The Influence of Different Sublevel Heights on the Stability of Faults under Sublevel-Filled Synergistic Mining

Abstract

At present, when addressing the problem of fault stability in mines, most attention is paid to the study of the impact of a single mining method on faults. In order to study fault stability in mines more comprehensively, this paper researches the effect of coordinated mining using multiple mining methods. For example, the sublevel caving method without the sill pillar and the lower-layer filling method of coordinated mining can be used to analyze the dynamic response law of the fault, as well as the stability of the fault in different mining conditions. In this paper, the stress field, displacement field and sliding trend index of the fault plane and orebody near the fault are obtained via numerical simulation and theoretical analysis methods, and the main factors affecting fault stability under different mining conditions are analyzed. The results show that under the influence of cooperative mining, the shear stress fluctuation of the fault surface and the ore body near the fault increase gradually with the sublevel height of the sublevel caving method without the sill pillar and the lower-layer filling method, and the indexes of slip tendency become larger, which may be a precursor of fault activation. In particular, the fault surface of the sublevel caving method without the sill pillar reflects the drastic change in the shear stress with the larger displacement, and the instability brought to the fault by the sublevel caving method without the sill pillar is greater than that brought to the fault by the lower layer filling method under the premise of only changing the height of the sublevel.

1. Introduction

With the development of society and improvements in the quality of life, humankind’s demand for energy is growing, and high-quality mineral resources are constantly being exploited. However, with the increasing depth of mining, the incidence of mining disasters is also increasing. Metal deposits are generally distributed in mountainous areas, so the geological formations encountered in the mining area tend to be well developed, and faults are widely distributed. The existence of these faults destroys the continuity and integrity of the rock mass around the mining area. Faults have an impact on roadway support and may also limit the mining scope and the direction of mining of the deposit, which is an important factor affecting the safety of underground mining in metal mines and is, therefore, of great significance in studying the effect of mining disturbances on faults [1,2].

