parameters were optimized based on zonation and grading. Using 3DEC software²⁶, stress, displacement, and fracture development before and after fracturing were analyzed. Theoretical results were validated through simulations and field tests, offering support for the application of zoned, hierarchical top-cutting and pressure-relief technologies in extra-thick seam mining.

Identification of the top plate and design of fracturing layers in extra-thick coal seams

During the coal mining process, one or multiple layers of hard and thick rock strata are frequently present above the coal seam. These strata play a pivotal role in governing the deformation and movement of the overlying rock layers and are commonly referred to as “key strata”. Key strata can be categorized into major key strata and minor key strata. In comparison to other strata, key strata are characterized by their substantial thickness, high strength, minimal deformation, synchronized failure with overburden rock, and the ability to retain a certain load-bearing capacity post-fracture. These distinctive attributes underscore the critical importance of key strata in the context of mining operations.

Under natural geological conditions unaffected by anthropogenic disturbances, the self-weight stress at any given point within a rock mass can be theoretically defined as the vertical stress induced by the weight of overlying strata per unit area, expressed as γh, where γ represents the unit weight of rock and h denotes the burial depth. However, for engineering scenarios involving excavated overburden strata, the mechanical behavior of rock layers exhibits greater complexity. Specifically, each stratum must sustain not only its intrinsic self-weight load but also additional stresses arising from interactive deformation with adjacent overlying layers. This stress superposition phenomenon necessitates a modified analytical approach for vertical stress calculation. Based on elastic layered system theory, a fundamental calculation method for vertical stress distribution in multi-layered overburden systems is proposed as follows:

According to the definition and deformation characteristics of the key layer, if there are m layers of rock that deform in a synchronous and coordinated manner, then the lowest rock layer is the key layer. Then the support characteristics of the key layer can be known:

1

where -load formed by layer m on laye 1; γ_i-capacitive weight of layer i, (i = 1,2,…,m); Ei—modulus of elasticity of layer i, (i = 1,2,…,m).

If the deformation of layer m + 1 is less than the deformation characteristic of layer m, then there must be no longer any need for the layer above layer m + 1 to carry any of the loads to which it is subjected:

2

When the m + 1st rock layer, the bending subsidence of its lower layers is greater than the bending subsidence of this layer, the layers above the m + 1st layer do not need the lower layers to carry the load, then there must be:

3

Equation (3) in the form of a load comparison is really a key layer of the stiffness (deformation) discrimination conditions. The geometrical significance of this is that the deflection of the m + 1st layer is less than the deflection of the lower layer.The specific method for identifying key rock layers is as follows: the overlying strata above the coal seam are sequentially numbered as the 1st, 2nd, 3rd, …, mth, …, nth layers (m < n). Starting from the 1st layer, calculations are performed layer by layer until the criterion defined by Eq. (3) is satisfied. The m-th layer that first meets this condition is identified as a key rock layer. Subsequently, all remaining key rock layers are determined using the same formula and method, starting from the first identified key layer. Once the positions of the key layers are determined, the corresponding weak rock layers governed by these key layers can also be identified. Therefore, by substituting the data in Table 1 into Eqs. (1)–(3), the associated weak rock layers were calculated.

Table 1.

Physical and mechanical parameters of roof and bottom plates.

Serial number	Lithology	Layer thickness (m)	Tensile (MPa)	Capacity (KN/m3)	Spring mold (GPa)
22	Siltstone	23.03	1.64	22.41	21.31
21	Sandy mudstone	2.00	1.34	23.62	27.67
20	Siltstone	15.67	1.64	22.41	21.31
19	Shale	1.65	1.06	24.21	25.50
18	Siltstone	4.46	1.56	22.41	19.80
17	Sandy mudstone	19.39	1.34	23.64	27.67
16	Siltstone	3.40	1.56	22.41	19.80
15	Sandy mudstone	3.66	1.34	23.64	27.67
14	Siltstone	5.43	1.56	22.41	19.80
13	Sandy mudstone	14.07	1.34	23.64	27.67
14	Siltstone	10.14	1.56	22.41	19.80
13	Sandy mudstone	3.35	1.34	23.64	27.67
12	Siltstone	1.70	1.56	22.41	19.80
11	Sandy mudstone	12.19	1.34	23.64	27.67
10	Siltstone	13.91	1.56	22.41	19.80
9	Shale	0.10	1.06	24.21	25.50
8	Coal	1.65	1.22	14.38	25.07
7	Shale	1.00	1.06	24.21	25.50
6	5 Coal	0.35	1.22	14.38	25.07
5	Sandy mudstone	3.77	1.34	23.64	27.67
4	Siltstone	8.48	1.56	22.41	19.80
3	Sandy mudstone	2.95	1.34	23.64	27.67
2	6 Coal Loading	1.15	1.22	14.38	25.07
1	Shale	4.85	1.06	24.21	25.50
0	6 Coal	20.00	1.22	14.38	25.07

