TY - JOUR
AU - Guo, L, Q
AB - Abstract The deformation and failure of overburden has been a major and challenging issue in the prevention of geological disasters in mines. In this study, Brillouin optical time-domain reflectometry is introduced to detect overburden deformation and failure. To use this technique, several boreholes are drilled in the top of a coal seam in a laneway, and optical cables are arranged in the boreholes to form a two-dimensional measurement cross section. During different periods throughout the mining process, the strain characteristics of the optical cables in the boreholes can be collected and dynamically analyzed to study the mining-induced changes in the strain in the overburden. This analysis can provide effective technical parameters for safe production in mines. A distributed optical fiber sensing (DOFS) test was conducted in the working face of a coal mine in Inner Mongolia, and a total of 25 sets of effective test data were obtained. The dynamic development characteristics of the overburden stress, deformation, and failure due to mining are analyzed. The characteristics of the strain distribution in the borehole are closely related to the orientation of the boreholes. Based on the strain distribution characteristics along the optical cable, the locations of the breakpoints, and the geological data of the 3-1 coal seam (where ‘3-1’ is the indication code of the coal seam), the caved zone and the fractured zone extend 17 m and 48 m above the coal seam, respectively. These results are basically consistent with those from borehole resistivity computed tomography analyses. The results indicate that DOFS technology provides high-resolution results and is a good monitoring approach. Additionally, this technique can accurately determine the characteristics of overburden damage and has favorable application prospects. overburden, distributed optical fiber sensing, borehole, strain, caved zone, fractured zone 1. Introduction During coal mining, the loss of the underlying coal and rock support considerably changes the structural state of the overburden. The overburden may experience fracturing and other deformation, changing the original equilibrium state of the overburden around the mining area. The overburden will consequently form a bending subsidence zone, fractured zone, and caved zone, from top to bottom (Wu et al2012). Accurately determining the height of the damage zones and their development characteristics can provide critical guidelines for controlling water damage and for extracting gas within the mining area (State Administration of Work State Administration of Coal Mine Safety 2011). Research on the deformation and failure of overburden can be traced back to the 1850s. Through progress in this area of research, scholars have proposed many research methods. At present, the most developed and widely used methods include the utilization of classic theory (Song 1988, Liu 1995, Qian et al1996, Lu 2002, Wu 2002), empirical formulas (State Bureau of Coal Industry 2000), numerical simulations (Hamdan et al2014), physical similarity simulations (Chen and Yu 2000, Zhang et al2011, Zhang et al2016), and field measurements. Additionally, scholars are very interested in direct field measurements. There are two major categories of field testing techniques for measuring the deformation and failure of overburden: direct and indirect. Direct techniques include borehole imaging (Yang 2016), drilling fluid testing (Kang et al2002), and borehole pressure testing (Zhang et al2000). Indirect techniques include transient electromagnetic methods (Shen and KongShen 2000), shock wave computed tomography (CT) (Zhang et al2004), ultrasound imaging (Sun and Xu 2008), resistivity methods (Zhang et al2009), conductivity imaging (Zhang et al2005), and microseismic methods (Sun et al2012). The application of these methods greatly improves the accuracy of detecting the height of overburden damage and provides some technical support for the safe and efficient production of coal mines. However, due to various practical constraints, coal mining stick lacks a method that allows continuous high-resolution monitoring of the deformation and damage process. Additionally, existing monitoring sensors are mostly low-resolution point sensors, which are vulnerable to leaks and false detection. Therefore, the existing monitoring sensors are insufficient to provide a comprehensive data set to reconstruct the relevant parameters of the monitoring area. Based on these observations, this paper proposes the use of Brillouin optical time-domain reflectometry (BOTDR) to dynamically test the deformation and failure of overburden. BOTDR is a newly developed and innovative measuring technique that can monitor the axial strain along an optical cable, and it can be applied in harsh detection environments without incurring electromagnetic interference (Shi et al2004). Through the deployment of a distributed optical cable during coal seam mining, the strain state of the overburden during deformation and failure can be determined. In addition, based on the geological conditions of the 3-1 coal seam and the results of borehole resistivity CT, the characteristic changes in the overburden are analyzed to derive the heights of the caved zone and fractured zone. The results show that the distributed optical fiber sensing (DOFS) system can be instrumental in testing and analyzing the deformation and changes in overburden. 