In different geological conditions, mining method selection is strictly controlled because different tectonic geological conditions influence the selection of different mining methods. Currently, the methods commonly used in mining are the stratified filling method, the sublevel caving method without the sill pillar, the room and pillar mining method, and the natural caving method [3,4,5]. Considering various needs, such as economy and safety production, there has also been extensive research on the optimization of structural parameters for different mining methods [6,7]. The complex tectonic geological background of the mining area is usually reflected in the variety of mining methods used in a synergistic form of operation. There are also abundant research results on the effects of different mining methods and co-mining methods on the instability of surrounding rock, high-level pillars, roof and overburden [8,9,10]. This paper takes the Gansu Jinchang Longshou Mine West Second Mining Area as an example for related research. The Jinchang Longshou Mine West Second Mining Area is a typical sharp-inclined metal mine where the sublevel caving method without the sill pillar and the lower-layer filling method are currently used for coordinated mining. An F8 fault exists in the mining area, which is the largest fault in the Longshou Mine. This special geological structure increases the risk of a mine disaster, and the dynamic response of the fault changes under the disturbance of mining, which makes instability slipping more likely, with the decrease in stability resulting in the collapse of the roof of the overlying rock, which damages the surrounding rock and seriously affects production safety. Domestic and international scholars have produced a wealth of research on the impact of mining disturbances on the stability of faults. W. F. Brace [11] used an experimental rock friction slip test as the basis for analyzing the viscous-slip characteristics of fault sliding in the seismic process, and their research was the beginning of the “fault viscous-slip destabilization theory”. Xun Z and Wang Tongxu et al. used numerical simulation to analyze the stress change in surrounding rock near the fault and concluded that the spread of rock stress became discontinuous at the interface between the fracture surface and the coal seam, which can easily lead to the accumulation of deformation energy in the coal body and the rapid release of gas pressure, thus causing an eruption event [12,13]. Wang H et al. carried out on-site physical tests and numerical simulations and found that rapid changes to the top plate anchor loading, roof separation, and roadway deformation may be related to the sudden destabilization of the fault structure [14]. Jiang Y et al. studied the evolution characteristics of the stress and displacement of surrounding rock, and faults under mining action are studied by using simulation experiments and numerical simulations, with the activation mechanism of the fault being analyzed [15,16,17,18,19,20]. Sainoki A et al. performed a dynamic numerical analysis of a model of faults running parallel to a plate-like deposit. The analysis results showed that the maximum dynamic shear displacement increment induced by quarry mining is significantly affected by the fault friction angle, the mining depth, and the position of the fault relative to the ore body, while the fault stiffness and expansion angle have little effect [21]. Wang H et al. investigated the characteristics of overlying strata collapse and mining pressure on a fault-influenced zone by creating a physical experimental model [22]. Recently, Sainoki and Mitri, 2014a; Sainoki and Mitri, 2014b; and Sainoki and Mitri, 2015, conducted a series of studies to model the effects of fault slip using various methods. The effects of mining depth, friction angle, expansion angle, and fault stiffness on fault slip were examined. In addition, the fault surface roughness and slip-weakening behavior were investigated, and the fault slip strength was estimated by using FLAC3D [23,24,25,26]. Vishal et al., 2012, investigated the value of critical velocity obtained by transforming the stick–slip motion into a steady motion through a series of rock friction experiments. The relationship between the coefficient of kinetic friction, coefficient of static friction, and the increase in the velocity of motion was obtained [27,28]. Xuebo Z et al. investigated the influence mechanism of different geological parameters on the activation of small faults in deep coal mining and face mining by using numerical simulation [29]. Guojian C et al. carried out a systematic superposition dynamic normal perturbation shear test on the surface of saw-cutting bare granite by utilizing a multifunctional shear test device developed independently to reveal the activation mechanism and characteristics of faults under the action of a certain stress environment and the external dynamics of the normal perturbation load [30]. Zhou Yang et al. [31] derived the mining stress evolution law when the working face crosses the faults through FLAC3D. Wang Zhanshuo [32] used the Ritz method to calculate and determine the bending deformation deflection function equations of the top plate of two types of faults and carried out the weakened decompression zoning for the area near the faults through numerical simulation. Song Weihua et al. [33] conducted sandstone double-shear friction tests to determine the sliding criterion of sandstone and found the fault sliding criterion of the Taiji Coal Mine. For the problem of fault stability caused by mining disturbances, scholars at home and abroad have used a variety of methods to analyze and study the problem of fault stability, such as numerical models, mechanical models, engineering verification and theoretical analysis. The evolution law of stress during mining, the influence of various fault parameters and structural parameters on fault stability, and the fault slip criterion are obtained. However, the geological structure of the mining area is complex, and it is often impossible to fully consider a variety of factors when conducting research and analysis. The stability of the fault is not only linked to one parameter; therefore, it is necessary to consider the comprehensive situation.

At present, most research has focused on the influence of a single mining method on fault stability and few research works have been conducted on the change rule of the dynamic response of faults under coordinated mining methods. In terms of numerical simulation, most of the studies reflect mining in the ore body but rarely reflect the real ore body, the fault model and the real mining situation. In order to further study the impact of mining on fault stability, in this paper, we take the West Second Mining Area of Jinchang Longshou Mine as the background; this is a typical sharply inclined metal mine, which has adopted the sublevel caving method without the sill pillar and the lower-layer filling method for coordinated mining. There are fault structures in the mining area, which will increase the risk of mining disasters. Using numerical simulation methods, different mining structure parameters are set to study the dynamic response analysis of faults under the disturbance of composite mining using the sublevel caving method without the sill pillar and the lower-layer filling method and to analyze the change rule of the stability of the faults by observing the changes in the shear stress, displacement and other parameters of the fault surface and the vicinity of the faults during composite mining. This provides a scientific basis for engineering under similar mining conditions and geological conditions and has certain reference significance.