According to the above formula, the load formed by each top rock layer on the lower key layer is calculated separately, and the location of its key layer is discriminated, and the results of the discrimination are shown in Table 2.

Table 2.

Identification of key layers of roof of 61,607 working surface.

Lithology	Thickness (m)	Height (m)
Siltstone	8.48	37.43
Siltstone	13.91	58.21
Siltstone	10.14	85.59

A comprehensive analysis was conducted to identify the key rock layers overlying the coal seam, which include an 8.48-m-thick coarse sandstone layer, a 13.91-m-thick coarse sandstone layer, and a 10.14-m-thick coarse sandstone layer. The calculations of the collapse zone revealed that, following the extraction of the No. 6 coal seam and the interbedded gangue, the development of the free-falling airspace zone extended up to the 8.48-m-thick coarse sandstone layer, the 13.91-m-thick coarse sandstone layer, the 10.14-m-thick coarse sandstone layer, and the 14.07-m-thick sandy mudstone layer.

The results indicate that the 8.48-m-thick coarse sandstone layer, the 13.91-m-thick coarse sandstone layer, the 12.19-m-thick sandy mudstone layer, the 10.14-m-thick coarse sandstone layer, and the 14.07-m-thick sandy mudstone layer constitute the key rock layers requiring targeted treatment. Consequently, the final borehole location must penetrate into the 14.07-m-thick sandy mudstone layer to ensure effective control and management of the overlying strata. This approach is critical for mitigating potential ground control issues and ensuring the stability of the mining environment.

Top hydraulic fracturing parameter design

Introduction to the hydraulic fracturing process

The double-end blocking and fracturing technology is characterized by the absence of a slit link, with both blocking and fracturing operations being executed through two distinct systems. This technology primarily segments the pre-pressure area into isolated compartments via blockers positioned at both ends, facilitating the fracturing of the targeted coal-rock mass through the application of high-pressure liquid. The operational flow of this system is illustrated in Fig. 1

Fig. 1.

Double-end plugging and fracturing technology system flow chart.

In light of the conditions at the 61,607 working face, the double-end blocking and fracturing technology was selected. The procedural workflow is delineated as follows:

1. Drilling Construction: A drilling rig was employed to construct fracturing drill holes with diameters of 63.5 mm and 75 mm, adhering to the design specifications. Upon reaching the designated depth, drilling was halted, and the drill holes were flushed with water for a minimum of 2 min to ensure the smoothness of the hole walls and the absence of coal-rock mud residue. Subsequently, the push rod operation was initiated.

2. Replacement of Single-Pipe Blocking Fracturing Unit: The A-type and B-type fracturing units were connected to form a single-pipe blocking fracturing unit. This assembly was securely attached to the high-pressure push drilling rod via a connector. Utilizing the drilling rig, the single-pipe blocking fracturing unit was advanced to the fracturing point near the bottom of the drilled hole, after which the drilling rig and high-pressure push drilling rod were fixed in place.

3. Blocking and Fracturing: The control valve installed in the pipeline was closed, and the high-pressure pump was activated. Upon the pressure reaching approximately 30 MPa, the control valve was opened to harness the high-pressure impact from the nozzle of the fracturing unit to disrupt the rock layer of the roof plate.

4. Pressure Relief and Shifting: Following the completion of single-point fracturing, the pump station was turned off, and the water relief valve was opened. Once the pressure gauge indicated that the pipeline pressure had decreased to "0", the drilling rig was activated to retract the rods to the subsequent fracturing point. The drilling rig was then turned off, and both the drilling rig and high-pressure push drilling rods were secured.

