Case studies of comprehensive gas control method during fully mechanized caving of low permeability ultra-thick coal seams

Slicing fully mechanized caving mining now is a common high-eciency mining method for ultra-thick coal seams. However, effective gas control has remained a diculty in fully mechanized top-coal caving mining of low permeability ultra-thick coal seams. This study focused on mining of the #9-15 coal in Liuhuanggou Coal Mine, Xinjiang Province, China, and combined theoretical analyses and eld test results for exploring comprehensive gas control methods for fully mechanized caving of low permeability ultra-thick coal seams. The No. (9-15)06 panel is a top slicing panel of the #9-15 coal with a mining height of 9 m, and the No. (4-5)02 goaf is located on the top of the panel. Through analysis, gas emissions in the No. (9-15)06 panel were mainly sourced from the coal wall, caving of top coal, goaf, and neighboring coal seams. A comprehensive gas control method based on source separation was proposed, which combined gas pre-drainage along the coal seam, high-position drilling on the top, pre-burial of pipes in the goaf, and pressure-balancing ventilation. Considering the poor gas pre-drainage effect for low permeability coal seams, the permeability of the coal seam was enhanced using hydraulic fracturing. According to coal seam and crustal stress distribution characteristics, the arrangement of the boreholes and backward segmented fracturing technology were designed. Field data show that coal underwent remarkable pre-fracturing under hydraulic fracturing. Mean gas pre-drainage from the boreholes was enhanced by nearly 4 times compared to the pre-hydraulic fracturing state. Finally, using the proposed comprehensive control method based on the gas sources, eld tests were performed in the No. (9-15)06 panel. Field measurement data demonstrate that gas concentration in the return airow uctuated within a range of 0.05%~0.35%, i.e., gas concentration did not exceed the standard. The proposed gas control method can provide insightful reference for the other similar projects.


