Introduction

With the development of deep mining continuing to increase at a rate of 10 ~ 20 m per year, the effect of ground stress of main coal seam is increasing1. Deep coal seam is more susceptible to negative affects by high stress and low-permeability, and the phenomenon of stress concentration can also lead to significant changes in the characteristics of coal cracks development2, thus affecting the flow capacity and extraction effect of gas in coal seam3. Therefore, it is necessary to further analyze the influence of stress on the characteristics of coal cracks development on the basis of studying the law of coal stress evolution. And it is particularly important to put forward the corresponding stress unloading measures and coal and gas outburst prevention technology on this basis.

Many scholars had analyzed the influence of stress evolution on characteristics of coal crack development from different angles. The main research could be divided into the following aspects: (1) Theoretical analysis: The research showed that the change of characteristics of coal crack development reflects the process of energy conversion. The input mechanical energy was continuously transformed into the elastic deformation energy accumulated in coal rock mass under the external loading. When the elastic deformation energy was stored to a certain limit, that is, when it exceeded the ultimate load capacity of coal rock mass, micro-crack initiation, crack penetration would occur gradually in coal rock mass4,5,6. (2) Experiment research: Some scholars had carried out uniaxial, triaxial and impact loading tests on raw coal, and found that the cracks, porosity7,8,9 and related permeability10 of raw coal will change. (3) Numerical simulation analysis: Some scholars had carried out numerical simulation research on the deformation, stress field and crack evolution of coal rock mass under different loading methods11,12. A large number of studies had shown that stress variation has a significant effect on the crack evolution of coal rock mass.

The cracks distributed in coal rock mass had significant effects on its gas permeability13 and permeability14. To a certain extent, the cracks development was related to the characteristics of gas and liquid seepage directly15. With the rapid development of monitoring technology for coal rock mechanics characteristics, CT scanning technology provides an effective means for observing the spatial distribution of internal fractures in coal and studying the real-time fracture development characteristics under loading conditions. Some scholars at home and abroad have used CT scanning systems to observe the initiation, propagation, and failure evolution of coal and rock masses under different loads such as uniaxial and triaxial, revealing the influence of permeability on gas drainage efficiency16,17.

In view of the need for pressure relief antireflection of coal rock mass in the compaction state due to stress concentration in the field, many scholars had applied different experimental methods. For example, methods such as mining protective layer18,19, waterflood softening20, presplitting blasting21, hydraulic fracturing22,23, ultra-high pressure hydraulic cutting24 and large diameter drilling pole25 were adopted to transform the whole coal rock mass. And then change the overall stress state of the coal rock mass by experimental means, thus promoting the growth and development of cracks, and increasing the gas permeability or permeability of coal rock mass.

Stress had an important effect on the crack development of coal rock mass, and the gas permeability and permeability of coal rock mass were often correlated with the crack development degree positively. Based on this, previous research results had been abundant, but most of the relevant studies focused on the influence of stress on the final result of crack evolution or the application of the final result, while few studies focused on the law od crack evolution during the loading process of coal rock mass, especially in the process of triaxial loading. In addition, fewer studies focused on the effect of stress induced crack development on gas permeability and extraction in field applications. Therefore, the stress state in the rock pillar region was analyzed by theoretical and numerical simulation methods, and the triaxial-CT scanning experiments were conducted on coal rock mass according to the change law of the stress state to explore the evolution law of cracks with stress changed. The pumping conditions affected by stress difference under different working conditions were verified and compared. And in view of the prominent danger of the working face increases caused by stress concentration in the rock pillar area, pressure relief was carried out by changing the stress distribution. The results could be used as reference for coal mine gas control under local stress concentration.

Theoretical analysis of static stress of underlie coal mass under the rock pillar affected area

The rock pillar left in the protective layer needed to withstand the lateral abutment pressure and self-weight transmitted from the nearby goaf, and the retaining width of the rock pillar was smaller than the width of the working face. The stress distribution on the surface of rock pillar could be simplified as uniform load. The stress value of any point in the semi planar body could be calculated from the part of the semi planar body subject to the normal distribution force on the boundary according to the elasticity theory26. Thus, the stress exerted by the rock pillar on each point above the protective coal layer could also be obtained.