Steps 3 and 4 were repeated to accomplish the fracturing operations at all designated fracturing points. The rig was subsequently moved to the next hole to continue the fracturing operations.

Design of hydraulic fracturing parameters

1. Preferred spacing of drill hole sets

Drawing on engineering experience from hydraulic fracturing operations in roof strata across mines in Shaanxi, Inner Mongolia, Shanxi, and other regions, it has been found that when the water injection pressure is maintained at no less than 30 MPa and the fracturing duration exceeds 30 min, the effective fracturing radius typically ranges from 5 to 10 m. In practical design schemes, an appropriate safety margin is generally incorporated, resulting in recommended drill hole spacing of 8 to 15 m. Field applications across various sites have consistently demonstrated favorable fracturing performance under these parameters.

Based on empirical analogical analysis, the hydraulic fracturing boreholes for the roof of the 61,607 working face are preliminarily designed with an inclination angle between 50° and 70°, and a vertical spacing ranging from 8 to 15 m.

• (2)Optimization of drill hole diameter

In double-end sealing hydraulic fracturing technology, the influence of borehole diameter on the propagation radius of fractures is relatively limited. During the initial application stages of this technology, borehole diameters typically ranged from 50 to 63.5 mm, primarily constrained by the dimensions of the borehole sealers. With advances in the materials used for expansion rubber and improvements in processing techniques, the Coal Academy of Sciences has significantly enhanced the performance of borehole sealers across a range of specifications, enabling customized production based on project requirements.

Considering the relatively long borehole lengths required for fracturing operations in the roof of the 61,607 working face, it is recommended to adopt drill bits with diameters ranging from 63.5 to 75 mm for construction purposes.

• (3)Water injection pressure preference

Given the uncertainty in the fracture toughness of the rock mass, the required water injection pressure is estimated based on its tensile strength. Physically, fracture occurs when the fluid pressure exceeds the combined effect of tensile strength and in-situ stress, leading to tensile failure and crack initiation.

4

where p₂-Estimated value of cracking pressure required for hydraulic fracturing;σ₁, σ₃-Maximum and minimum principal stresses at the fracturing point, influenced by depth, mining history of the coal seam in which it is located and neighboring seams, and mining geological conditions; R_t-Tensile Strength of Fractured Top Rock Formations.

Considering that no geostress test work has been carried out in Longwanggou at present, according to the hydraulic fracturing construction experience of the roof plate of multiple mines in Shaanxi, Inner Mongolia, Shanxi and other places, the preliminary design of water injection pressure ≥ 30 MPa (according to the site conditions, it is recommended to increase the fracturing pressure to 50 MPa and above, and the flow rate should meet the conditions of maximum fracturing pressure of no less than 80 L/min), fracturing time ≥ 30 min minutes, and later adjusted in real time according to the The fracturing time is ≥ 30 min minutes, and later adjusted in real time according to the fracturing results on site.

Hydraulic fracturing program design

In order to strengthen the decompression work in the two tunnels and the retreat area of 61,607 working face, suitable hydraulic fracturing construction of the roof plate is selected by zoning and grading, so as to reduce the impact dynamic load generated by the collapse of the roof plate and at the same time to eliminate the influence of stress concentration of the tectonics, prevent and control the disasters of pressurized frame and impact ground pressure, and ensure the safety of the working face for back-mining. The details are shown in Figs. 2 and 3.

Fig. 2.

Layout plan of hydraulic fracturing drilling holes on the roof of 61,607 two parallel troughs.

Fig. 3.

61,607 retreat channel roof fracturing drilling layout plan.

The above table shows the parameters of different types of boreholes in the main transport tunnels, auxiliary transport tunnels and withdrawal channels, which are divided into 3 types of boreholes, namely, A, B and C. They also include the fracturing pressure (more than or equal to 30 MPa), fracturing time (more than or equal to 30 min), borehole spacing, borehole diameter, effective spacing, depth of boreholes, inclination of boreholes, orientation of boreholes, as well as the designed number of times and modes of single-hole fracturing (double-end blocking fracturing). (double-end blocking fracturing). In the main and auxiliary transportation tunnels, there were 9 times of A and B type holes and 2 times of C type holes, all of which used double-end plugging and fracturing. In the withdrawal channel, there are 9 times of high and low holes respectively, and all of them adopt double-end plugging and fracturing. This is shown in Table 3 and Figs. 2 and 3.

