Case histories of rock bursts under complicated geological conditions

Abstract

During the past decade’s exploitation of coal seams in Muchengjian Mine in Jingxi Coalfield, there were nearly thirty rock burst events, which hindered the safe and efficient coal production. Two typical mining areas were selected for analysis where almost half of rock burst events occurred. The research was aimed at finding connections between the occurrence of rock bursts and geological characteristics. The temporal and spatial characteristics of rock bursts were described in detail, the geological characteristics were investigated carefully, and the possible reasons for rock bursts were analyzed. The details documented in these cases not only provide an essential reference value for understanding the development mechanism of rock bursts, but also provide a basis for selecting control measures and optimizing related technical parameters during tunneling or mining under complicated geological conditions.

Introduction

Rock burst, a common dynamic disaster often accompanied with sudden, quick and violent ejection of coal or rock during exploitation of coal seams, often happens in complex ways under special conditions, even without any warning information (Tan et al. 2011; Xiao et al. 2016). Such failure characteristics pose a great threat to the safety and efficiency of mining. Despite decades of research, the mechanism of the rock burst remains not adequately understood.

Researchers, however, have reached an agreement that there are primarily two key factors causing the rock burst phenomenon, namely, mining engineering conditions and geological conditions (Christopher 2016; Dou et al. 2014; Iannacchione and Tadolini 2016; Tan et al. 2012; Zhou et al. 2015). It is generally known that many geological conditions (such as folds, faults, seam pitching, hard roof, and mining depth, etc.) are generated by the geodynamic movement, and their characteristics directly determine the occurring conditions of rock bursts or other dynamic disasters (Aguado and González 2009; Jiang et al. 2014; Mazaira and Konicek 2015; Snelling et al. 2013; Tan et al. 2015). According to incomplete statistics, the rock burst events occurring near geological structures, account for nearly 70% of the total in China (Wang et al. 2012). Jingxi Coalfield, which is called “China’s Geological Encyclopedia”, has a very complicated geological condition. Figure 1 shows the geological structure sketch map of Jingxi Coalfield. One of its representations is Muchengjian Mine, which is rich in folds, faults, phase transition areas (thickness/angle change areas of coal seam), and other structures. Muchengjian Mine has witnessed more than thirty rock burst events during the past ten years. The analysis and summary of these events build a foundation for understanding, predicting, early warning, and preventing of rock bursts in this kind of complicated geological mines or other underground engineering projects.

Fig. 1

Geological structure sketch map of Jingxi Coalfield (Han et al. 2014)

It is not practical to describe every event because a large number of rock bursts occurred in the Muchengjian Mine. Two typical mining areas, where rock bursts occurred frequently, were selected for this research since the characteristics and mechanisms of rock bursts in the two areas are typical in the whole Muchengjian district. Firstly, basic engineering geological characteristics of the mining district, mining areas, and working faces are presented. Then, the corresponding information of every burst event in every working face is described in detail including spatial and temporal characteristics of rock bursts, damage mechanism characteristics of roadways and surrounding rock, and geological characteristics of accident locations, etc. Lastly, a summary discussion is presented. The information revealed in these rock bursts can provide essential references not only for the study of rock bursts and its mitigation, but also for the optimizing design and exploitation in other underground tunneling or mining engineering under complicated geological conditions.

Engineering geological characteristics and in-situ stress field

Overview of engineering geological characteristics

The geological structure is complicated in the Muchengjian Mine in the Jingxi Coalfield. The mine is located in the south-east wing of “Miaoan Ridge-Jiji Mountain Syncline”, as shown in Fig. 1. The most common structures are large or small folds, accompanied with fault structures. These folds show a series of secondary anticlines or synclines, whose formations are open and stretched in the south and north, but locked in the middle, causing great changes of the coal seam. This complicated geological condition not only makes difficulties for coal production, but also leads to frequent rock bursts.