Introduction
Slicing fully mechanized mining proves to be a high-e ciency mining method for ultra-thick seams that has been extensively applied in China. However, most of China's coal elds feature high gas and low permeability. 1,2 Ultra-thick seams with low permeability are generally poor in gas pre-drainage. When using fully mechanized mining methods, a great amount of gas is emitted from coal seams. The accumulation of high-concentrations of gas in fractures in the overlying strata of the goaf can easily trigger a serious accident, such as a gas explosion, which seriously threatens production safety in the mines. 3,4 In order to achieve effective gas control during fully mechanized mining of low permeability ultra-thick coal seams, numerous attempts to developed useful gas drainage and discharge methods have been made.
Fan et al. focused on fully mechanized mining of ultra-thick mines in Tashan Mine and proposed the gas in the goaf of the panel can be controlled well by combining surface vertical drilling and high position gas drainage roadways. 5 However, for the mines with great burial depth, gas drainage by drilling a great number of vertical holes. According to the characteristics of longwall coal mining in Australia, Guo et al. proposed a gas drainage and discharge method for the goaf using horizontal drilling, which exhibited favorable drilling performance during construction on the bottom of the roof fracturing zone. 6 Zhang et al. examined the fracture development rules for the roof when using high-dipping longwall coal caving and the related effect on gas drainage; 7 according to their results, they designed an optimal construction layer during high-position drilling for gas drainage and discharge from the No. 704 panel of Baojishan Mine. Liu et al. addressed gas accumulation in the upper corner of the goaf in a fully mechanized panel for tilted ultra-thick coal seams by pre-burying gas drainage and discharge pipes in the goaf. This technique was successfully applied in Wudong Mine. 8 In the above studies, straight line construction was performed for gas drainage in both surface and underground drilling process. In recent years, directional drilling has provided a new means of gas drainage in coal mines. Using directional drilling, holes can be drilled in coal seams along a preset direction. 9 Taking Daning Mine, Shanzi, China, as an example, Lu et al. carried out directional drilling in coal seams for gas pre-drainage and achieved favorable results. 10 Li et al. proposed a new inverse π-type drilling mode for gas drainage and discharge, which can realize comprehensive gas control after it passes through the caving zone, fracture zone, and coal seams in the working face. 11 In contrast with traditional drilling, directional coal seam drilling exhibits a series of advantages, including strong adaptability, high drilling proportion, and favorable gas drainage e ciency, which is particularly suited to gas control in single coal seams. However, directional drilling sets high demands on the coal seams, i.e., the coal seam should be quite permeable.
For low permeability coal seams, drilling along the coal seam has poor gas drainage performance; therefore, some fracturing techniques are necessary for pre-fracturing and permeability enhancement of coal seams. Deep-hole loosening blasting is a common pre-fracturing and permeability enhancement technique for coal seams; 12, 13 however, the technique has certain shortcomings. The dynamic shock wave produced in deep-hole blasting can easily cause damage to the surrounding rocks in the tunnel and make controlling rock deformation more di cult. In addition, unreasonable deep-hole blasting schemes also trigger numerous accidents such as gas explosions as well as coal and gas outbursts. Many scholars performed a numerous studies and explored some methods for permeability enhancement coal seams, such as hydraulic fracturing. [14][15][16] Hydraulic fracturing refers to drilling holes in the coal followed by generating fractures in the enclosed holes by utilizing a high-pressure water as medium and overcoming tensile strength of the rock and crustal stress using hydraulic pressure. 17 After hydraulic fracturing, the initial fractures in the coal form and numerous secondary fractures are produced. 17 Gas transport channels can be increased while promoting the desorption of adsorbed gas, thereby leading an increase in coal permeability and range. 18,19 In recent years, some scholars developed directional hydraulic fracturing and pulsed hydraulic fracturing techniques. 20,21 Effective gas control has always remained as a critical technical di culty in fully mechanized caving mining of low permeability ultra-thick coal seams. As described above, many scholars have proposed numerous gas control methods and ideas and gained favorable results. In fully mechanized mining of low permeability ultra-thick coal seams, gas has a variety of sources and is signi cantly di cult to control. Currently, one method addressing the control of multi-source gas in the working face is lacking.
This study focused on the mining of #9-15 coal in Liuhuanggou Coal Mine, Changji City, Xinjiang Province, China, analyzed the gas source in the working face and investigated a comprehensive gas control method for fully mechanized top-coal caving mining in low permeability ultra-thick coal seams by combining theoretical analysis and eld test results. Firstly, both mining and geological conditions of the mines were introduced, and the coal seam mechanical parameters and in-situ crustal stress were measured to provide basic data for the gas control scheme design. Next, the main gas sources in the No. (9-15) 06 panel were analyzed and a comprehensive gas control method based on the sources of gas was proposed. Considering that low permeability coal seams exhibited poor gas pre-drainage performance, coal seams were pre-fractured using hydraulic pressure to increase permeability. The hydraulic fracturing scheme and technology were designed and the pre-fracturing and permeability enhancement performance of the coal seams were veri ed through eld measurements. Finally, the effectiveness and reliability of the proposed source-separation comprehensive gas control method was validated using long-term eld tests. The proposed gas control concept and methods can provide useful reference for other similar projects. In the No. (9-15) 06 panel, a fully mechanized longwall caving method was used for top slicing of #9-15 coal, with a mining height of 9 m and a mechanized mining height of 3 m. The strike length and the inclined length of the working face were 1045 m and 100 m, respectively.

Coal seam mechanical parameters
In order to measure the related mechanical parameters of the #9-15 coal seam, #9-15 coal samples were collected along the tailgate. Because of the signi cant sample thickness, coal samples were sliced into three layers, namely, the upper slice, the middle slice, and the lower slice. Coal samples were processed into standard cylindrical specimens with different sizes (Φ50mm × 100 mm and Φ50mm × 25 mm). Next, using a MTS815 electro-hydraulic servo rock tester, uniaxial compressive, tri-axial compressive, and Brazilian disk splitting tests were performed on the standard samples to acquire the related parameters, including uniaxial compressive strength σ c , tensile strength σ t , elasticity modulus E, Poisson's ratio υ, cohesive force C, and internal friction angle φ. Each test was repeated 5 times and averaged. Table 1 lists the mechanical parameters of #9-15 coal. The mean uniaxial compressive strength of #9-15 coal was 35.35 MPa, suggesting great coal seam hardness. On account of poor permeability, gas pre-drainage performance along the coal seam was far from ideal. Gas control during mining in the No. (9-15)06 panel is of crucial importance.