Table 3.

Hydraulic fracturing parameters of two parallel channels on the 61,607 working face.

Location	Category	Borehole orientation	Single hole fracturing design
Primary Runner	B	Toward the direction of the working face hollow area, the working face side goes at an angle of 45°	33.5/50/54.5/59/69.5/72/86/90/100/104
	C	Toward the face, the face-side strike angle is 75°	36/39.5
Auxiliary Slot	A	Toward the working face, the working face side is oriented at an angle of 0°	33.5/50/54.5/59/69.5/72/86/90/100/104
	B	Toward the working face, the working face side is oriented at an angle of 45°	33.5/50/54.5/59/69.5/72/86/90/100/104
	C	Toward the working face, the working face side is oriented at an angle of 75°	36/39.5
Evacuation Channel	High pitched hole	Toward the working face, the working face side is oriented at an angle of 0°	31/46/50/54/64/79/83/92/96
	Low level hole		33.5/50/54.5/59/69.5/72/86/90/100/104

Numerical simulation

The numerical simulation model is shown in Fig. 4, which simulates the collapse process of the roof slabs of the two trenches and the withdrawal channel of the 61,607 working face in Longwanggou, the Moore-Coulomb model is selected for this structural model, and the size of the numerical simulation model is 300 m × 100 m × 100 m, with a vertical stress of 10 MPa applied to the upper boundary to simulate the loading of the overlying rock layer, and the model is fixed in the horizontal direction. The horizontal boundary is fixed, and the horizontal displacements in the left–right and front–back directions are set to 0 to simulate the boundary conditions. The roadway in the model is designed according to the real value, and the roadway is cut in the direction of the working face, and the angle of the cut is as shown in the previous chapter, and the depth of the cut is 8 m, firstly, the roadway is excavated and the roof is cut in the model, and after calculating and balancing, the working face is mined for 5 m in each step, and after each step of the mining, the working face waits for the same amount of time to carry out mining in the next step. The faults and slits in the model are all realised by the built-in nodal commands of 3DEC software. When setting the parameters of faults in the model, in order to simulate the fault crushing zones, the parameters of cohesion and friction angle are all referenced to the relevant parameters of the crushed rocks in the faults in the engineering site, while the parameters of slits are set to 0 to simulate the effect of cutting the roof plate in the site, and the rest of the parameters are consistent with the rock layer where the slits are located. Table 4 shows the relevant nodal parameters required for numerical simulation.

Fig. 4.

61,607 Before and after fracturing model of the top slab of the pullback channel.

Table 4.

Nodal mechanical parameters.

Lithology	Angle of internal friction(°)	Cohesion (MPa)	Tensile strength (MPa)
Sandy mudstone	16.00	1.30	1.60
Siltstone	18.00	2.80	2.90
Coal	10.00	0.60	0.13
Shale	16.00	1.30	1.60
Siltstone	17.00	2.50	2.70

Stress evolution law of hard top plate

Figures 5 and 6 illustrate the maximum principal stress distribution before and after hydraulic fracturing, revealing significant variations in stress concentration above the mined-out void and the coal pillar. These findings underscore the critical role of hydraulic fracturing in stress redistribution and roof stability enhancement.

Fig. 5.

Stress reduction trend before and after fracture of the roof of two slots 61,607.

Fig. 6.

Stress reduction trend of 61,607 60,607 stoping channel roof before and after rupture.

Prior to fracturing, intense compressive stress accumulates above the coal pillar, reaching a peak of 34.8 MPa, while tensile stress within the mined-out void induces roof bending without immediate fracturing. This stress concentration leads to increased risk of structural instability and potential mining-induced seismic events. Post-fracturing, the stress concentration in the coal pillar is significantly reduced to 19.2 MPa, representing a 44.8% decrease. Simultaneously, the tensile stress in the void drops from 12.7 to 5.8 MPa, marking a 54.3% reduction, which facilitates controlled roof subsidence and mitigates the potential for sudden roof collapses.

From a spatial perspective, the high-stress concentration zone extends 68 m before fracturing. Following hydraulic fracturing, this zone contracts to approximately 20 m, indicating a 70.6% reduction in the stress-affected region. This substantial decrease effectively redistributes the stress, alleviates peak loads, and reduces excessive deformation in the surrounding rock, thereby enhancing overall structural stability.