As shown in Fig. 1, take the midpoint below the coal pillar as the origin of coordinate, the vertical direction as the X-axis, the horizontal direction as the Y-axis, establishing a rectangular coordinate system. Set any point in the semi planar body to be M (x, y), and the stress generated by the normal distribution force q at point M was:

Figure 1
figure 1

Stress transfer diagram of the retaining rock pillar.

Full size image
$$\left. \begin{gathered} {\sigma _x}=\frac{2}{\pi }\int_{{ - a}}^{a} {\frac{{q{x^3}}}{{{{\left[ {{x^2}+{{(y - \xi )}^2}} \right]}^2}}}} d\xi \hfill \\ {\sigma _y}=\frac{2}{\pi }\int_{{ - a}}^{a} {\frac{{qx{{(y - \xi )}^2}}}{{{{\left[ {{x^2}+{{(y - \xi )}^2}} \right]}^2}}}} d\xi \hfill \\ {\tau _{xy}}=\frac{2}{\pi }\int_{{ - a}}^{a} {\frac{{q{x^2}(y - \xi )}}{{{{\left[ {{x^2}+{{(y - \xi )}^2}} \right]}^2}}}} d\xi \hfill \\ \end{gathered} \right\}$$
(1)

Where x was the distance of any point from the X-axis below the rock pillar, and y was the distance of any point from the Y-axis below the rock pillar. a was the distance from the midpoint to both sides of the coal pillar in unit m, ξ was the length of any point from point O in the interval [-a, a], dξ was the minimum length, σx was the stress component in the X direction, σy was the stress component in the Y direction, and τxy was the stress component in the X-Y plane.

The integration of the above formular could be obtained as follows:

$$\left. \begin{gathered} {\sigma _x}= - \frac{q}{\pi }\left[ {\arctan \frac{{y - a}}{x} - \arctan \frac{{y+a}}{x}+\frac{{x(y - a)}}{{{x^2}+{{(y - a)}^2}}} - \left. {\frac{{x(y+a)}}{{{x^2}+{{(y+a)}^2}}}} \right]} \right. \hfill \\ {\sigma _y}= - \frac{q}{\pi }\left[ {\arctan \frac{{y - a}}{x} - \arctan \frac{{y+a}}{x}+\frac{{x(y+a)}}{{{x^2}+{{(y+a)}^2}}} - \left. {\frac{{x(y - a)}}{{{x^2}+{{(y - a)}^2}}}} \right]} \right. \hfill \\ {\tau _{xy}}= - \frac{{q{x^2}}}{\pi }\left[ {\frac{1}{{{x^2}+{{(y+a)}^2}}} - \left. {\frac{1}{{{x^2}+{{(y - a)}^2}}}} \right]} \right. \hfill \\ \end{gathered} \right\}$$
(2)

Where x and y were the horizontal axis and vertical axis of the rectangular coordinate system respectively, a was the distance from the midpoint to both sides of the coal pillar in unit m, L was the influence range of coal pillar on the protected layer in unit m, q was the normal distribution force acting on the retaining rock pillar in unit MPa, H was the distance between the upper protective layer and the protected layer in unit m, and γ was mean weight of strata in unit t/m3.

When the stress was transferred to the coal body at a certain load on rock pillar, that is, x = H, the horizontal coordinate y determined the size of σx. At the same time, when the self-weight of coal and rock between the upper protective layer and the protected layer was considered, that is, y = 0, the self-weight stress of coal and rock was γH, and σx reached its maximum value.

$${\sigma _x}_{{\hbox{max} }}=\frac{{2q}}{\pi }\left( {\arctan \frac{a}{H}+\frac{{Ha}}{{{H^2}+{a^2}}}} \right)+\gamma H$$
(3)

Where a was the distance from the midpoint to both sides of the coal pillar in unit m, q was the normal distribution force acting on the retaining rock pillar in unit MPa, H was the distance between the upper protective layer and the protected layer in unit m, and γ was mean weight of strata in unit t/m3.

As shown in Fig. 2, it could be seen that with the increase of H, the coal stress shows a decreasing trend, that is, σxmax shows a decreasing trend from the top to the bottom of coal seam. Under a certain condition of q, the σxmax value tended to increase with the increase of coal pillar width.

Figure 2
figure 2

The variation curve of coal seam stress values under different rock pillar widths.