2. Overview of the Study Area

2.1. Tectonic Geological Background

The Longshou Mine West Second Mining Area belongs to Jinchang City, Gansu Province, in western China, geographically located in the western Heshi corridor and situated in the northern foothills of the eastern Longshou Mountain, the boundary of the Alashan Plateau. The mining area, in general, belongs to the area where the Alxa Plateau and Qilian Plateau edge make contact, and it not only has the significant geological characteristics of the Plateau area, but also the folded area of strong tectonic influence that characterizes the North Qilian Mountains. The mine has experienced Indo-Chinese, Yanshan, Xishan and other phases of tectonic movement, accompanied by strong magma intrusion activities, resulting in the development of a very complex mine fracture structure. The mine is located at the turning point of the Longshoushan uplift tectonic line from east–west to north–west, and the general tectonic features are characterized by a simple fold pattern, a developed fracture structure and a complex nature. The tectonic condition of the Longshou Mine is complicated, and its biggest fault is F8, as shown in Figure 1. The left picture of Figure 1 shows the entire Jinchuan copper–nickel deposit, the right picture of Figure 1 shows Longshou Mine, and the red box area in the right figure is the West Second Mining Area of Longshou Mine. The F8 fault, which is a translational reverse/thrust fault, is distributed on the southeast side of the mining area, extending about 5.3 km in a north–east–south–west direction, trending to the southeast, with an inclination angle of about 80–85°, and the width of the fault fracture zone is 15–32 m [34,35,36,37].

2.2. Mining Status of the Mining Area

The West Second Mining Area of the Longshou Mine was completed in 2010 and subsequently put into production with a design capacity of 1.65 million t/a, a design ore dilution rate of 7%, a loss rate of 5% and a service life of 20 years [38]. The West Second Mining Area of Longshou Mine was originally designed to adopt the natural caving method. At a late stage of the project construction, due to changes in the economic situation and for other reasons, it was finally decided to change the mining method used at the West Second Mining Area to the lower-layer filling method. The height of the segments is an important structural parameter and an important indicator of the amount of work to be carried out, as is the stability of the roadway and the faults, and multiple segments make up a central section. The lower-layer filling method was formally adopted in the West Second Mining Area in 2012, and the elevations of the two middle sections were 1554 m and 1430 m, and the elevations of the first mining section were 1630 m and 1534 m. The horizontal middle sections of the West Second Mining Area are 1430 m, 1554 m and 1640 m and are composed of multiple sections, with the mining area designed to be simultaneously mined in two middle sections, namely, at 1640 m and 1554 m. In March 2016, due to the influence of the mine-control fault F8 in the mining area, the elevation of the 1554 m middle section, which was east of the eight rows of the 1610 m level and the overburden, was changed to 1610 m. The collodion-filling body and overburdened rocks underwent a large overall collapse, and parts of the production activities went into a state of shutdown.

In 2018, we carried out an industrialized test of the sublevel caving method without the sill pillar, including research on its supporting system in the West Second Mining Area of Longshou Mine, and divided the remaining ore body in the 1554 m middle section into two stages for mining. The first stage was the test stage, with a mining ore body east of the 1595 m level for eight rows; the second stage was the promotion stage, where after the success of the test stage, the sublevel caving method without the sill pillar could be extended to the mining of all the remaining ore bodies in the 1554 m middle section. In April 2020, the first stage of the test work was completed, the expected indicators were reached, and the start of the second stage was planned. The actual height of the sublevel caving method without the sill pillar was 15 m. The specific structural parameters of the caving method are shown in Figure 2, and the height of the lower-layer filling method was 20 m [34]. The relative position of the ore body, the fault and the exploration line division are shown in Figure 3, below.

3. Model Building and Validation

3.1. Model Building

In this paper, the West Second Mining Area of Jinchang Longshou Mine was taken as the engineering background, and a large-scale 1:1 true 3D numerical model was built by using multiple software modeling methods, reflecting the real relative position of the ore body and the fault as well as the spatial distribution status of the two mining methods. The refined numerical simulation was used to analyze the changes in stress and displacement fields of the fault surface and the ore body near the fault, as well as the surface displacement and movement patterns caused by mining under different working conditions. We utilized AutoCAD 2020, the 3D modeling function of Rhino 7, the powerful meshing function of ANSYS 2021R2, and the excellent computational function of FLAC3D 6.0 to ensure the reliability of the results (Figure 4). Due to the complexity of modeling and the limited data available, the modeling in this paper focused on the main factors affecting the stability of the roadway, including the geological structure, fault morphology, ground stress state and strength parameters.