Numerical simulations further confirm that hydraulic fracturing not only accelerates roof collapse in a controlled manner but also mitigates stress accumulation, thereby preventing the formation of large overhanging structures. This controlled stress release significantly reduces the risk of mining-induced seismicity and enhances safety in underground operations. These findings highlight the efficacy of hydraulic fracturing as a proactive measure for improving stress management and maintaining long-term stability within the mining environment.

Characteristics of subsidence and deformation of hard top plate

In the numerical simulation, horizontal measurement lines were established along the length of the model within the middle and upper sections of the hard roof slab to monitor and analyze the variation in roof subsidence under both conventional mining and hydraulic fracturing pressure relief mining conditions. The results are illustrated in Figs. 7 and 8. The subsidence of the hard roof slab along the model length generally exhibits a symmetrical “concave” distribution. Under conventional mining conditions, the high integrity of the hard roof slab leads to the formation of a large overhanging roof area in the central mining region, spanning approximately 200 m. The subsidence of the roof in the two tunnels measures about 0.93 m, while the retreating passages experience a subsidence of approximately 0.94 m. This extensive overhanging roof area significantly increases the risk of dynamic disasters in the working face.

Fig. 7.

Decreasing trend of settlement deformation before and after hydraulic fracturing of the roof of two parallel grooving on 61,607 working face.

Fig. 8.

Decreasing trend of settlement deformation before and after hydraulic fracturing of the roof of the 61,607 working face pullback channel.

Following the implementation of hydraulic fracturing for pressure relief, the original overhanging roof area and its subsidence pattern were effectively altered. The large overhanging roof transformed into an overall subsidence and rotational subsidence of the roof slab, with the maximum subsidence occurring near the fractured zones. The peak subsidence values reached 1.83 m and 1.95 m, representing increases of 1.96 and 2.07 times, respectively, compared to conventional mining. Specifically, the subsidence near the fractured zones increased by 96.8% (from 0.93 to 1.83 m) and 107.5% (from 0.94 to 1.95 m) for the tunnel and retreating passage roofs, respectively. The application of hydraulic fracturing for pressure relief facilitates a more substantial and timely subsidence of the targeted hard rock layer, thereby mitigating the risk of roof overhang and enhancing the safety and stability of the mining operations.

These results demonstrate that hydraulic fracturing not only reduces the overhanging roof area but also induces a controlled and predictable subsidence pattern, which is critical for minimizing dynamic hazards and ensuring operational safety.

Numerical results of fracture field development

Figures 9 and 10 illustrate the development of roof cracks during the normal mining process of the working face. Figure (a) depicts the crack development in the roof during mining without hydraulic fracturing, while Figure (b) shows the distribution of the roof crack field under mining conditions following hydraulic roof cutting and pressure relief. During the advancement of the working face, the cracks on the roof above the mining area without fracturing are scattered and relatively uniformly distributed across the roof of the mining area. However, the roof fails to fracture in a timely manner, resulting in a significantly larger overhanging roof length.

Fig. 9.

61,607 Fracture distribution before and after fracturing the top plate of the two trenches.

Fig. 10.

Fracture distribution before and after fracturing of the top plate of the 61,607 retreat channel.

In contrast, after hydraulic roof cutting and pressure relief, the roof fractures and collapses promptly, releasing stress and effectively reducing the length of the overhanging roof. As the working face continues to advance, the development of cracks gradually stabilizes. The application of hydraulic roof cutting and pressure relief ensures that the roof collapses in sync with the mining void, preventing the formation of a long-distance overhanging roof. A compacted surface emerges above the void area, and the height of crack development is notably reduced compared to that observed under normal mining conditions. This reduction in crack height and the timely collapse of the roof contribute to enhanced stability and safety in the mining operations.

Industrial tests

Microseismic monitoring system layout program

1. Substation and backbone network design

Two substations and 16 vibration sensors were deployed to monitor the 61,607 working face. Each substation, equipped with eight high-frequency vibration sensors, was installed in one of the two troughs. Substations were positioned at one end of the monitoring area to avoid high-pressure radiation and ensure optimal dry conditions.