Full size image

Taking the 16,219 working face of Jinhe Coal Mine as an example, the buried depth of the protective layer covered on this working face was about 630 m, and the distance between the bottom plate and the top plate of the protective layer H was 20 m. The width of rock pillar was 6 m, so a was 3 m. The mean weight of strata γ was 2.5 t/m3. The stress concentration coefficient above the rock pillar was 3 ~ 3.5, and the estimated q value was 48 MPa. Put the above values into Eq. (3), and the σxmax value transmitted by rock pillar to underlie coal seams was 9.09 MPa.

Test scheme

Profile of test site

Jinhe Coal Mine was located 116 km northwest of Lanzhou City, Gansu Province and belonged to Yaojie Coal and Electricity Group Corporation. The main coal mining in Jinhe Coal Mine was two-layer coal with an average thickness of 22.9 m, which was a near-horizontal ultra-thick single coal seam mining. The maximum gas content (mainly carbon dioxide) was more than 20 cm3/g·r. It has the characteristics of coal and gas (carbon dioxide) outburst and rock burst combined dynamic disaster. The primary gas components of the two-layer coal seam were mainly carbon dioxide, followed by methane, of which carbon dioxide gas components account for more than 80%. The content of carbon dioxide gas, adsorption constant (a value) and Δp were 5 ~ 8 times, 2 ~ 3 times and 2.4 ~ 3.8 times of methane gas, respectively. Full-mechanized caving mining was in two layers. And the regional outburst prevention measures were carried out by exploiting the overlying oil shale protective layer and combing with pressure relief gas (carbon dioxide) extraction.

Figure 3 was the layout plan of the protective layer and the protected layer, and Fig. 4 was the profile diagram of the influence area of the retaining rock pillar in the protective layer (I-I profile). As shown in Figs. 3 and 4, oil shale was used as the upper protective layer on the 16,219 working face. The average layer spacing from the underlie ultra-thick coal seams was 20 m, the thickness of the protective layer was 3 m, and the underlie ultra-thick coal seams was divided into two full-mechanized caving mining layers. Due to the mining deployment, the rock pillar (which was 6 m wide) in the middle of the two protective layers was located in the middle area of the underlie working face. When the combined roadway was driven to point A in Fig. 4, the value of the prominent prediction index K1 increased from the normal value of 0.20 mL/(g·min1/2) to 0.57 mL/(g·min1/2) under the premise of taking regional pre-pumping anti-outburst measure, exceeding the critical value of K1 determined by the working face of 0.4 mL/(g·min1/2). The risk of outburst at the working face increased. Therefore, the influence of the retaining rock pillar in the protective layer on the stress evolution, crack development and gas extraction of the protected coal layer was studied by taking the gas geology of 16,219 working face as the engineering background.

Figure 3
figure 3

The layout plan of the protective layer and the protected layer in Jinhe Coal Mine.

Full size image
Figure 4
figure 4

The profile diagram of the influence area of the retaining rock pillar in the protective layer (I-I profile).

Full size image

Simulation scheme of dynamic evolution law of mining stress

In order to analyze the dynamic evolution process of mining stress in the affected area of rock pillar, a numerical model was built by using FLAC3D software according to the engineering geological conditions of 16219 working face and Figs. 2 and 3. And make the following assumptions : 1) coal rock medium as homogeneous, continuous isotropic body ; 2 ) The in-situ stress environment is in the static horizontal pressure state ; 3) without considering the influence of time factor ; 4) The failure of coal-rock medium is regarded as an ideal elastic-plastic body and satisfies the Mohr-Coulomb criterion.

The overall length, width and height of the model were 855 m, 528 m and 240 m respectively (as shown in Fig. 5). During the calculation process, the horizontal movement of the side of the model was limited, and the vertical movement of the bottom surface was limited. The vertical stress value of 12.5 MPa was loaded on the top surface, the average weight of the rock layer was 2.5 t/m3, and the original rock stress environment of the coal seam was calculated to be 16.25 MPa.There were 681,614 cells in the model. The boundary coal pillars with a width of at least 100 m are retained around the model. The buried depth of the protective layer bottom plate was about 630 m, and the distance between the protective layer and the protected layer was 20 m. The initial vertical stress of coal seam was 16.25 MPa. The thickness of coal seam was 30 m. Set three stress monitoring lines inside the coal seam from bottom to top at 7 m, 17 m and 27 m above the bottom plate. Stress monitoring line positions was shown in Fig. 6. The 16,219 gas drainage roadway was located in the middle of the coal seam, the bottom plate was 14 m away from the top of coal seam, and the height of the roadway was 3.5 m. The excavation sequence was as follows: upper protective layer 16,107 working face, protected layer 16,209 working face, 16,219 gas drainage roadway and upper protective layer 16,119 working face.

