Experimental research on the instability characteristics of the overlying strata structure that characterizes shallow interval goaf mining

Abstract

Here, we analyze the instability characteristics of the overlying strata structure that characterize shallow-depth seams under insufficient goaf in Yushenfu mining area. To simulate the No. 20107 longwall interval working face situated in Nanliang Coal Mine, physical simulation, an acoustic emission (AE) monitoring system, a stress acquisition system and a total station were used. The results indicate that during interval goaf formation, which is correlated with mining, the immediate roof collapses, and the main roof strata remains stable. Gradually, the stress that acts on the temporary coal pillar (TCP) gradually exhibits the ‘uniform increase–accelerated increase catastrophe instability’ change characteristics. Due to the concentrated load of the overlying strata, the bearing capacity of the TCP gradually deteriorates until the catastrophic instability occurs, and the unstable roof strata forms a ‘W-shaped voussoir beam’ structure. The research results provide evidence for the strata control that is associated with shallow-seam mining.

1. Introduction

The coal resources that characterize the Jurassic coalfield, which is located in north Shaanxi, are characterized by shallow burial, thin bedrocks, numerous recoverable seam layers and large reserves. With regard to the initial mining area development stages, we observed that outdated production support equipment was being used: the use of a single pillar cannot limit the immense mine pressure, which characterizes the longwall mining process that is used in shallow, buried coal seams (Shi & Hou 1996; Huang 2002). Most small and medium-sized mines that utilize upper coal seam mining implement traditional pillar mining (e.g. room and pillar mining, room mining, Wongawilli mining, and gateway and pillar mining). Mines such as the Changlebao Coal Mine (Sha et al.2017) and Jinniu Coal Mine (Xue 2022) perform knife pillar mining; the Yubujie Coal Mine (Shao et al.2009) and Xuemiaotan Coal Mine perform narrow strip water preserved mining and the Nanliang Coal Mine (Hou & Fu 2009) performs the longwall interval mining method that is based on knife pillar mining, and has been popularized in many neighboring mines. The use of different mining methods leads to the formation of different insufficient goaf overburden structures.

Scholars have effectively analyzed this research topic. Meng et al. (2007) proposed a calculation model that combined a beamless floor system with a column cap; thus, the failure and instability mechanism of the overlying rock that characterizes room and pillar goaf was analyzed. It is was observed that the failure that affects the room mining pillar yields a chain failure. Xie (2014), who investigated the stress–strength relationship of the pillars that characterize room-pillar goaf, constructed a mathematic model; thus, they analyzed roof-pillars that characterize shallow-seam room-pillar goaf, and they determined the instability process of the coal-pillar group, which induced large-area roof disaster under rheological action. Wang et al. (2008) constructed the mechanical model pertaining to the elastic foundation plate of the strip goaf roof, and they analyzed the breaking mode and sudden instability mechanical process that affects the goaf roof in different stages. To analyze the stress characteristics and distribution law of the coal rock that characterizes the pillar goaf, Tu et al. (2013) combined the preceding method with the elastic half-plane theory. To construct a mechanical model that can analyze the pillar-roof that characterizes pillar goafs, Sun et al. (2017) used the rheological properties of the coal pillar and the minimum potential energy principle, and they obtained the calculation formula that estimates the effective bearing time of the goaf coal pillar. To analyze the stability of room pillars as well as pillar mining goaf and interval goaf, Zhang & Wang (2020) and Zhang et al. (2017) utilized the safety factor method. Yu et al. (2017) established a non-uniform stripping model that analyzes coal pillars, and based on the progressive stripping behavior of the coal pillar and the stacking characteristics of the stripping body, they proposed a long-term stability evaluation method that considers strip coal pillars. To construct a time-dependent stability analysis model that considers coal pillars, Xu et al. (2005) used the minimum potential energy principle and, based on catastrophe theory, they considered roof stiffness and softening as well as the rheological properties of the coal pillar. Wang et al. (2012) constructed the cusp catastrophe model that analyzes room coal-pillar stability, and they considered the catastrophe instability law pertaining to the residual pillar that characterizes shallow-seam room mining. To evaluate the probability density function that affects room coal-pillar instability, Zhu et al. (2018) proposed a calculation method and used a renormalization group method. Zhang et al. (2022) and Zhang et al. (2019) summarized the progress, effect and limitation of the evaluation method pertaining to coal-pillar instability. Robert et al. (2016) used numerical simulation to analyze the stability of coal pillars in Australia, South Africa, the USA and India, and they developed an extended damage initiation limit method. Ebrahim et al. (2017) investigated the long-term stability of an abandoned room and pillar coal mine, and they indicated that pillar deterioration significantly affects the long-term stability of mining excavations. However, regarding shallow interval mining, the instability characteristics of the overlaying structure, which is characterized by insufficient mining, are rarely analyzed.