The size of the model was 3000 m (x) × 2300 m (y) × 1050 m (z), the number of cells was 862,524, the number of nodes was 147,117 and the Moore–Cullen strength criterion was adopted.

The rock mass in this simulation is an ideal elastoplastic stress–strain model, and the Moore–Cullen yield criterion describes:

$$f_s={\sigma }_1-{\sigma }_3\frac{1+\mathit{sin}\phi }{1-\mathit{sin}\phi }-2c\sqrt{\frac{1+\mathit{sin}\phi }{1-\mathit{sin}\phi }}$$

(1)

Inside the official: ${\sigma }_1$ and ${\sigma }_3$ are the maximum principal stress and the minimum principal stress. $c$ is the adhesion force, and $\phi$ is the friction angle. When >0, shear failure occurs.

The surface of the calculation model is a free boundary, which constrains two X-axis displacements perpendicular to the X-axis direction and two Y-axis displacements perpendicular to the Y-axis direction. Gradient stress is applied to the model boundary according to the actual ground stress:

$$\left\{\begin{array}{l}{\sigma }_1=2.13+0.037H \mathrm{M}\mathrm{P}\mathrm{a}\\ {\sigma }_2=1.12+0.0159H \mathrm{M}\mathrm{P}\mathrm{a}\\ {\sigma }_3=1.106+0.032H \mathrm{M}\mathrm{P}\mathrm{a}\end{array}\right.$$

(2)

where ${\sigma }_1$, ${\sigma }_2$ and ${\sigma }_3$ are the maximum principal stress, minimum principal stress and vertical principal stress (MPa), respectively; H is the depth (m).

The model is shown in Figure 5. The yellow part of Figure 5 shows the filling body, and the cemented backfill is usually composed of aggregate, gelling agent, water and other components, and the mechanical properties of the backfill are significantly improved due to the addition of gelling agent. The internal friction angle of the backfill is 38°, the cohesion of the backfill is 0.95 Mpa, and the tensile strength of the backfill is 0.8 Mpa [39]. The filling of a single approach in the downward horizontal layered cementation filling method is often divided into two parts: the lower part of the approach adopts the filler with higher strength (4~5 Mpa), and the upper part adopts the filler with relatively low strength (2~3 Mpa) [40].

The values of each mechanical parameter (

Table 1. Mechanical parameters of the rock mass.

Lithology	Densities/kg·m−3	Bulk Modulus/GPa	Shear Modulus/GPa	Internal Friction Angle/°	Cohesion/MPa	Tensile Strength/MPa
Surrounding rock	3090	2.67	1.52	34	1.1	2.1
orebody	3000	2.15	1.44	31	0.8	1
fault	2300	0.38	0.18	21	0.05	0.02

) refer to the results of the mine’s ore and rock mechanics test, and the mechanical parameters mainly include ore and rock density ρ, bulk modulus K, ore and rock cohesion c, internal friction angle φ and the tensile strength of rock mass.

The specific locations of the fault deployment monitoring points are shown in

Table 2. Layout of monitoring points.

Measurement Point Position	Away from the Fault Plane on the Mining Side			Near the Fault Plane on the Mining Side					Interior of Fault
	Bottomless Column Segmental Sublevel	Downward Layered Cementing and Filling Procedure	At the Perimeter Rock below the Ore Body	At the Perimeter Rock above the Ore Body	Bottomless Column Segmental Sublevel	Between the Two Methods	Downward Layered Cementing and Filling Procedure	At the Perimeter Rock below the Ore Body	Above the Ore Body	Bottomless Column Segmental Sublevel	Beneath the Ore Body
point number	1	2	3	4	5	6	7	8	9	10	11

.

The specific locations of the monitoring points of the faults are represented in the model in Figure 6.

3.2. Model Validation

There are a total of 138 surface rock movement GPS monitoring points in the West Second mining area, as shown in Figure 7. The rock shift monitoring updates the monitoring point measurements in real time, and the rows involved are the 5-row and 7-row monitoring points. From 2012 to March 2016, when the collapse occurred, the cumulative vertical subsidence at the 5–11 row lines (Figure 7) for 3 years and 9 months was 1279 mm. The subsidence characteristics of the area clearly show that the subsidence of the measurement points near the fault area was larger than that of the points far from the fault area, and the vertical displacement of the points with large horizontal displacements was also large [34].