Figure 5
figure 5

Model diagram.

Full size image
Figure 6
figure 6

Schematic diagram of stress monitoring lines inside the coal seam.

Full size image

Test scheme of CT scanning under conventional triaxial loading

When the 16,219 combined roadway was driven to the affected area of rock pillar ( as shown in the red part in Fig. 4), the coal blocks were processed into the standard specimens. The ultrasonic detector was used to measure the velocity of the specimens, and the patterns with similar wave velocity were selected (as shown in Fig. 7). The size of the specimen was shown in Table 1. Among them, three specimens were subjected to mechanical tests (as shown in Fig. 8), and three specimens were subjected to CT scanning tests (as shown in Fig. 9) under conventional triaxial loading.

Figure 7
figure 7

Coal blocks and specimens.

Full size image
Table 1 The size of the specimens.
Full size table
Figure 8
figure 8

Conventional triaxial loading test diagram.

Full size image
Figure 9
figure 9

Digital core X-ray scanning equipment.

Full size image

The ability of gas accumulation and migration was determined by the coal crack development27. In order to analyze the development of coal crack in the rock pillar affected area quantitatively, conventional triaxial loading tests under different confining pressures were carried out to obtain the stress-strain curves of coal specimens, and the CT scanning test was carried out to reconstruct the three-dimensional visual model of cracks inside specimens at different loading stages, so as to analysis the change law of crack volume and surface area quantitatively.

Verification scheme of gas extraction effects in different stress states

In order to test the gas extraction effects under different stress states, the coal seam residual gas content and the borehole gas extraction rules in pressure relief area, no mining affected area and rock pillar affected area were investigated in the 16,220, 16215and 16,219 working face of Jinhe Coal Mine. Under the conditions of analyzing the buried depth of working face and the influence of mining movement around it, the initial stress was calculated according to the mean weight of coal strata of 2.5 t/m3 to investigate the gas extraction effects and gas permeability coefficients under different stress states.

Results analysis

Simulation results of dynamic evolution law of mining stress

As shown in Fig. 10, after finishing the face mining of 16,107 protective layer and 16,209 protected layer, a stress increase area was formed in the area 25 m away from the right side of the 16,219 gas drainage roadway. The peak position was 45 m on the right side of this roadway, and the peak value was 25 MPa.

Figure 10
figure 10

Stress distribution in the stoping tendency direction at the 16,107 and 16,209 working face. (a) Cloud map of stress evolution. (b) Stress curve.

Full size image

As shown in Fig. 11, after finishing the mining of 16,119 working face, the stress of the retaining rock pillar further increased. The retaining rock pillar affected area covered on the protective layer was formed in the area 14 ~ 48 m away from the right side of the 16,219 gas drainage roadway. The peak position was 33 m on the right side of this roadway, and the peak value was 15 MPa. The peak stress in the rock pillar affected area was 5 ~ 15 times that of the coal body on both sides.

Figure 11
figure 11

Stress distribution in the stoping tendency direction at the 16,119 working face. OA, EF-Ingerated coal, AB-16209 Goaf, BC, DE-Protective layer mining pressure relief area, CD-Rock pillar stress influence zone

Full size image

The coal raw rock stress was 16.25 MPa. With the influence of the mining movement of the 16,107, 16,209 and 16,119 working face, the stress value of the underlie coal seam under the rock pillar increased first and then decreased, which were 21.4 MPa, 17.3 MPa and 15 MPa. While the stress concentration coefficients were 1.32, 1.06 and 0.92. It was necessary to further study the influence of stress on crack development by CT scanning test under conventional triaxial loading on basis of the study of stress evolution law.