In-situ measurement of crustal stress
In order to gain in-depth knowledge of the crustal stress distribution in Liuhuanggou Mine, in-situ measurements were performed. The measurement accuracy of in-situ stress is related to the selected method and sensors. In this study, using stress relief method, a CSIRO hollow inclusion stress gauge from Australia was used, which possesses numerous advantages including accuracy, reasonable distribution of numerous strain gauges, and a wide application range.
The selection of the measuring points should avoid the areas with complex geological structures and signi cant mining effects as far as possible. 5 measuring points of crustal stress were selected, as the arrangement displayed in Fig. 3. Table 2 lists the arrangement of the crustal stress measuring points and the boreholes.  (1) Gas emission quantity from coal walls (Q 1 ) When the working face moved forwards, fresh coal walls are exposed to the air, and pressure balance was broken after reliving the rock pressure. Due to the existence of a gas pressure gradient in the coal, gas was emitted along coal fractures and towards the working face.
(2) Gas emission quantity from top coal caving (Q 2 ) Before the caving of top coal in the working face, the top coal was rst fully broken under the combined action of supporting pressure and the supports, and the original gas pressure balance in the coal seam was broken during crushing of the top coal. Accordingly, the adsorbed gas in the coal in the stable state was converted into free gas, which was then emitted into the working face during top coal caving.
(3) Gas emission quantity from the goaf (Q 3 ) Using the fully mechanized caving method, the recovery ratio in the No. (9-15)06 panel was approximately 85%, and a great amount of coal would be left in the goaf. Therefore, gas from the remaining coal was the main source of the emitted gas in the goaf. In addition, mining of the top layer of the #9-15 coal seam affected the bottom coal seam to be mined, and numerous fractures were generated, thereby emitting a great amount of gas. In conclusion, total gas emission quantity from the No. (9-15)06 panel (Q) equaled to the sum of the gas emission quantities from above sources, i.e., Q = Q 1 + Q 2 + Q 3 + Q 4 . Q 1 and Q 2 can be lowered by gas predrainage along the coal seams. 23 The gas from latter two sources accumulated in fractures in the overlying goaf strata. The friction during the breaking movement of the overlying strata easily triggered gas explosion accidents in the goaf, which seriously threatened production safety in the working face. Therefore, for the gas accumulated on the top of the goaf (Q 3 + Q 4 ), effective measures are required for drainage or dilution.

Enhancement Of Coal Seam Permeability Via Hydraulic Fracturing
On account of the high strength and poor permeability of the #9-15 coal seam, gas pre-drainage performance was poor. Therefore, pre-fracturing and permeability enhancement should be rst performed on the coal seam via hydraulic fracturing for improving gas pre-drainage effect of the coal seam.

Crack initiation pressure
Crack initiation pressure (hereinafter referred to as the initiation pressure) refers to the maximum water pressure during hydraulic fracturing. Numerous factors including burial depth of the coal seam, crustal stress, coal mechanical properties, and original cracks can affect initiation pressure. According to hydraulic fracturing theory, 14, 24-26 tensile cracks dominate the hydraulic fracturing process. Under the action of water pressure in the boreholes and the crustal stress eld, when the tensile stress on the tip of the cracks exceeds the tensile strength σ t , cracks develop gradually. By ignoring the water seepage effect from the boreholes on the surrounding media, the initiation pressure can be estimated as [26] where σ 1 denotes the maximum principal stress, σ 3 denotes the minimum principal stress, and σ t denotes the coal tensile strength.
Additionally, gas pressure also signi cantly affects crack initialization and propagation under hydraulic fracturing. 27 Based on statistics from a large number of coal mines in China, Li et al. carried out regression analysis and determined the following relationship between crack initialization pressure P k2 , burial depth of the coal seam H, and gas pressure P 0 : 19 According to the minimum pressure principle, the minimum value between P k1 and P k2 was selected as the initialization pressure of the coal seam (P k ): In this study, mean burial depth of the working face was approximately 420 m, mean unit weight of the overlying strata was 23 kN/m 3 , and gas pressure of the coal seam was 0.5 Mpa. According to measurements, the maximum principal stress and minimum principal stress in the test region were 10.54 MPa and 6.34 MPa, respectively, through inversion. The average tensile strength of #9-15 coal was 1.97 MPa (Table 1). Therefore, P k1 = 10.45 MPa and P k2 =10.74 Mpa. According to Eq. (3), P k =10.45 MPa.