Regarding the research object, this study utilizes the overlying structure of the interval goaf that characterizes Nanliang Coal Mine. Based on a rock mechanical parameters test, a physical similarity model is constructed; thus, the stability of the pillars that characterize interval goaf is analyzed. To analyze the stress variation law and the collapse instability characteristics that characterize the interval goaf, we combined the stress sensor and AE monitoring system. To verify the stress distribution law and the interval goaf fracture, which provides a reference for the strata control that characterizes the shallow insufficient mining area, the UDEC program was used.

2. Working face conditions

This study considers the No. 20107 interval goaf that characterizes Seam 2–2, which is located in Nanliang Coal Mine, Yushenfu Mining Area. The No. 20107 longwall working (LW) face is located on the west side of Panel 201. With regard to Seam 2–2, the depth is set at 92 m, the longwall face dip length is set at 180 m, the strike length is set at 1650 m, the mining height is set at 2 m and the seam angle is set at 0–3°. The support entails a DVJ-22 hydraulic prop that comprises a I-IDJA.1200 metal hinged roof beam, and a 1.0 × 0.6 m row spacing is used. The maximum roof control distance is set at 4.0 m. The main roof comprises an 8.4-m thick siltstone, the immediate roof comprises a 0.5–3.7-m siltstone, the immediate bottom comprises a 1-m thick mudstone and the basic floor strata comprises a 9-m siltstone. We observed that the actual size of the interval coal pillar (ICP) that characterizes the No. 20 107 LW face is 6–23.8 m. Figure 1 depicts the distribution diagram of the ICP that characterizes the No. 20 107 LW face.

3. Physical simulation analysis of the overburden caving that characterizes interval mining

3.1. Physical similarity simulation experiment design

3.1.1. Coal rock mechanical parameters test

Using a core drilling machine, the rock samples are processed into standard specimens, and using a WE-60 universal material testing machine, the mechanical parameters are tested. Tables 1 and 2 depict the test results pertaining to the uniaxial compressive strength and tensile strength of the overlying rock.

3.1.2. Determination of physical model similarity constants

To accurately reflect the actual caving characteristics of the overlying strata, the model laying process should match the actual in-situ working conditions; with regard to the model and prototype, the similar conditions (e.g. the geometric conditions, initial conditions and boundary conditions) should be considered.

Based on the similarity principle and experimental requirements (Li 1998), the similarity constants are obtained as in Table 3, where the parameter subscript p denotes the prototype, whereas m denotes the model:

3.1.3. Physical simulation scheme

We analyze the actual geological conditions pertaining to the working face design model that simulates the pavement derived from the Nanliang Coal Mine No. 20 107; furthermore, we use the interval mining method. With respect to Seam 2–2, the buried depth is 92 m, the bedrock thickness is 25 m, the weathered rock is 8.5 m, the loose layer thickness is 67 m and the coal seam thickness is 2 m. Figure 2 depicts the model design. The model selects a 3000 × 200 × 2000 mm (length × width × height) plane physical similarity simulation experiment platform, and it combines the actual mining conditions of the interval goaf. The simulation material is composed of an aggregate, which comprises river sand, quartz sand and loess; a cementing agent, which comprises gypsum and calcium carbonate, and water. To divide the weak structural planes (e.g. bedding and joints), the 8–10 mesh mica powder is used. The model’s actual laying height is 113 cm. Table 4 depicts the material ratio.