2. DOFS technology From a novel sensing technology proposed in the early 1980s, optical fiber sensing has become the fastest growing advanced sensing technology because of its durability, stability, and insensitivity to external electromagnetic perturbations. DOFS technologies can measure data at any point along the optical cable through light scattering, thereby enabling continuous and real-time measurements along the entire optic cable (Horiguchi et al1989, Ohno et al2001, Piao et al2009, Bernini et al2012). The continuous measurement of the optical fiber can provide global strain measurements, which are more advantageous than data extrapolated from a few point measurements collected with point sensors and can help reconstruct the multidimensional strain distributions. Although most photons move forward as light propagates along an optical fiber, some photons are scattered back toward the optical source via a number of mechanisms. The basis of BOTDR is the spatiotemporal analysis of the backscattered light by an optical detector as the forward-propagating light travels along the fiber. BOTDR focuses on the backscattered Brillouin light, which is produced by the interaction between the pulsed light and acoustic phonons. Figure 1 shows the working principle of BOTDR. In figure 1(a), the pulsed light is transmitted from one end of the fiber at a certain frequency. The Brillouin scattered light is scattered back to the incident end and enters the optical signal processing unit of the test instrument, which converts the optical signal into an electrical signal via the photodiode. The electrical signal is then processed to obtain the scattering spectrum, as shown in figure 1(b). Figure 1. View largeDownload slide Working principle of BOTDR. Figure 1. View largeDownload slide Working principle of BOTDR. As shown in figure 1(c), the Brillouin backscatter spectrum is generally a Lorentz-type spectrum, in which the peak power is the Brillouin frequency shift. Generally, the Brillouin frequency shift is a linear function with respect to the axial strain and fiber temperature. The influence of temperature on the Brillouin frequency shift can be neglected if the temperature differences are less than 5 °C, and the optical fiber strain and Brillouin frequency shift can be expressed as equation (1) (Xu et al2015, Zhang et al2015): νB(ε)=νB(0)+dνB(ε)dεε, 1 where νB(ε) is the Brillouin frequency shift with an axial strain of ε along the optical fiber, νB(0) is the Brillouin frequency shift with zero axial strain along the optical fiber, dνB(ε)/dε is a constant of approximately 493 MHz, and ε is the axial strain of the optical fiber. In this paper, to measure the overburden deformation in the working face of the 3-1 coal seam, we use an AV6419 fiber strain distribution tester as our BOTDR instrument. The AV6419 is made in China and has been used for measurement and monitoring in various civil engineering fields. Table 1 displays the main parameters of the AV6419. The sampling resolution is set to 0.1 m, and the range is configured to 0.5 km. Table 1. Test instrument parameters. Project Parameter Working wavelength (nm) 1550 ± 5 Fiber type Single-mode optical fiber (SMF) Maximum dynamic range (dB) 15 Spatial resolution (m) 1 Maximum sampling resolution (m) 0.05 Number of sampling points 20 000 Strain test accuracy (με) ±50(10–20 ns); ±10(50–200 ns) Strain test repeatability (με) ≤ ±100 Strain test range (με) (-15 000)–(+15 000) Test range (km) 0.5, 1, 2, 5, 10, 20, 40, 80 Average number of readings 210–224 Frequency sweep interval (MHz) 1, 2, 5, 10, 20, 50 Frequency sweep range (GHz) 9.9–12.0 Interface VGA, RS232C, USB, ethernet Light output interface FC/APC Maximum power consumption (W) 100 Project Parameter Working wavelength (nm) 1550 ± 5 Fiber type Single-mode optical fiber (SMF) Maximum dynamic range (dB) 15 Spatial resolution (m) 1 Maximum sampling resolution (m) 0.05 Number of sampling points 20 000 Strain test accuracy (με) ±50(10–20 ns); ±10(50–200 ns) Strain test repeatability (με) ≤ ±100 Strain test range (με) (-15 000)–(+15 000) Test range (km) 0.5, 1, 2, 5, 10, 20, 40, 80 Average number of readings 210–224 Frequency sweep interval (MHz) 1, 2, 5, 10, 20, 50 Frequency sweep range (GHz) 9.9–12.0 Interface VGA, RS232C, USB, ethernet Light output interface FC/APC Maximum power consumption (W) 100 View Large Table 1. Test instrument parameters. Project Parameter Working wavelength (nm) 1550 ± 5 Fiber type Single-mode optical fiber (SMF) Maximum dynamic range (dB) 15 Spatial resolution (m) 1 Maximum sampling resolution (m) 0.05 Number of sampling points 20 000 Strain test accuracy (με) ±50(10–20 ns); ±10(50–200 ns) Strain test repeatability (με) ≤ ±100 Strain test range (με) (-15 000)–(+15 000) Test range (km) 0.5, 1, 2, 5, 10, 20, 40, 80 Average number of readings 210–224 Frequency sweep interval (MHz) 1, 2, 5, 10, 20, 50 Frequency sweep range (GHz) 9.9–12.0 Interface VGA, RS232C, USB, ethernet Light output interface FC/APC Maximum power consumption (W) 100 Project Parameter Working wavelength (nm) 1550 ± 5 Fiber type Single-mode optical fiber (SMF) Maximum dynamic range (dB) 15 Spatial resolution (m) 1 Maximum sampling resolution (m) 0.05 Number of sampling points 20 000 Strain test accuracy (με) ±50(10–20 ns); ±10(50–200 ns) Strain test repeatability (με) ≤ ±100 Strain test range (με) (-15 000)–(+15 000) Test range (km) 0.5, 1, 2, 5, 10, 20, 40, 80 Average number of readings 210–224 Frequency sweep interval (MHz) 1, 2, 5, 10, 20, 50 Frequency sweep range (GHz) 9.9–12.0 Interface VGA, RS232C, USB, ethernet Light output interface FC/APC Maximum power consumption (W) 100 View Large 3. Working area 3.1. Geological conditions of the mining working face The measurements are performed on the first mining face of the 3-1 coal seam. The working face is 2819 m long and 240 m wide. The coal seam dips between 0° and 4°, with an average inclination of 1°, and the working face is a monoclinic structure. The thickness of the coal seam is 1.5–3.5 m, and the average recoverable thickness is 2.5 m. According to the geological data collected during the coal mining, the working face contains two normal faults, which have a negligible effect on the mining. The overall structure of the working face is relatively simple. 