Coal seam permeability enhancement via hydraulic fracturing
(1) Arrangement of hydraulic fracturing boreholes Based on engineering and geological conditions of the No. (9-15)06 panel and coal seam occurrence characteristics, the spatial positions of the hydraulic fracturing boreholes and high-position gas drainage boreholes on the roof of the working face were analyzed to avoid damages to gas drainage boreholes. Figure 6 displays the arrangement of hydraulic fracturing boreholes.
A set of boreholes with a diameter of 75 mm were arranged on the inner sides of tailgate and headgate. The rst group boreholes was arranged 15 m from the set-up room of the No. (9-15)06 panel, and the interval between every two groups of boreholes is 8 m. Table 3 lists the detailed parameters of the boreholes.
(2) Field equipment for hydraulic fracturing included a high-pressure water-injection pump, water tank, ow controller, pressure gauge, and high-pressure water pipes. The maximum water injection pressure of the water injection pump was 35 MPa, and the water ow velocity was 200 L/min.

(3) Sealing of boreholes
Borehole sealing is a critical make-or-break step that determines the effectiveness of hydraulic fracturing.
Currently, people always use cement and other special hole packers for sealing. 19 The latter was used in the present eld test. Under high-pressure water, the rubber gasket in the hole packer was compressed, then expanded radially and reached the inner wall of the hole to achieve a seal. Using this method, borehole sealing e ciency was high and the whole operating process was simple. However, this process set high requirements on the boreholes, i.e., the construction quality should be high and the borehole wall should be smooth.

(4) Fracturing technology
Spatial propagation characteristics of the hydraulic fractures directly affect permeability enhancement effect of the coal seam. The hydraulic fractures always propagated along the direction parallel to the maximum principal stress and perpendicular to the minimum principal stress. 17,28,29 According to in-situ measurement results, vertical stress dominated the crustal stress eld in Liuhuanggou Mine, and the dominant angle of the minimum horizontal principal stress was 140°, i.e., the angle between dominant direction of the minimum horizontal principal stress and the gate axis of the panel was approximately 84°. Hydraulic fractures in the boreholes are more likely to propagate radially from the borehole; therefore, in order to enhance hydraulic fracturing performance, backward segmented fracturing technology was adopted. Figure 7 illustrates the present borehole sealing method. Two hole packers were connected in series, and the spacing distance between them (denoted as L) was the sectional fracturing length. In this study, L = 8 ~ 10 m. In Fig. 8, the opening pressure of the pressure valve in hole packer A equaled that in hole sealer B, and both were smaller than the opening pressure of pressure valve C. Accordingly, local borehole sealing and fracturing can be easily achieved.