As can be seen from Figure 8, after mining the ore body, there was an obvious subsidence center on the surface, located near line 7 of the exploration line, and the subsidence range was between line 5 and line 7 of the exploration line, which can be seen to be consistent with the actual range of rock movement on the surface and the mining range of the ore body. The shape of the subsidence contour was a concentric ellipse, like the actual shape in reality, and the long axis of the ellipse pointed to the strike direction of the ore body; furthermore, with the mining of the ore body, the subsidence area expands, the maximum amount of subsidence also expands, and the subsidence area gradually expands in the direction of the fault, which is consistent with the deformation and damage area observed on the surface. A numerical simulation could be used to calculate that the maximum vertical displacement was about 1.3 m, and the settlement center was near rows 5 and 6.

3.3. Different Sublevel Heights for the Sublevel Caving Method without the Sill Pillar

The sublevel height is an important index that affects the workload when mining, the stability of the return roadway, and the stability of the fault, and the sublevel height of the sublevel method is usually the height of the return roadway plus the height of the fan-shaped shell hole. The segmentation height of the filling method consists of a number of layered heights. Different sublevel heights give different patterns of dynamic response to faults. By comparing the fault-related parameter indexes under the different segmentation heights of the bottomless column segmental sublevel method and the downward layered cementation filling method, we analyzed the evolution of the stress field and displacement field of the surrounding rock and the fault surface in the process of mining the adjacent F8 fault, and then we analyzed the appropriate segmentation heights and the factors that have a great influence on the stability of the fault.

The numerical simulation of mining was carried out by keeping an actual segmentation height of 20 m for the lower-layer filling method. The segmentation height of the sublevel caving method without the sill pillar was set to 10 m, 15 m and 20 m, respectively. Due to the complexity of the mining situation in the field, the sublevel height was changed in the numerical simulation, but the height of the approach was not changed. The change in the fault dynamic response was analyzed, and it is shown in Figure 9.

From the analysis of the above diagrams, it can be seen that the displacement range of the fault plane near the mining side was mainly distributed in the sublevel caving method without the sill pillar and the filling method. The vertical displacement of the fault plane in the disintegration method was the largest, and it gradually decreased to the surrounding area. After mining, the vertical displacement of the fault surface at the filling method was about 0.6 m, and the maximum vertical displacement of the fault plane on the mining side could be up to 1.2977 m when the sublevel height of the sublevel caving method without the sill pillar was 10 m; the maximum vertical displacement of the fault surface when the segmentation height of the sublevel caving method without the sill pillar was 15 m was 1.4504 m, and that of the fault surface when the segmentation height was 20 m was 1.8139 m. With the increase in the segmentation height of the bottomless column segmented sublevel method, the fault surface would have the maximum displacement of the bottomless column segmented sublevel method, and the maximum displacement of the fault surface would be 1.8139 m. The vertical displacement of the fault face increases with the increase in the sublevel height, which indicates that the increase in the sublevel height of the sublevel method will lead to an increase in the instability of the fault; the positive stress on the fault face near the mining side was within the range of 10 MPa~25 MPa, and the local stress would suddenly increase to several tens of megapascals due to the mining process destroying the original stress field, and the value of positive stress would increase to 10 Mpa with the increase in the sublevel height of the sublevel method without the column. As the height of the sublevel caving method without the sill pillar increased, the value of positive stress also gradually increased, but there was no sudden change in the positive stress value.

3.4. Different Sublevel Heights of the Lower-Layer Filling Method

The actual sublevel height of 15 m was maintained for the sublevel caving method without the sill pillar for the numerical simulation of mining. The mining sublevel height of the lower-layer filling method was set to 15 m and 20 m, and the fault’s dynamic response change was analyzed, as shown in Figure 10.