3.2. Test system To investigate the characteristics of the overburden damage, we designed a test system in the working face based on the construction conditions. In this test system, two overburden failure monitoring boreholes were installed. The #1 and #2 boreholes are on the face of the auxiliary roadway roof. Figure 2 shows the layout of the #1 and #2 boreholes in plan view. The difference in the orientations of the #1 and #2 boreholes and the working face is 40°. The elevation angles of the #1 and #2 boreholes are 30° and 15°, respectively. Both boreholes are oriented in the mining direction. The #1 borehole was designed to have a depth of 160 m, but the actual installation depth is 145 m. The #2 borehole was designed to have a depth of 140 m, but the actual installation depth is 120 m. Figure 3 shows the borehole profiles. Note that the borehole lengths in the borehole profiles are not the true borehole lengths and instead represent the apparent lengths of the boreholes. Because the cross section of the profile is not parallel to the borehole orientation (the cross section is oriented parallel to the auxiliary transportation roadway of the working face), the borehole lengths in figure 3 are the projections of the true borehole lengths, and the relation between the apparent and actual lengths is described by equation (2). According to equation (2), the apparent length of the #1 borehole is 120.5 m, and the apparent length of the #2 borehole is 94.1 m. l0=(l·cosα·cosβ)2+(l·sinα)2. 2 In equation (2), l0 is the apparent length of the borehole, l is the true length of the borehole, α is the elevation angle of the borehole, and β is the angle between the borehole and the auxiliary transportation roadway of the working face. Figure 2. View largeDownload slide Plan view of the borehole layout. Figure 2. View largeDownload slide Plan view of the borehole layout. Figure 3. View largeDownload slide Cross section of the borehole layout. Figure 3. View largeDownload slide Cross section of the borehole layout. 3.3. Distributed sensor cable installation To install a distributed optical cable in a borehole, the optical cable is attached to the outer wall of a pipe with fasteners. First, the optical cable is attached to the outer wall and forms a catenary with fixed elements. To avoid twisting and entanglement of the optical cables, the fasteners are separated by 1 m. Second, the catheter is extended with a catheter adapter device. Using this method, the optical cable is fixed to the pipe and installed at the target borehole depth. Finally, an orifice gate valve is used to seal the orifice, and the whole borehole is grouted with a special slurry. The grouting material ratio of the slurry depends on the strength of the rock formation and the mechanical rock properties determined by a fracturing test. The grouting is completed section by section from the bottom to the top of the borehole. In each section, a slurry with an appropriate grouting material ratio is injected to ensure the durability of the grouting and the effective coupling between the cable and the surrounding rock; then, the strain fields are obtained (Zhang et al2016, Sun et al2017). During grouting, the borehole is equipped with a pressure grouting device. The grouting is completed when the slurry reaches the top of the borehole. Figure 4 illustrates the installation process of the distributed sensor cable. As shown in figure 5, the installed distributed optical cable is specially designed to monitor deformation in the mining environment. The main structure of this optical cable includes an optical fiber, metal reinforcement and a jacket. When the overburden is deformed or destroyed during mining, the distributed optical cable simultaneously experiences microscopic deformation. The deformation of the surrounding rock can be analyzed using equation (1) to calculate the changes in the surrounding rock. Figure 4. View largeDownload slide Installation diagram for a distributed sensor cable. Figure 4. View largeDownload slide Installation diagram for a distributed sensor cable. Figure 5. View largeDownload slide Schematic diagram of the structure of a distributed sensor cable. Figure 5. View largeDownload slide Schematic diagram of the structure of a distributed sensor cable. 4. Field measurements 4.1. Data collection From 14 July 2016 to 8 August 2016, we collected a total of 25 data sets from the #1 and #2 optical cables. The first group of data was collected when the distance between the working face and the monitoring section was 350 m. The first group of data is treated as the set of initial values in our interpolation. The working face moved forward as mining progressed until the distance between the working face and the monitoring section was -50 m. We also gradually increased the measuring frequency as the mining progressed. In the later mining stages, when the working face was approaching the sensitive monitoring section, we performed intensive data collection and real-time monitoring of the overburden deformation and strain distribution. Throughout the entire real-time monitoring process, we effectively obtained the overburden strain data before the mining face was affected, the deformation of the roof during the coal mining process, and the strain field distribution as the working face approached the monitoring borehole measurement area. Table 2 lists the collected data, including the distance that the working face advanced and the measurement times. In combination with the mining progress data, these measured data allow further exploration and analysis of the strain distribution in the surrounding rocks during the mining process. The measurement accuracy of the failure in the overburden is significantly improved by performing a detailed analysis with the in situ measurements of the optical fiber strain parameters. The entire process of site construction, field measuring and field monitoring lasted approximately one month. Table 2. Data acquisition conditions. Collection number Date Advance distance (m) Distance between working face and orifice (m) C1 14 June 2016 PM 03:01 0 350 C2 15 June 2016 PM 03:05 10 340 C3 14 July 2016 PM 03:00 220 120 C4 16 July 2016 PM 03:00 10 110 C5 19 July 2016 PM 03:00 32 78 C6 20 July 2016 PM 03:06 3 75 C7 21 July 2016 PM 03:10 20 55 C8 22 July 2016 PM 03:11 7 48 C9 23 July 2016 PM 03:03 6 42 C10 24 July 2016 PM 03:03 2 40 C11 25 July 2016 PM 03:00 5 35 C12 26 July 2016 PM 03:00 5 30 C13 27 July 2016 PM 03:00 2 28 C14 28 July 2016 PM 03:05 5 23 C15 29 July 2016 PM 03:00 6 17 C16 30 July 2016 PM 03:03 7 10 C17 31 July 2016 PM 03:10 5 5 C18 1 August 2016 PM 03:20 0 5 C19 2 August 2016 