3.1.4. Monitoring methods

• Stope pressure and working face support resistance monitoring

  The micropressure sensor is laid on the floor of Seam 2–2; thus, the change pertaining to the overburden stress, which characterizes mining, is monitored. To monitor the abutment pressure that affects the surrounding rock before and after mining, the 128-channel pressure-data automatic acquisition system is used. We monitor the variation characteristics pertaining to the pillar stress, which affects the goaf after interval mining. The simulated excavation step distance is 4 m, and the monitoring system is arranged as per figure 2.

• Working face roof damage fracture AE characteristic signal monitoring

  To monitor the elastic strain energy release and the event count signals that characterize the micro-damage fracture process that affects the overburden, the AE monitoring system is used. By analyzing the number of micro-damage fractures that affect the overlying strata and the energy release amplitude, we can predict and explain the macroscopic fracture mechanism that characterizes the rock strata. The AE measurement points are arranged as in figure 2, and five AE signal sensors are arranged. The 1#, 2#, 4# and 5# sensors are placed 10 cm away from the boundary sides; the 1# and 4# sensors are arranged 30 cm from the top boundary and the 2# and 5# sensors are arranged 10 cm from the bottom of Seam 2–2. The 3# sensor is placed in the middle of the model, and it is situated 30 cm above the ICP.

• Model surface displacement monitoring

  With respect to the surface of the model, two measuring lines pertaining to roof strata movement are arranged at 13 and 25 m from the upper surface of Seam 2–2. The measuring points are spaced at 10 m, and the bottom–top distribution is denoted as C and D. To monitor the movement of the strata, which characterizes the mining process, the total station is used.

3.2. Physical simulation results

3.2.1. Stability analysis of the interval coal pillar

Using the interval mining method, Seam 2–2 is successively mined. Figure 3 depicts the initial roof caving characteristics of the interval goaf. The boundary coal pillar is set at 36.5 m. To form a first MS (mining strip), 50 m is excavated and a 6-m TCP (temporary coal pillar) width is set; subsequently, to form a second MS with the TCP as the center, 50 m is excavate and a 15-m ICP width is set. It is worth noting that two adjacent MS constitute a mining section. To form the third and fourth MS, we repeat this process. Finally, the initial interval goaf overlying space structure is formed, and the immediate roof caving and main roof strata remain stable. With regard to the face propulsion speed, the cycle footage is 0.8 m, the working face completes five cycles each day and two classes of coal mining each day, the actual production time is 16 h and the MS completion time is 12.5 d. In combination with the time similarity constant, the single excavation step of the model is 4 m, the interval of each step is 1.6 h and the model completes a MS every 31.25 h.

Figure 4 depicts the variation curve pertaining to the abutment stress, and it considers different Seam 2–2 mining durations. After the first MS is mined, the immediate roof collapses, and the main roof remains stable. The influence zone of the advance abutment pressure exists 17.5 m in front of the LW face. The 6-m TCP is retained and, thus, the second MS can be continually mined. The 1# TCP stress attains 5.86 MPa. Compared with the first MS, when the second MS is mined, the caving characteristics of the overlying strata remain constant, and the 1# TCP stress increases to 7.82 MPa. After setting the ICP to 15 m, the 1# TCP stress increases to 9.94 MPa when the third MS is mined. The ICP stress exhibits a saddle shape distribution, and the maximum ICP stress attains 6.95 MPa. When the 2# TCP was assembled, the TCP stress attained 5.78 MPa. After the fourth MS was excavated, the immediate roof collapsed; by contrast, the main roof remained stable, and the 1# TCP stress attained 12.78 MPa. The pressure pertaining to the ICP showed an asymmetric saddle-shaped distribution (i.e. a high left side and low right side). The maximum ICP stress attained 10.88 MPa, whereas the 2# TCP stress attained 7.58 MPa.