From the analysis of the above diagrams, it can be seen that the displacement range of the fault plane near the mining side was mainly distributed in the sublevel caving method without the sill pillar and in the filling method. The vertical displacement of the fault surface using the disintegration method was the largest, and it gradually decreased in the surrounding area. The maximum vertical displacement of the fault contact surface on the mining side was 1.4352 m when the sublevel height of the filling method was 15 m. The maximum vertical displacement of the fault surface was 1.4504 m when the sublevel height of the filling method was 20 m. With an increase in the sublevel height of the filling method, the vertical displacement of the fault surface also increased; the range of the change in the positive stresses was consistent with the trend in the above analysis. Through the comparison of shear stress and displacement, it can be determined that the collapse method has a greater influence on the stability of the fault than the filling method in the case of changing the sublevel height only.

4. Numerical Simulation Analysis

4.1. Fault Dynamic Response Law under Different Mining Conditions

In order to conform to the real effects of mining, we performed a numerical simulation of single-filling-method mining before sublevel-filling synergistic mining. As shown in Figure 11, the area before the dotted line box is the shear stress change of 11 measurement points for single-filling mining, and inside the box is the shear stress change of measurement points shown by collapse-filling coordinated mining. Figure 11 shows that the shear stress at measurement point 2 and measurement point 7 dropped abruptly during the single-fill mining period, and local slip may have occurred where measurement points 2 and 7 are located, so measurement points 2 and 7 can be ignored in chipping-filling coordinated mining; however, the status of the rest of the measurement points needs to be combined with the dynamic response of the mining during the single-fill mining period to judge them together.

In order to investigate the dynamic response law of the fault under different mining conditions and to analyze the stability of the fault, the changes in shear stress and vertical displacement of the 11 measurement points under different mining conditions were analyzed and compared, and conclusions were drawn.

4.1.1. Shear Stress

Removing the measuring point 2 and the measuring point 7, according to Figure 11 and Figure 12, it can be seen that the dynamic change in the shear stress at the measuring point can be divided into three categories, namely, almost no effect, small fluctuation and large fluctuation. Measuring points 3, 4 and 8 are located in the fault plane of the non-mining area, and their shear stress does not change and is in a stable state. The shear stress of measuring points 5, 6 and 10 fluctuates greatly, and the shear stress of measuring point 6 increases continuously with the increase in segmental height, which indicates that this stage is the stage of elastic energy accumulation, and the shear stress of measuring point 10 decreases from 3 MPa to 0.8 MPa in the early stage of collaborative mining with a section height of 20 m when using the caving method, indicating that the position of this measuring point may be in a rapid slip stage until it is in a stable state. When the segmented height of the collapse method exceeds 15 m, which is 20 m, the shear stress at measuring point 10 shows a greater decrease.

From Figure 13, it can be analyzed that points 3, 4, 8 and 11 are in a stable state. The measuring points with the largest fluctuations in shear stress are still 5, 6 and 10. There is a high probability of slippage at the early stage of mining at measuring point 10, and the change form of these measuring points is consistent with the trend of the caving method at different segmental heights, but with the increase in the segmented height of the filling method, the variation range of the shear stress at all measuring points is small. This may indicate that the sublevel caving method without a sill pillar has a greater impact on the stability of the fault than the filling method.

4.1.2. Displacement

In order to be realistic, the single-fill method of mining before sublevel-fill synergistic mining was also simulated. As shown in Figure 14, in front of the dotted line box is the displacement change of the 11 measuring points of single-filling mining, and inside the box is the displacement change of the measuring points shown by sublevel-filling collaborative mining. The state of the measurement points needs to be judged together with the dynamic response law of mining during the single-filling method.

From Figure 15, it can be seen that the displacement of each measurement point is basically stable when the segmentation height of the sublevel method is 10 m; however, with an increase in the segmentation height of the sublevel method, the vertical displacement of each measurement point shows a different degree of increase in the magnitude of the displacement (except for measurement point 4). There was a sublevel height of 10 m when the vertical displacement figures of measuring points 7 and 5 were larger than other points, for about 0.7~0.8 m; at a sublevel height of 15 m, when the settlement rate of each point increased slightly, there was no sharp decline in the situation, and it was still within the controllable range. At a sublevel height of 20 m, the displacement values of measuring points 5 and 10 regarding the settlement rate appeared to show a substantial increase, with a difference of half a meter or so, and for measuring point 5, after the mining simulation was completed, the settlement displacement reached 1.2 m. This indicates that the fault is in a stable state, and a small amount of local slow misalignment has occurred.