PM 03:00 5 0 C20 3 August 2016 PM 03:00 10 -10 C21 4 August 2016 PM 03:10 10 -20 C22 5 August 2016 PM 03:06 5 -25 C23 6 August 2016 PM 03:00 10 -35 C24 7 August 2016 PM 03:00 5 -40 C25 8 August 2016 PM 03:00 10 -50 Collection number Date Advance distance (m) Distance between working face and orifice (m) C1 14 June 2016 PM 03:01 0 350 C2 15 June 2016 PM 03:05 10 340 C3 14 July 2016 PM 03:00 220 120 C4 16 July 2016 PM 03:00 10 110 C5 19 July 2016 PM 03:00 32 78 C6 20 July 2016 PM 03:06 3 75 C7 21 July 2016 PM 03:10 20 55 C8 22 July 2016 PM 03:11 7 48 C9 23 July 2016 PM 03:03 6 42 C10 24 July 2016 PM 03:03 2 40 C11 25 July 2016 PM 03:00 5 35 C12 26 July 2016 PM 03:00 5 30 C13 27 July 2016 PM 03:00 2 28 C14 28 July 2016 PM 03:05 5 23 C15 29 July 2016 PM 03:00 6 17 C16 30 July 2016 PM 03:03 7 10 C17 31 July 2016 PM 03:10 5 5 C18 1 August 2016 PM 03:20 0 5 C19 2 August 2016 PM 03:00 5 0 C20 3 August 2016 PM 03:00 10 -10 C21 4 August 2016 PM 03:10 10 -20 C22 5 August 2016 PM 03:06 5 -25 C23 6 August 2016 PM 03:00 10 -35 C24 7 August 2016 PM 03:00 5 -40 C25 8 August 2016 PM 03:00 10 -50 Note. C1 represents the first acquisition of data. View Large Table 2. Data acquisition conditions. Collection number Date Advance distance (m) Distance between working face and orifice (m) C1 14 June 2016 PM 03:01 0 350 C2 15 June 2016 PM 03:05 10 340 C3 14 July 2016 PM 03:00 220 120 C4 16 July 2016 PM 03:00 10 110 C5 19 July 2016 PM 03:00 32 78 C6 20 July 2016 PM 03:06 3 75 C7 21 July 2016 PM 03:10 20 55 C8 22 July 2016 PM 03:11 7 48 C9 23 July 2016 PM 03:03 6 42 C10 24 July 2016 PM 03:03 2 40 C11 25 July 2016 PM 03:00 5 35 C12 26 July 2016 PM 03:00 5 30 C13 27 July 2016 PM 03:00 2 28 C14 28 July 2016 PM 03:05 5 23 C15 29 July 2016 PM 03:00 6 17 C16 30 July 2016 PM 03:03 7 10 C17 31 July 2016 PM 03:10 5 5 C18 1 August 2016 PM 03:20 0 5 C19 2 August 2016 PM 03:00 5 0 C20 3 August 2016 PM 03:00 10 -10 C21 4 August 2016 PM 03:10 10 -20 C22 5 August 2016 PM 03:06 5 -25 C23 6 August 2016 PM 03:00 10 -35 C24 7 August 2016 PM 03:00 5 -40 C25 8 August 2016 PM 03:00 10 -50 Collection number Date Advance distance (m) Distance between working face and orifice (m) C1 14 June 2016 PM 03:01 0 350 C2 15 June 2016 PM 03:05 10 340 C3 14 July 2016 PM 03:00 220 120 C4 16 July 2016 PM 03:00 10 110 C5 19 July 2016 PM 03:00 32 78 C6 20 July 2016 PM 03:06 3 75 C7 21 July 2016 PM 03:10 20 55 C8 22 July 2016 PM 03:11 7 48 C9 23 July 2016 PM 03:03 6 42 C10 24 July 2016 PM 03:03 2 40 C11 25 July 2016 PM 03:00 5 35 C12 26 July 2016 PM 03:00 5 30 C13 27 July 2016 PM 03:00 2 28 C14 28 July 2016 PM 03:05 5 23 C15 29 July 2016 PM 03:00 6 17 C16 30 July 2016 PM 03:03 7 10 C17 31 July 2016 PM 03:10 5 5 C18 1 August 2016 PM 03:20 0 5 C19 2 August 2016 PM 03:00 5 0 C20 3 August 2016 PM 03:00 10 -10 C21 4 August 2016 PM 03:10 10 -20 C22 5 August 2016 PM 03:06 5 -25 C23 6 August 2016 PM 03:00 10 -35 C24 7 August 2016 PM 03:00 5 -40 C25 8 August 2016 PM 03:00 10 -50 Note. C1 represents the first acquisition of data. View Large 4.2. Data processing and analysis Using BOTDR, pulsed light transmitted from one end of the optical cable provides a measurement of the strain along the entire optical fiber. If the optical cable breaks at any point along its length, we can still obtain the strain distribution from the transmitting end to the broken point. When the temperature variation in the working environment is maintained within a 5 °C range, the relative strain in the optical fiber caused by temperature variation is negligible. During data acquisition, we used the mine temperature tester to measure the temperature; the results showed that the temperature of the working environment varied within 5 °C. Therefore, temperature correction is not necessary in the data processing step. The optical fiber strain can be directly calculated using equation (1). 4.2.1. Strain distribution characteristics of the optical cables in the boreholes Figures 6(a) and (b) show the strain distributions measured from monitoring boreholes #1 and #2, respectively (for clarity, the data shown in figure 6 are only part of the characteristic curve). Positive and negative values correspond to tensile strain and compressive strain, respectively. The #1 and #2 optical cables generally recorded tensile strain, indicating that the overburden is under tensile conditions within the borehole area. However, the strain of the cable varies with the depth of the borehole. The maximum strain magnitude of the tensile force in the #1 borehole is approximately 5555 με, whereas that of the compressive force is approximately -865 με. The maximum tensile strain of the #2 borehole is approximately 6707 με, while the maximum compressive strain is approximately -540 με. The #1 borehole was closer to the working face than the #2 borehole, so the top of the #1 borehole was preferentially affected by the stress of the working face, resulting in the corresponding strain characteristics. With the continuous advancement of the working face, the #2 monitoring borehole became increasingly affected, which led to strain-related structural changes in the overburden. After the coal seam was excavated, the overburden was subjected to the self-weight effect, forming a caved zone, a fractured zone and a bending subsidence zone. Additionally, the deformation and displacement of the overburden caused the optical cables to become stretched or compressed. When the deformation displacement was large, a breakpoint formed. Based on the analysis of the optical cable strain test results at different depths within the borehole, we can determine the height of the caved zone and fractured zone during our observation period. Figure 6. View largeDownload slide Strain distribution of the optical cables. Figure 6. View largeDownload slide Strain distribution of the optical cables. 