Results of CT scanning under conventional triaxial loading

As shown in Fig. 12, MTS816 testing machine was used to carry out conventional triaxial loading tests under the confining pressure of 4 MPa, 8 MPa and 12 MPa. With the increase of confining pressure, the conventional triaxial compressive strength and strain of coal specimens increased. The strength of the coal specimens at different loading stages was shown in Table 2.

Figure 12
figure 12

Results of Conventional triaxial loading stress-strain curve under different confining pressure conditions.

Full size image
Table 2 Results of conventional triaxial loading tests with different confining pressures.
Full size table

① The confining pressure selection: The outline of coal roadway in the rock pillar affected area was smooth relatively. And the cutting quantity index S, which reflected the stress of the working face, was 1.8 ~ 2.2 kg/m and stable relatively, indicating that the coal roadway was not subject to high stress, and the whole coal body was in the approximate linear elastic stage. Combined with numerical simulation, it could be seen that the maximum stress value of coal body in the rock pillar affected area is 15 ~ 21.4 MPa after the mining disturbance of the adjacent working face. With the results of conventional triaxial loading tests under three confining pressures, the confining pressure of 8 MPa, which was more appropriate to the field, was selected as the test condition, and the CT scanning test under conventional triaxial loading was carried out.

② Scanning point ___location selection: The scanning points were selected in the middle of different loading stages on the stress-strain curve of conventional triaxial loading, and no less than 5 scanning points were set.

Combined with the numerical simulation results and the field conditions, CT scanning tests were carried out under the confining pressure of 8 MPa under triaxial loading for quantitatively analyzing the changes of crack surface area and crack volume at different loading stages. As shown in Figs. 13 and 14, scanning points were selected in four different loading stages: compaction, approximate elasticity, softening and post-peak failure. The statistical data obtained were shown in Table 3.

Figure 13
figure 13

Determination of CT scanning points and stress-strain curves under conventional triaxial loading.

Full size image

O-Initial state (before loading), AB-Compaction stage, BC-Softening stage, CD-Post-peak failure stage, 1 ~ 5: Scanning point.

Table 3 Internal cracks development of specimens at different loading stages.
Full size table
Figure 14
figure 14

CT scanning test under conventional triaxial loading.

Full size image

According to Fig. 1414, at the second scanning point of the coal specimens, that is, at the crack compaction stage, the crack surface area and crack volume of the coal specimens were reduced by 87% and 91% respectively compared with the ones at the initial state. While at the third scanning point of the coal specimens, that is, at the approximate elasticity stage, the crack surface area and crack volume of the coal specimens were reduced by 74% and 71% respectively compared with the ones at the initial state.

In order to further quantify the evolution and law of cracks expansion of specimens per unit volume, the crack density φn was defined as:

$${\varphi _n}=\frac{{{V_n}}}{V}$$
(4)

Where n was the scanning point, ranging from 1 to 5, Vn was the crack volume at each scanning point in unit mm3, and V was the total volume of coal specimen, which was 195095.45 mm3.

According to the CT scanning results from the compaction stage to the post-peak failure stage under conventional triaxial loading conditions, as shown in Figs. 13 and 14, the expression between stress value in different loading stages σn and crack density φn was as follows:

$${\sigma _n}=0.029{e^{0.025{\varphi _n}}}$$
(5)

Field verification of gas extraction effect under different stress states

The coal seam residual gas content and the borehole gas extraction rules in pressure relief area, no mining affected area and rock pillar affected area were investigated in the 16,220, 16215and 16,219 working face of Jinhe Coal Mine. The initial stress was calculated according to the mean weight of coal strata of 2.5 t/m3. And combined with the mining numerical simulation results, the coal stress values of the three sites were determined to be 3.5 MPa, 15 MPa and 21.4 MPa, respectively. The investigate sites and specific conditions were shown in Table 4.

Table 4 Investigate sites overviws.
Full size table

Under the conditions of 113 mm borehole diameter, 10 m×10 m layout and 30 kPa negative pressure extraction, the fitted extraction curves of pressure relief area, no mining affected area and stress concentration area within 60 days were shown in Fig. 15. After 15 months of extraction, the residual gas contents in the three test areas were 3.56 cm3/g·r, 6.84 cm3/g·r and 8.51 cm3/g·r, and the extraction rates were 80.8%, 67.4% and 50.4%, respectively.

Figure 15
figure 15

Extraction volume of extraction volume under different stress states.