From Figure 16, it can be seen that the displacement of each measurement point was basically stable when the sublevel height of the sublevel method was 10 m, and it can be clearly seen from this group of graphs that the influence of filling on the fault displacement and settlement was smaller than that of the sublevel, and there was basically no change in the displacement of the fault surface from the segment heights of 15 m to 20 m in the filling method. The vertical settlement value of measurement point 5 was the largest at 1 m.

4.2. Fault Stability Analysis

In this paper, the fault slip propensity index was adopted to select the ratio of shear stress/positive stress for evaluating the slip hazard trend and degree [24].

$$\frac{\tau }{\sigma }$$

(3)

which reflects the shear deformation stage in which the fault unit is located. The change rule of the slip susceptibility index affected by mining can not only reflect the fault slip activation state, but also provide precursor stress information regarding fault slip instability. The larger the slip tendency index, the greater the risk of fault activation (see Figure 17).

When the section height of the filling method was maintained at the actual mining height and the section height of the sublevel caving method without the sill pillar continuously increased, the frequency of fluctuation of the slip tendency indicator increased, and the maximum value of the indicator increased continuously, except for measurement points 3, 4, 6 and 8, which are quasi-static and had no change. The slip propensity index of measurement point 10 decreased significantly during mining, then recovered with increasing frequency as the height of the sublevel segmentation increased before stabilizing at the end. On the other hand, the slip propensity index of measuring point 5 first showed an increase in different fluctuation frequencies with the increase in the height of the chipping section during mining, and then it stabilized. This indicates that the slip hazard of measuring point 5 continued to increase, and the slip tendency index could reach 0.55 at the height of 20 m in the bottomless column segmented sublevel method, which represents a certain degree of danger. Measurement point 10 may have had a small local slip because it was the only point with a large drop in indicators. The slip indicators of measurement points 1, 2, 7 and 11 had a small increase at first and then stabilized; therefore, the fault as a whole was in a stable state. When the segmented height of the sublevel method was not changed to the actual mining height, increasing the segmented height of the filling method did not create a difference in the slip tendency indexes of each measurement point, and there was only a very small change.

5. Conclusions

In most previous studies, scholars often studied the impact of single mining methods on faults, but the impact of co-mining on faults was rarely studied. In this paper, the following main conclusions are drawn through numerical simulation and theoretical analysis:

Upon increasing the height of segments of the sublevel caving method without the sill pillar and the lower-layer filling method, the changes in shear stress and displacement showed similar patterns. The displacement and shear stress of the fault face near the mining side increased continuously, but the impact of the disturbance of the sublevel caving method without the sill pillar on the fault face was smaller than that of the lower-layer filling method.

By analyzing the dynamic responses of 11 measurement points, it can be concluded that the instability was greatest with the sublevel caving method without the sill pillar within the fault, and a small slip may occur, but no slip phenomenon occurred at the other measurement points, so the fault is in a stable state as a whole. Additionally, the impact of the sublevel height change in the sublevel caving method without the sill pillar is larger than that of the filling method, which is more likely to put the fault into an unstable state.

When the sublevel height of the sublevel caving method without the sill pillar increased from 10 m to 15 m, the changes in the shear stress and displacement of the fault surface did not show any rapid changes; however, when it grew to 20 m, the rate of displacement continued to increase, which increased the index of slip tendency, and the risk of activation of the fault became larger. When the sublevel height of the filling method increased from 15 m to 20 m, the shear stress, displacement and slip tendency indexes did not change much.

When the mining face is close to the fault, the shear stress of some measuring points on the fault plane will increase by about 1.4 times, and the stress concentration phenomenon will occur, so when the working face is advanced to the fault, attention should be paid to the change of fault thickness and the change of surrounding rock stress, and preventive measures should be taken immediately when sudden changes occur to ensure the safe and efficient production of the mine.