4.2.2. Analysis of the strain characteristics of the overburden Figure 7 shows the changes in the strain from the #1 and #2 optical cables over time. With the orifice as the origin, the x-axis represents the length of the sensor cable within the borehole, and the y-axis represents the strain magnitude. Overall, this figure reflects the variation in the strain with the advancement of the working face. Figure 7. View largeDownload slide View largeDownload slide Borehole optical cable strain curves. Figure 7. View largeDownload slide View largeDownload slide Borehole optical cable strain curves. Figure 7 shows that due to their spatial differences, the #1 optical cable and the #2 optical cable exhibit substantially different characteristics as the overburden deforms. The strain magnitudes of the #2 borehole optical cable are greater than those of the #1 borehole, but the strain sensing time of the #2 borehole lags behind that of the #1 borehole. During the measurement process, most of the #2 optical cable is located within the theoretical caved and fractured zones, so the #2 optical cable is the first to exhibit breakage of the sensor cable. With the continuous advancement of the working face, four breakages occur in the #2 borehole. The fracture position and the strain distribution characteristics of the optical cable provide robust information on the stratification of the caved zone and fractured zone during the mining process. In figure 7(a), the distance between the working face and orifice is 120 m. Due to the effect of advancing stress, overburden deformation in the form of tensile strain occurs at depth of 103.8–120.5 m in #1 borehole; however, the strain is only 300 με. Based on the sensitivity of the optical cable to deformation, we conclude that the deformation characteristics of the optical cable in the #1 borehole do not exhibit any evidence of elasto-plastic deformation in the overburden, and only a minor disturbance of the overburden is detected. The overburden structure remains relatively intact and impermeable. Meanwhile, the strata controlled by the #2 borehole do not incur obvious deformation. As mining of the coal seam continues to slowly advance toward the monitoring section, as shown in figure 7(b) (i.e., when the working face is 78 m from the orifice), the coal seam in the lower part of the depth range 99.7–120.5 m in the #1 borehole and the depth range 76.8–94.1 m in the #2 borehole had already been mined. The strain in the #1 optical cable in the 103.8–120.5 m depth section continues to increase, reaching a maximum strain value of 1798 με at 114.6 m. In addition, tensile strain is also present at the depth of 62.4–91.4 m in the #1 borehole. Momentarily, there is a subtle compressive strain at the top of the depth range 76.8–94.1 m in the #2 borehole. When the working face is 55 m from the orifice, as shown in figure 7(c), half of the #1 and #2 boreholes are above the mined-out area. The tensile strain along the depth range of 103.8–120.5 m in the #1 borehole decreases overall because the collapse of the lower rock causes movement of the upper rock, and then the upper rock is compressed again. At a depth of 81.8 m, the #1 borehole exhibits greater deformation, with a strain of 2785 με. However, the adjacent area has not yet deformed considerably. The cross-sectional view shows that the lithology corresponding to a depth of 81.8 m is an interface between a mudstone and a sandstone. Therefore, separation of the rock strata may occur at this interface. Simultaneously, the #2 borehole optical cable incurs the first fracture, and the breakpoint strain reaches 4786 με at a depth of 74.5 m in the borehole. Generally, the breakpoint of an optical cable is within the caved zone. However, the accurate caved zone position results at this time need to be combined with the later data for comprehensive analysis. In addition, the tensile strain in the range of 49.8–74.5 m in the #2 borehole increases. In figure 7(d), the distance between the working face and orifice is 30 m. The maximum strain is 4914 με at a depth of 80.2 m in the #1 borehole, and the strain at 81.8 m does not differ from that shown in figure 7(c). In addition, the typical delamination development occurs at a depth of 74.8–111.3 m in the #1 borehole. Furthermore, the depths of 0–56.1 m in the #1 borehole also present tensile strain. However, the tensile strain along 56.1–74.8 m in the #1 borehole is significantly less than that shown in figure 7(c). Additionally, the second fracture in the #2 optical cable occurs at a depth of 51.0 m, corresponding to a strain of 2630 με. The second breakpoint is near the interface between a sandstone and a mudstone and is probably due to rock collapse. In figure 7(e), the working face is 0 m from the orifice. The optical cable in the #1 borehole still shows the largest deformation at a depth of 80.2 m, and relatively large deformation occurs from 52.3 to 80.2 m. Meanwhile, the extent of delamination development gradually expands from 74.8–111.3 m to 52.3–111.3 m. In addition, the maximum strain point (at a borehole depth of 80.2 m) should correspond to the top interface of a developing fractured zone; thus, the vertical height of the fractured zone is 48.5 m from the coal seam. However, this is not the final height because the rock continues to move. The third breakpoint in the #2 borehole occurs at a depth of 34.3 m, with a strain of 873 με. 4.2.3. Determining the height of the two zones within the overburden From the above analysis, as shown in figure 7, we have concluded that the top of the fractured zone is 48.5 m from the coal seam. To determine the heights of the fractured and caved zones as they develop, we analyze the strain characteristics of the working face when it is -50 m from the orifice. At this moment, the overburden above the mined-out area is not affected by the coal mining, so the deformation and failure in the overburden are stable; additionally, the measured heights of the two zones should be more accurate. Figure 8 shows that the optical cable in the #1 borehole at a depth of 52.3–111.3 m exhibits typical delamination development. Additionally, a relatively high strain value occurs at 79.8 m. Based on the strain extremes and geological data, we conclude that a fractured zone developed at 48 m, i.e., within the sandstone. The variation in strain at depths of 28.2–49.8 m in the #1 borehole is subtle. Therefore, this layer is the key layer. The