3.2.2. Caving characteristics of the overlying strata that characterizes interval mining

• Caving law of overlying strata after the 1# TCP instability

  Seventy-two hours after the fourth MS is mined, the 1# TCP is destroyed, and this destruction is caused by excessive loading. Figure 5 indicates that the 1# TCP was subjected to shear and tensile failure under the action of overburden load. The coal blocks that are adjacent to the coal-pillar collapse, and the main roof of the first and second MS rotates and becomes disjointed; thus, a ‘W-shaped voussoir beam’ hinged structure is formed. The caving height of the overburden rock that characterizes the first MS is 19.5 m, the development height of the separation layer is 39 m and the fracture span 27 m. The caving height of the overburden rock that characterizes the second MS is 17 m, the development height of the overburden separation layer was 40 m and the separation span was 27 m. The length of the fracture that the overlying soil layer exerted on the 1# TCP increased to 56 m, and it formed a self-stable fracture arch.

  After the instability of the 1# TCP, the measuring points C1–C15 exhibit apparent displacement changes. The subsidence curve of the displacement measuring point is depicted in figure 6. The subsidence of the overlying strata that is above the 1# TCP attains 0.3 m. With regard to the first and second MS, the subsidence of the overlying strata attains 0.93 and 0.84 m, respectively, and the fracture development height attains 40 m.

• Caving law of the overlying strata after the instability of the 2# TCP

  When the model had continually stood for 56 hours, the 2# TCP was also subjected to cracking damage, which was caused by excessive loading. Figure 7 shows that the 2# TCP was subjected to shear failure, which was caused by the overburden load, and that the coal blocks that were adjacent to the coal pillar collapsed. Subsequently, the main roofs of the third and fourth MS broke and formed a W-shaped voussoir beam-hinged structure. With regard to the third MS, the caving height of the overlying strata was 21.5 m, the development height of the separation layer was 42 m and the fracture span was 32 m. In regard to the fourth MS, the caving height of the overburden rock was 22 m, the development height of the overburden separation layer was 44 m and the separation layer span was 32 m. The overlying strata that exhibited the main roof fracture gradually collapsed to form a ladder-shaped caving structure, whereas the length of the 2# TCP overburden cracks increases to 54 m, and it forms a self-stabilizing fracture arch.

After the instability of the 2# TCP, the C15–C29 measuring points that characterize the C line exhibit apparent displacement changes. Figure 8 illustrates the subsidence curve of the displacement measuring point. In regard to the TCP, the subsidence of the overburden attains 0.6 m, in regard to the first MS, the subsidence of line C attains 1.56 m, whereas in regard to the second MS, it attains 1.35 m. Moreover, the fracture development height attains 54 m.

3.2.3. Interval mining coal-pillar stress variation characteristics

• Vertical TCP stress variation

  Using the real-time monitoring data pertaining to the Seam 2–2 floor stress sensor, which measures the stress that acts on the goaf coal pillar after interval mining, the variation curve that temporally considers the TCP stress is shown in figure 9. After the completion of the fourth MS, the 2–2 coal seam pressure sensor is continuously monitored and the vertical stress pertaining to the two TCPs gradually increases. After 72 hours of monitoring, the vertical stress that acts on the 1# TCP attains 21.37 MPa and the vertical stress that acts on the 2# TCP attains 13.24 MPa. Within a short duration, the vertical stress decreased drastically (5.74 MPa), and the main roof strata fractured above the 1# TCP. Subsequently, secondary fractures occurred in the main roof strata (first mining zone fracture depth 22 m; second mining zone fracture depth 27m). The analysis indicates that due to the small size of the TCP, when the mining range continually increases, the stress that is transmitted from the overlying strata load to the TCP gradually increases. Due to the concentrated stress, the transient release of the elastic energy that accumulates inside the 1# TCP occurs; furthermore, the TCP collapses and the stress decreases. After 128 h of continuous monitoring, the vertical stress of 2# TCP attained 21.81 MPa and caving instability occurred. The coal-pillar stress decreased to 6.14 MPa. The main roof strata fractured above the 2# TCP, and secondary fractures occurred at the 26.5 m level in the third MS and at the 22 m level in the fourth MS.