Veri cation of permeability enhancement of the coal seam via hydraulic fracturing
In order to evaluate permeability enhancement performance of the coal seam under hydraulic fracturing, this study used the following two methods for veri cation-boroscope examination and borehole gas emission quantity measurement.
(1) Method 1: boroscope examination A stratum detector (Fig. 8) was employed for examining fracturing results in the drilling hole before and after hydraulic fracturing. Both crack initiation and propagation behaviors in the holes were the research emphasis in this study. Figure 9 displays the distribution of cracks in the borehole before and after hydraulic fracturing. The inner wall of the initial borehole was very complete in shape while few original cracks were found in the borehole. After hydraulic fracturing, cracks mainly propagated radially from the borehole wall or in spiral pattern along the borehole wall, which correlates with the distribution of crustal stress in the test region, which was consistent with the predicted results as described above. In conclusion, the coal seam showed satisfactory pre-fracturing performance.
(2) Method 2: Measurement of gas emission quantity from the borehole Gas pre-drainage ows before and after hydraulic fracturing were monitored using an ori ce plate owmeter. 30 The gas ow in the borehole can be calculated as: The readings from the U-shaped differential pressure gauge were obtained and substituted into Eq. (4) and Eq. (5) for calculating the exhausted gas mixture and pure gas. According to eld test results, the variation rules of gas pre-drainage quality from a single borehole with time before and after hydraulic fracturing were plotted (Fig. 10).
It can be observed from Fig. 10 that after the implementation of hydraulic fracturing, mean gas predrainage quantity from the borehole was enhanced by nearly 4 times the original value. In addition, for low permeability hard coal from Liuhuanggou Mine, fractures in the coal around the borehole were propagated after hydraulic fracturing, thereby enhancing the permeability of the surrounding coal seams and remarkably improving gas pre-drainage performance.

Comprehensive Gas Control Based On Sources
By analyzing four main gas emission sources in the No. (9-15)06 panel, a comprehensive gas control method based was proposed (Fig. 11). Before mining in the panel, the gas emission quantity from the coal wall (Q 1 ) and the gas emission quantity form top coal caving (Q 2 ) were lowered by performing gas pre-drainage along the coal seam. During mining of the working face, the gas accumulated in the caving zone and the fracture zone (Q 3 + Q 4 ) was discharged via high-position boreholes on the roof. Meanwhile, drainage pipes were buried along the tailgate in the panel to address the accumulation of gas in the upper corner. Additionally, pressure-balancing ventilation was implemented during mining. The pressure difference between the two gate roads was reduced by adjusting the ventilation parameters to reduce air leakage and gas emissions in the goaf.
(1) Gas pre-drainage along the coal seam Hydraulic fractured boreholes were also used as gas pre-drainage holes before the mining in the panel for gas pre-drainage along the coal seam (Fig. 12). Accordingly, a hole achieved two purposes-enhancement of gas pre-drainage and reduction of construction volume.
The gas pre-drainage system along the coal seam mainly consisted of drainage pipe, lter pipe, collecting pipe, and hole-sealing materials (Fig. 12). The auxiliary devices mainly included the owmeter, the concentration meter, and the pressure gauges. Metal sleeves with a diameter of 40 mm and a wall thickness of 3 mm were put in each hydraulic fractured borehole and some small-diameter holes were arranged on one end of the sleeves. Two sections of sleeves were arranged around each borehole, and each section was 3 ~ 4 m in length. After installation of the sleeves, boreholes were sealed using polyurethane (PU). It should be noted that hole-sealing depth should not exceed the burial depth of the gas drainage pipe in the borehole. Finally, the sleeves were connected with the gas drainage header pipes.
The gas drainage head pipes provided a negative pressure of 25 kPa.
(2) Drainage of gas from the caving zone via high-position boreholes Gas emitted from pressure relief in the coal seam oor was an important source of gas in the panel during mining. After the mining, the overlying strata underwent fracturing, nally forming the bent subsidence zone, the fracture zone, and the caving zone; 31-34 meanwhile, a great number of layered and vertical fractures were produced, 35,36 and gas accumulated in these regions. Accumulation of the gas emitted from the #4-5 coal, the #7 coal and the #9-15 coal increased the possibility of gas accidents. Therefore, on the basis of gas pre-drainage along the coal seam before the mining, long-strike holes were drilled in the high-position drilling site on the roof for discharging the gas accumulated in the caving zone and the fracture zone during mining (Fig. 13). The high-position drilling site was located in the tailgate of the No. (9-15)06 panel. Drilling sites were set at an interval of 60 m. In each drilling site, 16 long boreholes in 4 rows were arranged along the strike of the coal seam in a fan-shaped distribution.
(3) Gas drainage at the upper corner of the panel The gas produced in pressure relief from the oor can rush into the goaf and enter the upper corner, which can result in an abundance of gas in the corner. In order to avoid above problem, gas drainage pipes were buried at the air return corner on the top of the goaf during mining, which assists gas drainage in the goaf (Fig. 13).
(4) Pressure-balancing ventilation for reducing air leakage in the goaf People generally increased the air supply for diluting the gas concentration in the return air ow for the panels with U-shaped ventilation. Unfortunately, it is an incorrect practice. Increasing air supply quantity in the panel blindly can increase the air pressure difference between the goaf and the panel as well as air leakage in the panel, leading to an increase in gas emission in the goaf and gas concentration in the return air ow. 37 On that basis, pressure-balancing ventilation 38 was adopted to reduce the pressure difference between two sides in both the air intake and return roadways and changing the pressure distribution in the ventilation system. Accordingly, air leakage in the goaf can be lowered to inhibit gas emission from the upper corner of the goaf.