key layer of overburden moves as a whole, and no damage occurs within this layer. In addition, the bottom of the key layer may be the top interface of the caved zone. Therefore, the vertical height of the caved zone above the coal seam is most likely 17 m. Additionally, the second breakpoint in the #2 borehole occurs at a depth 51.0 m, which is also a vertical height of 17 m. Therefore, through the comprehensive analysis of the location of the key layer in the #1 borehole and the breakpoint of the #2 borehole optical cable, the top of the overburden caved zone above the coal seam is determined to be 17 m. The fourth breakpoint occurs at a depth of 33.3 m in the #2 borehole, with a strain value of 4043 με, and this breakpoint is in the caved zone. Therefore, based on the analysis of the strain distributions along the #1 and #2 borehole optical cables and the geological data, we conclude that the caved zone extends 17 m above the coal seam and that the fractured zone extends to 48 m above the coal seam. The working face of the 3-1 coal layer has a minable thickness of 2.5 m. Therefore, the ratio between the thicknesses of the caved zone and the minable layer is 6.8, and the ratio between the thicknesses of the combined caved zone and fractured zone and the minable layer is 19.2. Figure 8. View largeDownload slide The borehole optical cable strain curve measured at a working face distance of -50 m from the orifice. Figure 8. View largeDownload slide The borehole optical cable strain curve measured at a working face distance of -50 m from the orifice. 4.2.4. Force analysis of overburden in the direction of coal seam mining During working face mining, due to the influence of the additional mining stress, the stress distribution law and plastic failure style of the overburden vary. After long-term research, experts and scholars at home and abroad have concluded that the stress distribution in the vertical cross section of the overburden exhibits a saddle shape that extends in the direction of the coal seam mining (Zhang 2007, Li 2012), as shown in figure 9. Based on the force analysis of the overburden in figure 9, we conclude that the saddle-shaped frontal zone I will be the first to be affected by advancement-related stress, deform and fail. Because of the support of overburden, zone II will be protected from the stress and will not undergo deformation and failure at this stage. However, with the advancement of the working face, the stress will gradually be transmitted forward, leading to periodic changes in the stress in the overburden. By comparing the strain data from the boreholes, we have discovered that because the angle of the #1 borehole is greater and more aligned with the mining face than that of the #2 borehole, the stress of these two boreholes exhibit a hysteresis. This finding demonstrates that our test outcome can be combined with the saddle shape theory. Moreover, in figure 6, the strain of the low-angle borehole is higher than that of the high angle borehole in some layers. This pattern arises because the low-angle borehole is mainly located in the caved zone, in which the overburden stress is considerable and causes extensive failure. Therefore, we conclude that the stress state of the optical cable is correlated with the parameters of the borehole location. Figure 9. View largeDownload slide Stress distribution with deformation and failure in the overburden during mining of a coal seam. Figure 9. View largeDownload slide Stress distribution with deformation and failure in the overburden during mining of a coal seam. 4.3. Borehole resistivity CT 4.3.1. Testing system and principle of resistivity CT To improve the accuracy of the test results using distributed optical fibers, we arranged an observation system to perform borehole resistivity CT in the #1 and #2 boreholes in the detection zone, as shown in figure 10(c). We utilized the resistivity method to evaluate the reaction of the geoelectric field in the overburden under the influences of coal seam mining. Borehole resistivity CT is a geophysical prospecting method that uses borehole electrodes in a detection zone to create a point-source emission zone or measurement zone from which to obtain the spatial distribution of the electric field. A CT test involves arranging a certain number of electrodes within two boreholes, and these electrodes serve as the emission sources and receivers. The receivers record the electric potential values, which are used to construct an image of the differences in the physical properties of the media between the two boreholes. These data can then be used to assess the failure characteristics of the media, as shown in figure 10(b). We used a high-density electrical method apparatus to collect the data. The apparatus, shown in figure 10(a), contains a power box and a collector that can collect 64 channels simultaneously. Figure 10. View largeDownload slide Image of the resistivity CT test system (a): acquisition instrument and parameters; (b): schematic diagram of the resistivity CT method; (c): resistivity CT test system in a roof borehole. Figure 10. View largeDownload slide Image of the resistivity CT test system (a): acquisition instrument and parameters; (b): schematic diagram of the resistivity CT method; (c): resistivity CT test system in a roof borehole. 