• (2)Sensor arrangement design

Vibration sensors were spaced at 200 m intervals, with an additional 150–200 m coverage ahead of the working face in the 61,607 troughs. Eight sensors were installed in each trough, totaling 16 units, forming a spatially distributed monitoring network. Sensor locations were determined based on field surveys, avoiding interference from fans and other vibration sources. When the working face advanced within 150 m of the last sensor, the sensor was relocated 200 m further to ensure continuous monitoring coverage.

Analysis of decompression effect of microseismic event statistics in 61,607 working face

Energy or magnitude is the most commonly used recording index when impact ground pressure accidents occur. Dou²⁷ believe that energy > 10,000 J belongs to high-energy vibration type mine earthquake, there are strong vibration and sound in the underground, and sometimes there is a tremor on the ground, which is easy to induce shock ground pressure, combined with the history of the occurrence of shock ground pressure in this mining area and the monitoring of the high-energy events to make corrections, so the monitoring of microseismic events of 9500 J or more.

1. Analysis of microseismic data before and after depressurization in the depressurized area.

The microseismic monitoring data of KJ768 high-precision microseismic monitoring system in August-October were retrieved, and microseismic events above 9500 J were extracted, and the energy distribution is shown in Figs. 11, 12 and 13:

Fig. 11.

Planar distribution of energy events above 9500 J in August.

Fig. 12.

Planar distribution of energy events above 9500 J in September.

Fig. 13.

Planar distribution of energy events above 9500 J in October.

In August, the main transport chute of the working face advanced 15 m into the roof fracturing zone, with the fracturing borehole extending 186 m ahead of the working face. Simultaneously, the auxiliary transport chute was positioned 128 m from the fracturing borehole, which extended 436 m ahead of the working face. Microseismic events exceeding 9500 J were concentrated in two ranges ahead of the working face: 93–164 m and 236–321 m, spanning a total of 156 m.

In September, the main transport chute of the working face advanced 44 m into the roof fracturing zone, with the fracturing borehole extending 352 m ahead of the working face. Meanwhile, the auxiliary transport chute was positioned 100 m from the fracturing borehole, which extended 449 m ahead of the working face. Microseismic events exceeding 9500 J were concentrated in a single range of 95–158 m ahead of the working face, spanning a total of 63 m.

In October, the main transport chute of the working face advanced 86 m into the roof fracturing zone, with the fracturing borehole extending 502 m ahead of the working face. Concurrently, the auxiliary transport chute was positioned 58 m from the fracturing borehole, which extended 407 m ahead of the working face. Microseismic events exceeding 9500 J were concentrated in a narrow range of 93–106 m ahead of the working face, spanning a total of 13 m.

From the above microseismic event statistics Table 5, energy event planar distribution map, energy event profile distribution map, it can be seen that compared with August, the microseismic event of 9500 J in October is 143 m less than the concentrated range in front of the workplace, a decrease of 92%. Compared with September, the microseismic event of 9500 J in October decreased by 50 m from the front of the work area, which is 79% lower. It can be seen that after the implementation of the pressure relief program, the microseismic events in the main and auxiliary transportation tunnels were reduced, indicating that the implementation of the pressure relief technology has achieved obvious results.

Table 5.

Statistics of energy events above 9500 J and 10000 J on the 61,607 working face from August to October.

Months	Number of microseismic events above 9500	Number of microseismic events above 10,000 J	Monthly progress
August	19	1	61.2
September	9	4	29.9
October	9	1	39.0

Drill hole peeping at the working face

To evaluate the effectiveness of hydraulic fracturing, a borehole panoramic imaging system was employed to observe fracture distribution in the borehole wall post-drilling and fracturing. The pre-designed hydraulic fracturing program for the hard roof of the working face was implemented by drilling and fracturing in two designated slots. After fracturing, the borehole imaging system was used to sequentially assess the fracturing effects of each group of drill holes. Figure 14 presents the observation results for the 23rd group of fracturing drill holes in the auxiliary transportation tunnel, located 635 m from the cut-off eye.

Fig. 14.

Drilling observation results.

According to the analysis of borehole observation data, the fractures induced at various depths after hydraulic fracturing are predominantly radial in nature, with a limited presence of vertical fractures in localized zones. In certain sections, both radial and vertical fractures coexist, indicating the development of horizontally oriented fractures alongside a small proportion of vertical fractures, resulting in a complex fracture network morphology. The comparative analysis confirms the successful initiation and propagation of fractures during staged fracturing operations.