• Vertical stress variation law of the ICP

  Figure 10 depicts the variation curve pertaining to the vertical stress that acts on the ICP during TCP instability. After the longwall face mining that characterizes the third MS, the ICP pressure exhibits a saddle-shaped distribution and the peak stress of the coal pillar attains 6.95 MPa. After the fourth MS is completed, the ICP pressure increases again, which indicates an asymmetric saddle-shaped distribution that exhibits a high left side and low right side, and the peak stress of the coal pillar attains 10.88 MPa. After the instability of the 1# TCP, owing to the collapse of the roof strata, the vertical stress that affects the ICP increases in a unified manner. The stress that acts on the left side of the ICP increases more apparently than the stress that acts on the right side, and the peak stress attains 12.5 MPa. The ICP stress generally exhibits an asymmetric saddle-shaped stress distribution that is characterized by a high left side and a low right side. After 56 h of continuous monitoring, the 2# TCP became unstable and the roof strata collapsed in a wide range; furthermore, the vertical stress of the ICP increased again. The stress that acts on the right side of the coal pillar increased more significantly than that on the left side, and the peak stress attained 13.5 MPa. The stress that acts on the ICP generally exhibits symmetrical saddle-shaped distribution characteristics.

  Figure 11 depicts the vertical stress variation characteristics of the rock that surrounds the goaf, and it considers the period in which the fourth MS of the 2–2 coal seam is mined. After the fourth MS is mined, the 1# TCP stress attains 12.78 MPa, and the pressure that acts on the ICP exhibits an asymmetric saddle-shaped distribution (curve shape: high left side and low right side). The peak ICP stress attains 10.88 MPa, and the 2# TCP stress attains 7.58 MPa. After the model was statically monitored for 72 h, the 1# TCP was broken and the soil layer fractures increased (56 m). The 1# TCP stress decreases to 5.74 MPa. After 56 h, the 2# TCP is broken, the soil layer fracture increases (54 m) and the 2# TCP stress is reduced to 5.73 MPa (Zhang & Wang 2021).

3.2.4. Interval mining overlying caving AE monitoring

To monitor the acoustic signal pertaining to the fracture process and the instability of the overlying strata, which characterize the process that is associated with the interval simulation experiment, the acoustic emission (AE) was utilized (Lai et al.2014). The AE signal data pertaining to the roof strata fracture that characterizes the early-stage sudden instability of the 1# and 2# TCPs were obtained. The variation characteristics of the AE signal characteristic parameters such as the AE energy and ring count, which characterize the process pertaining to roof strata fracture and instability were analyzed.