Analysis of gas comprehensive control effect
Finally, a long-term led test was performed in the No. (9-15)06 panel using the comprehensive control method based on the sources of gas. Overall, gas was adequately controlled. Figure 14 displays the monitoring gas concentrations in the panel during the mining period. The gas concentration in the return air ow uctuated within a range of 0.05%~0.35%, and the maximum gas concentration was no greater than 0.35%, which is far below the early warning value 0.5%.

Conclusions
Gas control has always been a key problem during slice mining of ultra-thick coal seams with high gas content. This study examined the No. Based on the arrangement in the No. (9-15)06 panel and the surrounding mining condition, gas in the panel had four emission sources-gas emission from the coal wall, gas emission from top coal caving, gas emission from the goaf, and gas emission from the neighboring coal seams. Gas emitted during mining easily accumulated in the fractures in the overlying goaf, which posed great threat to production safety in the panel. Therefore, a comprehensive control method based on the sources of gas for the panel was proposed in this study. Before mining, gas emissions from the coal walls and top coal caving were lowered via gas pre-drainage along the coal seam. After mining, gas accumulated in the caving zone and the fracturing zone was drained by using long-strike boreholes in the high-position drilling site. Drainage pipes were buried in the upper corner of goaf to avoid gas accumulation emitted from de-stressed oor coal seams. During mining, by means of pressure-balancing ventilation, air leakage in the goaf and gas emission from the goaf can be reduced.
Due to the great hardness and poor permeability of the #9-15 coal seam, conventional gas pre-drainage methods along the seam do not perform well. This study enhanced the permeability of the coal seam via hydraulic fracturing. Based on the measured results of crustal stress, the crack initiation pressure of the #9 ~ 15 coal seam under hydraulic fracturing was 10.45 MPa, and the arrangement scheme of the boreholes for hydraulic fracturing in the coal seam was designed. According to the measured crustal stress eld, it was predicted that hydraulic fractures were more likely to propagate radially from the borehole. Backward segmented hydraulic fracturing technology was designed for improving hydraulic fracturing performance. In order to validate the superiority of the proposed method, the pre-fracturing performance of the coal seam was evaluated using boroscope examination and gas emission tests. Field test results demonstrate that after hydraulic fracturing, obvious fracture development can be observed in the coal around the boreholes. Fractures mainly initiated radially from the hole wall and then propagated along the hole wall in spiral pattern, which was consistent with crustal stress measurements. After hydraulic fracturing, mean gas pre-drainage was enhanced by nearly 4 times.
Finally, a long-term eld test was performed in the No. (9)(10)(11)(12)(13)(14)(15) panel to validate the superiority of the proposed comprehensive control method based on the gas sources. Long-term monitoring results indicate that gas concentration in the return air ow uctuated within a range of 0.05%~0.35%. No warning of excessive gas appeared in the panel. Adequate gas control can effectively ensure mining safety in the panel.