Development of the fracture zone water flow level in the overburden

Free-water level calculations in fracture zones are usually performed using the empirical formula recommended in Regulations of Buildings, Water, Railway and Main Well Lane Leaving Coal Pillar and Press Coal Mining (State Bureau of the Coal Industry 2017). The formula applies to coal seams that can be mined in single layers of between about 1 and 3 m thick, and whose cumulative thickness does not exceed 15 m. Fully-mechanized extraction has now been implemented for thick, single coal seams, with rapid working face advance and extensive destruction of roof rock strata (Teng 2011). Consequently, the traditional empirical, predictive formula may no longer be applicable. Thus, studying the changes in the free-water level in the overburden fractured zone with respect to fully-mechanized mining is significant in coal mine safety.

There have been many studies of free-water surface level development in overburden. Kang (1998) suggested that decreasing primary mining thickness might lower the water level in the fractured zone, after studying the effects of various exploitation methods on water levels in the overburden fractured zone. Gao et al. (2012) used two arcs to fit the edges of the subsidence trough, and studied the relationship between fracture zone permeability and the interlayer tensile deformation of the overlying strata. In other studies the relationship was deduced between mining parameters and generalized damage factors by combining the generalized damage-factor definition of an interlayer stratum and the free water level in the fractured zone (Zhao & Yu 2010). Du & Bai (2012) elaborated the phenomena of intense static pressure, moderate dynamic pressure, and insufficient fracture growth to investigate the fracture mechanisms in overburden caused by fully-mechanized mining. Chen et al. (2006) determined the free-water level by simulating the failure of overburden caused by such mining using RFPA software. Liu et al. (2014) used various methods to find the free-water development level, by studying its dynamic characteristics in relation to fully-mechanized mining. Zhang et al. (2014) analyzed failures of overlying strata in a fully-mechanized extraction face and believed that the key stratum (KS) controlled overburden failure development. Xu et al. (2011) developed an empirical formula to calculate the ‘two zones’ height of fully-mechanized extraction faces using regression analysis. Xu et al. (2012, 2009) showed that the location of a primary key stratum (PKS) affected the development of the free-water level when they studied PKS impact.

Coal Seam No. 3 in Guojiahe mine is a stable coal bed, with an average thickness of 15 m. It dips at between 1 and 15° (average 7°). The working face is 235 m long, and between about 441 and 583 m below ground level. The overlying strata comprise 33 to 149 m of loess and laterite, and then about 407 to 446 m of consolidated strata. The immediate roof is dense and hard, and includes dark grey and sandy mudstones, and grey-white sandstone. Hydrogeological conditions are simple, the working face's main water source is the Luohe Formation sandstone. The principal overlying aquiclude is a sandy mudstone about 203 m above. The mechanical characteristics of the overburden strata are given in Table 1.

Using a mineable thickness of 15 m, empirical formulae (1), (2), and (3) yield different free-water surface maximum levels for the fractured zone – i.e., 59.9, 87.46 and 150.64 m, respectively – showing that the results from the traditional formulae for fully-mechanized upper coal extraction differ significantly. In other words, the traditional formulae do not apply to thick coal seams with fully-mechanized extraction. It is thought that both the differing lithology within the overburden and the KS locations might cause differential free-water surface level development, even at the same depth and with the same mining height. Low- and high-level key layers have different effects on free water surface development in the overburden strata. Thus, only with the correct combination of ratios of bedrock to overburden thickness (Jz) and bedrock thickness to mining height (Jc) with a KS, the maximum height of water conducting fractured zone can be estimated.

The PKS in the overlying strata is the main control on fracture zone height development. When the PKS is within the excavation zone and at shallow depth, the free-water zone of the fractured zone above develops to ground level. When it is within the fracture zone, the free-water surface is above it. On the other hand, when the PKS is in the bending subsidence zone, the top of the free-water zone is below it. Thus, determination of the PKS location within the ‘three zones’ of the overlying strata is crucial to any study of the free-water level. Zhao et al. (2015a, 2015b) and Huang (2002), report that the preliminary location of the PKS in ‘three zones’ was initially determined by calculating Jz and Jc, after which the water level range in the fractured zone was determined. Finally, formula (6) was adopted to calculate the tensile deformation of the rock strata within the range and to determine the maximum free water level in the fractured zone can be calculated. The detailed method is shown in Figure 2.

Field measurements and calculations show that the bedrock displacement angle in working faces 1304 and 1305 in Guojiahe mine is 65°, the surface soil layer displacement angle 45°, and the tangent of the major influence angle tanβ 1.64. The surface soil layer and rock strata above the coal seam are about 148 and 403 m thick, respectively. The average dip angle of coal seam is 7°. The comprehensive residual rock strata bulking coefficient, Ks, is 1.065; and the tensile deformation limit [ε] of the overburden strata is taken as 2.9 mm/m. Using the method indicated in Figure 2, the calculated maximum thickness of the fractured zone was determined as in Table 2.

• (1)

  According to the KS theory, No. 3, No. 5, No. 11, No. 14 and No. 19 rock strata were determined as sub key strata while No. 34 rock stratum was the PKS. The deformation and fracture parameters of key strata are as shown in Table 2.

• (2)

  The calculated ratios of bedrock to overburden thickness, J_Z, and of bedrock thickness to mining height, J_C, are: J_Z = 403.57/148.58 = 2.71 > 0.8, and J_C = 403.47/15 = 26.9 > 25. It can be inferred, therefore, that the PKS is in the bending and subsidence zones, and that the highest free-water level in the fractured zone was less than the vertical distance from the PKS to the coal seam roof.