4.3.2. Resistivity CT data collection and analysis of the results As shown in figure 10(c), we arranged 32 electrodes in the #1 and #2 boreholes. The resistivity data were collected at the same time as the BOTDR test was performed, i.e., starting on 14 July 2016 and ending on 8 August 2016. In total, 25 sets of data were collected. Similarly, we define the data collected on 14 July as the starting data and compare it with later data to analyze the changes over time. Based on the analysis of field measurements of the electrical characteristics of the overburden deformation and failure, Zhang et al (2009) found that the resistivity in an overburden failure zone is several to ten times higher than that in a normal field. We take P S Zhang's findings as the electrical property standards for both zones and use these values to determine the heights of the caved zone and fractured zone above the 3-1 coal seam. Representative resistivity data are shown in figure 11. Figure 11. View largeDownload slide Distribution of the overburden resistivity (a): distribution of the background resistivity in the overburden; (b): working face distance from the orifice is 0 m; (c): working face distance from the orifice is -50 m; (d): resistivity color scale. Figure 11. View largeDownload slide Distribution of the overburden resistivity (a): distribution of the background resistivity in the overburden; (b): working face distance from the orifice is 0 m; (c): working face distance from the orifice is -50 m; (d): resistivity color scale. As shown in figure 11(a), when the distance between the working face and orifice is 120 m, the resistivity magnitudes in the borehole resistivity profiles are relatively low. The average resistivity is approximately 100 Ω m, with local values reaching 300 Ω m. The working face gradually advances, and the working face is more than 0 m from the orifice in figure 11(b). Relatively stable high-resistivity zones are located from 0 to 19.5 m and from 30 to 51 m above the coal mining face. In these sections, the resistivity is greater than 800 Ω m and exhibits a clear vertical distribution. In figure 11(c), the working face is 50 m past the detection borehole. At this moment, the high-resistivity zones in the detection zone far from the orifice have stabilized, the vertical height of the local high-resistivity zone has decreased to some extent, and the resistivity in the high-resistivity zone remains over 800 Ω m at a height of 50 m in the overburden. The other zone of high-resistivity is located consistently at a vertical height of approximately 17.5 m. Additionally, the resistivity at a vertical height between 17.5 and 34 m is relatively unchanged, indicating that this zone is a key layer. Therefore, by using the borehole resistivity CT method, we have determined that the tops of the caved zone and the fractured zone in the overburden are 17.5 m and 50 m above the 3-1 coal seam, respectively. 4.4. Synthesized determination of the heights of the two zones To determine the heights of the deformation and failure zones in the overburden, we utilized both testing techniques, DOFS and borehole resistivity CT. This analysis indicates that the results of the resistivity CT are basically consistent with those of the DOFS within error. However, the two-dimensional resistivity has a remarkable volume effect in the inversion process, which tends to extend the abnormal zones (Yan 2009). After a comprehensive analysis, we concluded that the tops of the caved zone and fractured zone in the overburden are 17 m and 48 m above the 3-1 coal seam, respectively. 5. Conclusion During coal seam mining, the failure of the overburden is affected by many factors, such as the coal seam depth, thickness and lithology. Based on the spatial and temporal variations in the strain along two borehole optical cables, we can clearly identify the progression of structural deformation and failure and the development of fractures in the overburden. Measurement data from different periods can be used to analyze the basic trends in the deformation and failure of the overburden through time. According to certain characteristics of the strain distribution in the two boreholes (the maximum strain, the breakpoints of the sensing optical cable, etc), the caved zone and the fractured zone extend 17 m and 48 m above the coal seam, respectively. Hence, the ratio between the caved and mining zone thicknesses is 6.8, and the thickness ratio between the combined caved and fractured zone and the mining zone is 19.2. The test results of the DOFS are basically consistent with those of the resistivity CT within error. The characteristics of the strain distribution in the optical cables in the two boreholes are closely related to the orientations of the boreholes. The strain in the borehole at a low-angle to the corresponding horizon is greater than that at a high angle, and the data from the higher-angle optical cable exhibits a hysteresis. Furthermore, large deformation is more likely to occur at lithologic interfaces in the boreholes and may sever the optical cable. The use of BOTDR with a distributed optical cable provides only a one-dimensional perspective. For different working face and overburden conditions, a full-space three-dimensional observation system is needed to obtain further insights through measurement and practical operation of the mining working face. Acknowledgments The authors acknowledge the support of the Anhui University of Science and Technology Graduate Innovation Fund Project (No. 2017CX2002), the Major Project of the Natural Science Research through the Department of Education Anhui Province Science Research Project (No. KJ2016SD17), the Scientific Research Activities of Academic and Technological Leaders of Anhui Province (No. 2016D079) and the Key Research and Development Plan Projects of Anhui Province (No. 1804a0808213). References Bernini R , Minardo A , Zeni L . , 2012 Distributed sensing at centimeter-scale spatial resolution by BOFDA: measurements and signal processing , IEEE Photonics J. , vol. 4 (pg. 48 - 56 ) https://doi.org/10.1109/JPHOT.2011.2179024 Google Scholar Crossref Search ADS Chen J L , Yu S J . , 2000 Simulation experiment on the response of resistivity to deformation and failure of overburden , Chin. J. Geophys. , vol. 43 (pg. 699 - 706 ) Hamdan H , et al. , 2014 Contribution of electrical tomography methods in geotechnical investigations at Mavropigi lignite open pit mine , Environ. Earth Sci. , vol. 72 (pg. 1589 - 1598 ) https://doi.org/10.1007/s12665-014-3063-6 Google Scholar Crossref Search ADS Horiguchi T , Kurashima T , Tateda M . , 1989 Tensile strain dependence of Brillouin frequency shift in silica optical fibers , IEEE Photonics Technol. Lett. , vol. 1 (pg. 107 - 108 ) https://doi.org/10.1109/68.34756 Google Scholar Crossref Search ADS Kang Y H , Wang J Z , Kong F M , Liu X E , Jin R C , Zheng J Q . , 2002 Bore hole survey method for