Through the analysis of the above peep results, it is concluded that hydraulic fracturing of the roof plate can produce fissures in the hard rock layer overlying the working face, destroying the integrity of the hard rock layer, and effectively reducing the probability of the mine quake triggered by the overall spanning down of the hard rock layer during the process of mining back in the working face.

Pressure relief mechanisms and factors reducing the effects of high-energy microseismic events

Cantilever beam structure is easy to form after mining in hard roof face of coal mine, which leads to overlaying of dynamic mining pressure and high static pressure caused by overhanging rock strata and transfer to mining roadway through key layers, resulting in instability and large deformation of surrounding rock structure of mining roadway adjacent to the working face, thus preventing high-energy events caused by cantilever beam collapse or disturbance.

Comparison of displacement before and after hydraulic fracturing (Figs. 5 and 6) reveals that, in the unloading zone, high-angle fractures (HFs) in the roof and activated natural fractures (NFs) effectively disrupt the transmission path of supporting pressure. Their activation significantly reduces the maximum principal stress, primarily through shear sliding along the fracture surfaces. This "cut-off" effect is analogous to stress release observed around faults after seismic events. Most HF–NF intersections exhibit shear-sliding failure when stimulated, whereas such failure is limited without stimulation. This confirms that the pressure relief zone is optimally located along the supporting pressure path, promoting shear deformation and reducing load transfer to the inner rock mass.At the same time, this is consistent with the results of the borehole snooping.

In extremely thick coal seams, high in-situ stress accumulates due to elastic energy storage. Continuous mining expands the goaf and intensifies stress disturbances, increasing rock burst risk. Prior to fracturing, the intact roof is prone to sudden rupture, resulting in high-energy microseismic events. As shown in Table 5, the fracture network induced after fracturing weakens the integrity of the top plate, leading to more frequent low and medium energy events and reduced individual energy releases. Combined with Figs. 7 and 8, the observed stress drop in the caving zone is a clear geotechnical precursor of large-scale roof collapse. Hydraulic fracturing thus facilitates timely energy release and mitigates dynamic disturbances, offering an effective means to prevent rock bursts.

Conclusions

1. Stress Reduction and Pressure Path Disruption: Hydraulic fracturing effectively mitigated stress concentrations by disrupting supporting pressure transmission through shear sliding along high-angle fractures (HFs) and activating natural fractures (NFs). This "cut-off" effect, similar to post-seismic stress release near fault zones, reduced peak compressive stress in the coal pillar by 44.8% (from 34.8 to 19.2 MPa) and tensile stress in the void zone by 54.3% (from 12.7 to 5.8 MPa). The resulting stress redistribution alleviated pressure in critical structural areas, creating a low-stress, impact-resistant mining environment.

2. Subsidence Enhancement and Controlled Collapse: Post-fracturing roof subsidence increased by a factor of 1.96–2.07, enabling controlled collapse of overlying strata. This mechanism minimized large overhanging roof sections, improved structural stability, and aligned with observed stress reductions in the caving zone, which are precursors to significant roof collapse.

3. Seismic Hazard Mitigation: The fracture network compromised roof integrity, redirecting energy release from sudden high-energy microseismic events (> 9500 J) to more frequent, low-to-medium energy events. The range of high-energy events was reduced by 50 m, with a 79% decrease in frequency. Hydraulic fracturing enabled timely energy dissipation, mitigating dynamic disturbances and reducing the risk of rock bursts.

4. Fracture Optimization and Shear Activation: Borehole imaging confirmed that hydraulic fracturing expanded pre-existing fractures and induced new ones, especially at the intersections of high-angle fractures (HF) and natural fractures (NF), where shear sliding was predominant. This shear-driven activation significantly reduced load transfer to inner rock masses, thus ensuring safer mining conditions by preventing sudden roof ruptures.

Author contributions

Z.L. was responsible for revising the thesis, D.Z. was responsible for writing the thesis,L.G. was responsible for industrial experiments,L.G. was responsible for drawing pictures, and L.X. was responsible for revising the format.

Data availability

All data generated or analysed during this study are included in this published article.

Declarations

Competing interests

The authors declare no competing interests.