Full size image

In addition, standard specimens were taken from the three working faces mentioned above, and the crack volume of the coal samples was tested using CT scanning, the changes in crack volume, permeability coefficient, and residual gas content under different stress states were shown in Fig. 1616. With the increase of stress concentration in coal seams, the fracture volume and permeability coefficient show a decreasing trend, while the residual gas content shows an increasing trend.

Figure 16
figure 16

Changes of crack volume, gas permeability coefficient and residual gas content under different stress states.

Full size image

Engineering application

As shown in Figs. 17 and 18, from August 25 to 31, The K1 value was lower than 0.20 mL/(g·min1/2) in the first 8 m range of the 16,219 combined roadway, and increased to 0.31 mL/(g·min1/2) when the roadway was driven to 12 m. At this time, the concentration of alley gas (carbon dioxide) was lower than 0.2% and stable relatively. On September 3 (driving 20 m to point A in Fig 0.4), the roadway entered to the affected area of the overlying rock pillar, and the K1 value was 0.57 mL/(g·min1/2), exceeding the standard. During this period, the cutting quantity was 1.9~2.0 kg/m. And there were still a lot of gas (carbon dioxide) in the rock pillar affected area has not been desorbed and released.

Figure 17
figure 17

Change of K1 value index of combined roadway heading face.

Full size image

The essence of pre-splitting blasting to enhance gas extraction is to use explosive explosion to crack coal seam, so as to improve the permeability of coal seam and promote efficient gas extraction. After blasting, the permeability coefficient of coal body will be greatly increased, and a large amount of gas will be desorbed from the adsorption state to the free state, which can reduce the gas pressure of coal seam, reduce the gas content, and reduce the gas internal energy of coal seam. After blasting, the stress of coal seam ( pressure relief ) was reduced, and the concentrated stress is transferred to the depth of coal seam, so as to achieve the purpose of rapid prevention and control of coal and gas outburst28.

Aiming at the research on the technology and effect of pre-splitting blasting in high-gas and low-permeability coal seams, the dynamic stress field in the coal body around the blasthole was established in the literature29, and the formation and expansion radius of the radial crack area in the blasting source area were analyzed. The radius of the crack area was applied to the design of the layout parameters of the blasting hole and the extraction hole in the coal seam, and the reasonable spacing between the blasting hole and the extraction hole was determined. After blasting, the permeability coefficient of the coal seam increased from 1.03 m/d to 1670 m/d, an increase of more than 1600 times.The literature30 conducted pre splitting blasting and permeability enhancement tests on the 4# coal seam and 5# coal seam of Xinji No.2 Mine. The average gas extraction purity of the borehole after blasting was 0.60m3/min, and the average gas extraction concentration was 2.2%, which was 3 ~ 4 times and 2 ~ 3 times higher than before blasting, respectively. The literature31 addresses the problems of long gas extraction cycles and poor extraction effects in high gas and low permeability coal seams. The study found that the influence radius of pre splitting blasting was 4.5 ~ 5.3 m, and the average gas extraction concentration after blasting increased by 2.17 times, the average gas extraction purity increased by 2.02 times, and the coal seam permeability coefficient increased by nearly 5.3 times, effectively reducing the risk of coal seam outburst. In general, according to the results of pre-splitting blasting permeability test in Huainan of Anhui Province, Pingmei and Jiaozuo of Henan Province32,33,34, the permeability of pre-splitting blasting coal seam in soft and low permeability coal seam could be increased by about 10 ~ 20 times, and the permeability of deep hole controlled pre-splitting blasting in medium hardness coal seam could be increased by more than 100 times.

Blasting anti-reflection technology had become one of the main and mature outburst prevention technologies for low permeability coal and gas outburst coal seams in China due to its advantages of short organization time and good anti-reflection effect. It was widely used in many mining areas in China. Therefore, pre-splitting blasting technology was proposed to be used in 16,219 working face of Jinhe Coal Mine to increase the permeability of coal body in the area affected by rock pillar, so as to reduce the stress of coal body and increase the permeability of coal body. The details were as follows:

  1. 1)

    The coal seam, which was 10 m in front of the excavation, was discharged from shallow holes by small aperture dense drilling to form a safety barrier.