overburden failure , Coal Sci. Technol. , vol. 30 (pg. 26 - 28 ) https://doi.org/10.13199/j.cst.2002.12.26.kangyh.009 Li S Z . , 2012 Study on mining-induced fractures field evolution and gas migration rule , M.S. Thesis Chongqing University, Chongqing, China Liu T Q . , 1995 Influence of mining activities on mine rock mass and control engineering , J. China Coal Soc. , vol. 20 (pg. 1 - 5 ) https://doi.org/10.13225/j.cnki.jccs.1995.01.001 Lu Y Z . , 2002 Study on overlying strata movement law based on coal gangue paste cushion filling method , M.S. Thesis Central South University, Changsha, China Ohno H , Naruse H , Kihara M . , 2001 Industrial applications of the BOTDR optical fiber strain sensor , Opt. Fiber Technol. , vol. 7 (pg. 45 - 64 ) https://doi.org/10.1006/ofte.2000.0344 Google Scholar Crossref Search ADS Piao C D , Shi B , Zhu Y Q , Lu Y . , 2009 Experimental study on BOTDR temperature compensation in bored pile detection , J. Disater Prevention Mitigation Eng. , vol. 29 (pg. 162 - 164 ) https://doi.org/10.13409/j.cnki.jdpme.2009.02.019 Qian M G , Miao X X , Xu J L . , 1996 Theoretical study of key stratum in ground control , J. China Coal Soc. , vol. 21 (pg. 225 - 230 ) Shen B H , Kong Q J . , 2000 Exploration and research on overlay broken law on mechanization caving mining face , Coal Geol. Explor. , vol. 28 (pg. 42 - 44 ) Shi B , Xu H Z , Zhan D , Ding Y , Cui H L , Chen B , Gao J Q . , 2004 Feasibility study on application of BOTDR to health monitoring for large infrastructure engineering , Chin. J. Rock Mech. Eng. , vol. 23 (pg. 493 - 499 ) Song Z Q . , 1988 , Practical Ground Pressure XuZhou China University of Mining and Technology Press State Administration of Work, State Administration of Coal Mine Safety . , 2011 , Coal Mine Safety Regulations Beijing Coal Industry Publishing House State Bureau of Coal Industry . , 2000 , Regulations of Buildings, Water, Rail Way and Main Well Lane Leaving Coal Pillar and Press Coal Mining Beijing China Coal Industry Publishing House Sun B Y , Zhang P S , Fu M R , Xu S A . , 2017 Fiber optic test and results of failure law of floor strata in coal mining site , J. Hefei Univ. Technol.(Nat. Sci.) , vol. 40 (pg. 701 - 707 ) https://doi.org/10.3969/j.issn.1003-5060.2017.05.025 Sun J , Wang L G , Hou H Q . , 2012 Application of micro-seismic monitoring technology in mining engineering , Int. J. Min. Sci. Technol. , vol. 22 (pg. 79 - 83 ) https://doi.org/10.1016/j.ijmst.2011.06.007 Google Scholar Crossref Search ADS Sun Y J , Xu Z M . , 2008 Research on compositive geophysical methods in overburden rock failure detection for mining under Xiaolangdi reservoir , Coal Saf. , vol. 28 (pg. 25 - 28 ) Wu R X , Zhang W , Zhang P S . , 2012 Exploration of parallel electrical technology for the dynamic variation of caving zone strata in coal face , J. China Coal Soc. , vol. 37 (pg. 571 - 577 ) https://doi.org/10.13225/j.cnki.jccs.2012.04.016 Wu S L . , 2002 Movement evolution of failured overburden influenced obviously to the ground pressure around working , PhD Dissertation Shandong University of Science and Technology, Qingdao, China Xu S A , Zhang P S , Zhang D , Wu R X , Guo L Q . , 2015 Simulation study of fiber optic monitoring technology of surrounding rock deformation under deep mining conditions , J. Civ. Struct. Health Monit. , vol. 5 (pg. 1 - 9 ) https://doi.org/10.1007/s13349-015-0125-8 Google Scholar Crossref Search ADS Yan J F . , 2009 Research on application of multi-electrodes resisitivity imagining survey , M.S. Thesis Jilin University, Jilin, China Yang Z J . , 2016 Study on development law of water conducted zone in fully-mechanized mining face with 7 m mining height , Coal Sci. Technol. , vol. 44 (pg. 61 - 66 ) Zhang D , Wang J C , Zhang P S , Shi B . , 2016 Internal strain monitoring for coal mining similarity model based on distributed fiber optical sensing , Measurement , vol. 97 (pg. 234 - 241 ) https://doi.org/10.1016/j.measurement.2016.11.017 Google Scholar Crossref Search ADS Zhang D , Zhang P S , Shi B , Wang H X , Li C S . , 2015 Monitoring and analysis of overburden deformation and failure using distributed fiber optic sensing , Chin. J. Geotech. Eng. , vol. 37 (pg. 952 - 957 ) https://doi.org/10.3969/j.issn.1003-5060.2017.05.025 Zhang H X , Zhang G Y , Xu Y C . , 2005 New development on probing and measuring technology for failed cracking of overburden rock , Coal Sci. Technol. , vol. 33 (pg. 60 - 62 ) Zhang P S , Hu X W , Liu S D . , 2011 Study on dynamic detection simulation of overburden failure in model workface , Chin. J. Rock Mech. Eng. , vol. 30 (pg. 78 - 83 ) Google Scholar Crossref Search ADS Zhang P S , Liu S D , Wu R X . , 2004 Observation of overburden failure of coal seam by CT of sesmic wave , Chin. J. Rock Mech. Eng. , vol. 23 (pg. 2510 - 2513 ) Zhang P S , Liu S D , Wu R X , Cao Y . , 2009 Dynamic detection of overburden deformation and failure in mining workface by 3D resistivity method , Chin. J. Rock Mech. Eng. , vol. 28 (pg. 1870 - 1875 ) Zhang P S , Sun B Y , Xu S A . , 2016 Simulation research on BOTDR-based monitoring over temperature field of water inrushing from coal floor , J. Chongqing Jiaotong Univ.(Nat. Sci.) , vol. 35 (pg. 28 - 31 ) https://doi.org/10.3969/j.issn.1674-0696.2016.05.07 Zhang W Q , Zhang H R , Xu F J . , 2000 Continuous exploration for the mining failure form of the incline and thin coal seam's floor under the high depth , Coal Geol. Explor. , vol. 28 (pg. 39 - 42 ) Zhang Z Y . , 2007 Dynamic analysis on stability of hydraulic powered support in deep inclined fully mechanized wall and prevention slipsm easures , J. China Coal Soc. , vol. 32 (pg. 1 - 5 ) https://doi.org/10.13225/j .cnki .jccs .2007.07.007 © 2018 Sinopec Geophysical Research Institute
TI - Dynamic detection and analysis of overburden deformation and failure in a mining face using distributed optical fiber sensing
JF - Journal of Geophysics and Engineering
DO - 10.1088/1742-2140/aad1c6
DA - 2018-12-01
UR - https://www.deepdyve.com/lp/oxford-university-press/dynamic-detection-and-analysis-of-overburden-deformation-and-failure-r4PAF7lYuj
SP - 2545
VL - 15
IS - 6
DP - DeepDyve
ER -