• AE characteristics of the 1# TCP instability overlying caving

  Figure 12 illustrates the signal characteristics of the 2# AE sensor when the roof stratum breaks during the sudden instability of the 1# TCP. The analysis considers the period after which the fourth MS is mined. During the monitoring process, the vertical stress of the 1# TCP is increasing and debris is falling from the coal-pillar surface, which consists of fragmented coal. The AE energy and the corresponding ringing count accumulate continuously, which indicates that during the stress redistribution process that affects the overlying strata, the elastic strain energy that characterizes the 1# TCP accumulates continuously and the internal cracks that characterize the TCP and roof rock stratum are also developing continuously. During the 11–39 s period, the AE energy signal was suddenly released in large quantities; the change fluctuation was apparent; the ringing count changed drastically; the total number of events and signal intensity were large and the maximum AE energy attained 9273 ms·mV⁻¹, whereas the ringing count attained 1099 times per second. Consequently, AE energy and ringing count tend to be stable, which indicates that due to the sudden instability of the 1# TCP, the roof stratum disintegrates and loses stability within a short duration. Although the internal part of the model remains damaged after the fracture and instability of the overlying strata, the overall trend declines, the energy and ringing count fluctuate and the surrounding rock remains fundamentally stable.

• AE characteristics of the 2# TCP instability overlying caving

  Figure 13 depicts the signal characteristics of the 4# AE sensor when the roof rock layer disintegrates during the sudden instability of 2# TCP. The analysis indicates that after the instability of the 1# TCP, the 2# TCP is continuously monitored in real time. In regard to the initial monitoring stage, the stress that affects the 2# TCP and roof is gradually redistributed, and the AE sensor signal is not apparent. Due to the increase that affects the vertical stress pertaining to the 2# TCP, the surface of the coal pillar is characterized by fallen debris, the AE energy signal and the corresponding ringing count accumulate, which indicates that during the stress redistribution process, which affects the overlying strata, the elastic strain energy inside the 2# TCP is constantly accumulating and the internal cracks that characterize the TCP and the roof strata are also constantly developing. From 57 to 120 s, the AE energy signal was suddenly released in large quantities, apparent fluctuations occurred, the ringing count changed drastically, the total number of events and signal strength became larger and the AE energy and ringing count maximum value attained 11 428 ms·mV⁻¹, 982times/s. Subsequently, the AE energy and ringing count tended to be stable, which indicates that due to the sudden instability of the 2# TCP, the roof strata also experienced fracture instability within a short duration. When the overlying strata was subjected to fracture instability, although damage continuation still affected the model, an overall downward trend was observed, the energy and ringing count fluctuated, and the surrounding rock remained fundamentally stable.

4. Numerical simulation analysis of the overlying strata caving that characterizes interval mining

To effectively analyze the discontinuous fracture and caving phenomenon of the mining rock mass, we utilized UDEC (2013b); thus, we simulated and analyzed the instability characteristics of the strata after shallow interval mining, and we simulated the evolution law pertaining to the overlying rock stress and fracture that characterizes the interval goaf at different mining periods.

If we consider working face No. 20107, we used UDEC discrete element software; thus, we established the same numerical calculation model as the actual mining scenario. The simulated thickness of Seam 2–2 is 2 m, and it shows a 67-m loose layer thickness. Table 5 shows the model parameters that are adjusted during the laboratory test (Wang et al.2003; Chen et al.2006; Jaiswal & Shrivastva 2009). To analyze the block, the Moore–Coulomb criterion is used, whereas to analyze the joint surface, the Coulomb slip failure criterion is used. The model size is 260 × 108 m (length by height). The upper surface is free, whereas the sides and the bottom are fixed boundary conditions and thus, the horizontal and vertical movement are limited.