  2. 2)

    Small aperture presplitting blasting measures should be taken ahead. Construction of 4 blasting pressure relief holes for every 5 m drilled, and whose position was controlled at 1.5 m in the middle and on both sides of the contour line of the roadway. The aperture was 42 mm, the depth of the hole was 10 m, and the charge of each hole was 4.8 kg, so as to promote the development of coal cracks until the area affected by the rock pillar was excavated out the roadway.

  3. 3)

    Water injection was used to displace and dissolve part of the gas in the coal.

Figure 18
figure 18

The concentration of gas (carbon dioxide) in combined roadway heading face.

Full size image

After taken the measures, sampled the coal seam again in the rock pillar affected area. At this time, the macroscopic cracks of the coal block developed, and it was difficult to make standard cylinder standard specimens. So make the cylindrical coal specimens with a diameter of 49.02 mm and a height of 50.34 mm, the results of CT scanning under no-load condition showed that the cracks volume of coal specimen is 988.06 mm3. According to Eq. (4), the crack density φ was 1.04%. After roadway mining on September 4, carbon dioxide from the coal was released and poured into the roadway. The concentration of gas was increased by 2 ~ 2.5 times. After the concentration of carbon dioxide gas increased and kept stable emission, the K1 value in the combined roadway decreased gradually from 0.57 mL/(g·min1/2) to 0.17 mL/(g·min1/2).

Discussion

According to the stress analysis results of coal in rock pillar affected area, the maximum stress of coal was 23.4 MPa, which was in the linear elastic stage of conventional triaxial loading under the confining pressure of 8 MPa. This stage was the crack compaction stage, and the cracks were mainly in closed state, and the gas flow capacity and permeability were reduced significantly compared with the ones in initial state, which was not conducive to gas extraction35. It could be seen from formula (5) that the crack density exponential increases approximately with the increase of stress in the softening and post-peak stages, which the increase of stress could promote crack development theoretically. However, this did not mean that the purpose of promoting cracks development can achieved by increasing the stress of coal seams in the rock pillar affected area. The effect in engineering practices was even reversed. Under the conditions of true triaxial stress, the increase of stress could often lead to the further closure of cracks in coal seam, resulting in poor hole formation effect, and thus the poor gas extraction effect36. The field verification results of gas extraction effect under different stress states also showed that the mining stress environment has a great influence on the gas extraction effect. With the increase of stress, the extraction rate and permeability coefficient of coal seam show a decreasing trend, while the residual gas content shows an increasing trend.

Therefore, it was imperative to study the development path of coal cracks under low stress states in the rock pillar affected area of 16,219 combined roadway. Based on the studies on the stress evolution, crack development and gas flow law of coal seam in rock pillar affected area, it could be seen that pressure relief antireflection was a necessary means to improve the gas flow capacity and gas extraction effect of coal seam37. Combined with the Eq. (5) and Fig. 14, it could be seen that the crack density value of coal seam in rock pillar affected area is equivalent to the softening stage under conventional triaxial loading by adopting the pressure relief antireflection measures mainly on the presplitting blasting means. But the stress value after pressure relief was reduced greatly, realizing the coal cracks development under low stress conditions. Based on the analysis of the crack development, gas desorption flow and sensitive index test of working face outburst risk prediction before and after taking the pressure relief antireflection measures, it could be seen that the adopted measures can achieve the dual effects of stress reduction and cracks increasement, and the outburst risk of coal seam in rock pillar affected area can been reduced effectively, and the safe driving of coal roadway can also be realized.

Conclusion

  1. 1.

    The stress evolution law of coal seam in the rock pillar affected area of 16,219 working face in Jinhe Coal Mine was obtained. And the theoretical calculation and numerical simulation results showed that the stress state of coal seam in the rock pillar affected area is in the approximated linear elastic stage.

  2. 2.

    The expression between the stress and crack density was established according to the results of CT scanning test in conventional triaxial loading. The crack surface area and crack volume of specimens after undergoing the stage of compaction and approximate elasticity were lower than the ones in the initial state, which was not conductive to gas extraction.

  3. 3.

    The coal cracks in the rock pillar affected area was mainly in closed state. By adopting the comprehensive measures of intensive drilling discharge, presplitting blasting antireflection and coal seam water injection, the cracks could develop, the gas could enter the roadway through the cracks, and the outburst prediction index K1 was also reduced.