• (3)
  The subsidence coefficient of No. 27 rock stratum η27 was calculated as:

  

The calculation results, ε₂₇ = 2.7 mm/m < [ε] = 2.9 mm/m, and ε₂₆ = 3.65 mm/m > [ε], indicate that No. 26 rock stratum was fractured while No. 27 was not. Thus, under the geological and mining conditions of this working face, the height of the fractured zone in the overlying strata developed to the level of stratum 26, which means the maximum development height of water-flowing fractured zone in the overlying strata was 183.06 m and the thickness of the water-flowing fractured zone was around 183.06/15 = 12.20 M.

A model 5,000 (X) × 2,000 (Y) × 200 mm (Z) was chosen to meet the experimental conditions and objectives. The geometry, volume weight, and stress similarity constants were 200, 1.56 and 312, respectively, according to similarity theory. Working faces 1303 and 1305, both with an average width of 1,225 mm, were arranged in a coal seam with 100 mm section coal pillars between them, to match the experimental similarity ratio. The working height of the coal face was 75 mm. To eliminate boundary effects, 1,225 mm boundary coal pillars were placed on both sides of the model, and were 1,095 mm tall. The rest of the overburden strata were loaded using the gravity compensation load (because the experimental model does not extend to the surface).

Figure 3 presents the failure characteristics of the overburden strata during mining at the working face. Failures occurred periodically as the face advanced. When failure developed to the KS, an abscission space formed under it. As the KS fractured, the abscission space closed gradually and the gob was compacted. The periodic KS failure controlled development of the free-water flow fracture. The free-flowing fracture zone thickness variation curves that arose during mining are shown in Figure 4.

When mining of face 1303 was finished, the failure zone and free-flowing fracture zone heights were 70 and 144 m respectively (Figure 3(a)). When face 1305 was finished, the respective zone heights were 66 and 146 m (Figure 3(b)). After failure of the section coal pillars, the fractures in the strata above them developed gradually upward during mining, and the maximum free-water level in the fractured zone was 188 m (around 12.53 M) (Figure 3(c)).

Figure 4 shows the step-like variation of the relationship between the free-flowing fracture zone thickness and the distance ahead of the working face. When overlying strata failure developed as far upward as the KS, fracture zone development stopped due to the KS' strength. As the working face advanced, however, the length of the suspended span of the KS exceeded its maximum unsupported capacity, and fracture zone development continued upward again. In other words, fracture zone development was retarded when the failure of the overlying strata reached KS level. The relationship between the height of the free-flowing fracture zone and the distance ahead of the working face display a step-like variation, revealing the controlling effects of the KS.

The UDEC numerical model was designed to study development of the free-flowing fracture zone height in the overlying rock under sub-level extraction mining in the Guojiahe mine. The model dimensions are 275 (h) × 600 m (l). The thicknesses of the coal seam, bedrock, and coal seam floor are 15, 240, and 20 m, respectively, and working face length is 300 m. To eliminate boundary effects, 150 m boundary coal pillars were placed both sides of the working face. The model adopted the Mohr-Coulomb criterion, in which horizontal displacements are limited by the left and right boundaries, and vertical displacements by the upper and lower boundaries. The strata at the top of the model were simulated with a 6.46 MPa load, except at the boundary.

The failure forms of the overburden during mining are shown in Figure 5. When the working face reached 70 m, the overburden failure form became conspicuous (Figure 5(a)). When it reached 150 m, the failure height was around 73 m and an abscission space formed gradually in No. 5 sub-KS (Figure 5(b)). As the working face advanced to 230 m, No. 5 sub-KS fractured. At this time the water-flow fissured zone level was around 133 m (Figure 5(c)). The mine was almost fully exploited when the length of mining face reached 300 m, the gob compacted, and the water-flow fractured zone in the overlying strata developed further upward; the maximum water-flow fractured zone level was 193 m (around 12.87 M) (Figure 5(d)).

The empirical formulae, predictive model of movement and deformation in the upper rock-soil layer in the water-flow fractured zone, and the similarity and numerical models all yielded different maximum heights for the water-flowing fractured zone – Table 3. The values calculated using the traditional empirical formulae differ significantly, while those derived from the other models are close and show good consistency.

• (1)

  The predictive model of movement and deformation of the upper rock-soil layer in the water-flowing fractured zone of the overlying strata shows that the maximum height of the water-flowing fractured zone in the fully-mechanized upper coal face is 183.06 m, which is 12.2 times the mining height. In the overlying strata, periodic failures of lower level key strata control the height development of the water-flowing fractured zone, which is restrained by high level key strata.

• (2)

  The similarity model showed that, after the section coal pillars failed in the middle of the working face, the maximum fractured zone water level was 188 m, while the numerical simulation showed that, under full mining conditions, the water level was 193 m. The relationship between the water level in the fractured zone and the distance forward to the working face varies step-wise, revealing the controlling effects of the overlying KS on water level.

• (3)

  The maximum water level in the fractured zone, calculated by the predictive model, agrees well with results from the similarity and numerical simulations. The predictive model can yield reasonable estimates of the water level in the fractured zone above thick coal seams under fully-mechanized extraction conditions.

Financial support for this work, provided by the National Natural Science Foundation of China (No. 51874231), the Key Scientific and Technological Innovation Team Project of Shaanxi Province (No. 2018TD-038), the Key Laboratory Scientific Research Project Funding Project of Shaanxi Province (No.13JS066), is gratefully acknowledged.