Figure 14 depicts the evolution law of the vertical stress that characterizes interval mining, and it considers the No. 20107 L W face. In regard to figure 14a, after the first MS is mined, a stress release zone is formed inside the surrounding strata of the longwall face, and this stress release zone, which develops along the coal wall to the middle of the goaf, forms an arched shape. The abutment pressure influence range is 3–14m , and the peak stress attains 3.8 MPa. After the second MS is mined (figure 14b), a new stress release zone is formed on the longwall face. The strata load is gradually transferred onto the 1# TCP; thus, stress concentration occurs and the concentrated stress attains 4.5 MPa. After the third MS is mined (figure 14c), the vertical stress above the ICP presents an asymmetric saddle-shaped distribution and the concentrated stress of the 1# TCP attains 5.5 MPa. After the fourth MS is mined, the concentrated stress of 1# TCP attains 6 MPa and the concentrated stress of the 2# TCP attains 5 MPa (figure 14d). Gradually, the calculation steps increase and the concentrated stress on the 1# TCP increases continuously. When the model is calculated, the 1# TCP suddenly forms a cave-like structure and the main roof layer undergoes rotational instability and fracture. The main roof collapses to form a stable W-shaped voussoir beam structure, and the hinged joint of the main roof broken rock block is subjected to local stress concentration (figure 14e). The model runs until the instability of the 2# TCP occurs. The failure characteristics of the roof are similar to the collapse process of the strata, which occurs after the 1# TCP instability. Due to instability of the 2# TCP, the main roof rotates and breaks and, owing to the collapse, a stable W-shaped voussoir beam structure is formed. Furthermore, the stress release zone is formed in the mining section (figure 14f). The model continues to run, and the overlying strata attains a new stress equilibrium state. The segments that constitute the broken main roof of the interval goaf are hinged to each other and thus the W-shaped voussoir beam structure is formed. The ICP represents the constraint body to cooperate with each other; thus, the bearing of the overlying strata is formed. A local stress release zone that is adjacent to the fractured rock masses is formed in the lower part of the goaf (figure 14g).

5. Discussion and conclusions

The instability of the overlying strata situated in the insufficient mining area of an upper coal seam is sudden, and the failure of the overlying strata that characterizes goaf crucially influences the control of strata in lower coal seam mining. Herein, physical similarity simulation and numerical simulation are used; thus, the stress distribution law pertaining to the overburden instability process that characterizes the interval goaf of shallow coal seams is analyzed from the macro and micro perspectives. After interval goaf formation, the vertical TCP stress attains 21.37 MPa, and the TCP fails; thus, the fracture instability of the overlying strata situated in the interval goaf is induced, and a ‘W-shaped masonry beam’ structure is formed. The vertical stress distribution that characterizes the TCP exhibits ‘uniform increase-accelerating increase-mutation instability’ variation characteristics, and the ICP remains stable. Therefore, when mining the lower coal seam, the working face should pass through the isolated coal pillar, which represents the key area; thus, the occurrence of dynamic pressure can be prevented. Based on the preceding analyses, the following conclusions were obtained:

• The initial formation of interval goaf is characterized by the following phenomenon: when mining is performed, the immediate roof collapses and the main roof strata remain stable. The ICP stress exhibits an asymmetric saddle shape, and the TCP stress exhibits a single hump shape. Gradually, the vertical stress that acts on the TCP exhibits the ‘uniform increase-acceleration increase-abrupt instability’ change rule.

• Under the long-term action of the concentrated load, the bearing capacity of the TCP gradually deteriorates and sudden instability occurs, which induces main roof fracture and instability; thus, a W-shaped voussoir beam structure is formed. After the fracture and instability that affect the overlying strata of the mining section and the collapse arch of the soil layer are superimposed, a trapezoidal-semicircular arch fracture zone is formed.

• When the interval goaf TCP loses stability, the AE energy signal is suddenly released in large quantities, the change fluctuation is apparent, the ringing count changes drastically and the total number of events as well as the signal intensity are great; furthermore, when the coal pillar is abruptly destabilized, the signal attains the peak value and, due to coal-pillar destabilization, it decreases linearly.

• To simulate the instability and stress distribution characteristics of the overlying strata, which characterize the interval mining process, the UDEC program was used and the formation of the W-type masonry beam structure, which characterizes the interval goaf at different mining stages, was evolved; thus, the rationality of the physical simulation was verified.

Acknowledgements

The National Natural Science Foundation of China (Grant No. 51774229) and the Innovation Capacity Support Program (Science and Technology Innovation Team) of Shaanxi Province (No. 2018TD-038).

Conflict of interest statement. The authors declare no conflict of interest.

References