The main mining coal seam of Muchengjian Mine is the No. 3 coal seam. Figure 2 shows the main part of the geological occurrence of No. 3 coal seam (Fig. 2 only reveals the large geological structures). From Fig. 2, we can find the two typical mining areas: the West No. 1 Mining Area of +250 Level and the No. 1 Mining Area of +150 Level (W1 Area and M1 Area for short). The north of W1 Area is located in the bottom of a fold, and two large faults exist near the south boundary. The M1 Area is also located near a fold, but the fold curvature is smaller. The average mining depth of M1 area, however, reaches more than 900 m, much deeper than that of W1 Area, which is only 700 m.

Fig. 2

Part of the geological occurrence of No. 3 coal seam and layout of in situ stress measure stations

The exploratory boreholes reveal that the coal seam thickness in the W1 Area varies from 1.0 to 2.5 m with an average of 1.8 m. The dip angle varies from 3° to 20°, and the average value is 15°. The lithological characteristics of roof and floor are listed in Table 1.

Table 1 Lithological characteristics of roof and floor in the W1 Area

In M1 Area, the cored boreholes show that the coal seam thickness is different, but it has similar rock parting bands with M1 Area as listed in Table 2. The coal seam thickness varies from 1.6 to 4.0 m, and the average thickness is 3.4 m. The dip angle varies from 18° to 40°, and the average value is 20°.

Table 2 Lithological characteristics of roof and floor in the M1 Area

Coal burst liability

The coal burst liability (CBL) is the inherent property of coal, which arises with the increase of stored strain energy in the coal seam. It is the intrinsic factor causing rock bursts. Researchers have proposed various burst liability indexes based on energy, stiffness, strength, and failure duration, etc. The CBL indexes widely used in China as standards include the uniaxial compressive strength (R _C), elastic strain energy index (W _ET), bursting energy index (K _E), and dynamic failure duration (D _T). CBL classifications according to the Chinese standard (GB/T 25217.2-2010 2010) are listed in Table 3.

Table 3 Index values of CBL classification (GB/T 25217.2-2010, 2010)

The CBL indexes were measured through coal samples obtained from the W1 Area and the M1 Area, and the test results are listed in Table 4. In the W1 Area, the ranges of D _T, K _E, W _ET, and R _C are 15.1–46.3 ms, 11.5–24.0, 4.0–6.6, and 17.5–27.1 MPa, respectively, with the corresponding average values of 25.9 ms, 17.9, 5.1, and 22.0 MPa. In the M1 Area, the ranges of D _T, K _E, W _ET, and R _C are 17.4–52.4 ms, 10.5–19.2, 8.1–13.9, and 14.8–21.2 MPa, respectively, with the corresponding average values of 30.0 ms, 14.8, 10.8, and 17.5 MPa. These results indicate that the No. 3 coal seam has a strong CBL, meaning that it has the capacity to store energy and is prone to dynamic breakage.

Table 4 Test results of CBL

In-situ stress field

The in situ stress field is a fundamental factor causing dynamic disasters in underground mining, especially when mining depth increases or a geological structure exists (Tan et al. 2012). Thus, three measure stations for measuring the in situ stress were laid out, as shown in Fig. 2. The overcoring method of hollow inclusion was used in this test, and test results are listed in Table 5. According to the field test, the primary characteristics of the geo-stress field are summarized as follows:

Table 5 Test results of in situ stress

• 1. The maximum principal stress increases with the increase of depth except for the No. 3 measure station where the stress value is the largest (28.70 MPa) while the buried depth is only 581 m. Obviously this is caused by the fold syncline—a high in situ stress concentration area.

• 2. The dip angle of the maximum principal stress varies from −19.30° to 0.43°. This result means that the direction of in situ stress is near-horizontal. Thus, the tectonic stress is the main stress field in the Muchengjian Mine.

• 3. The azimuth angle of the maximum principal stress is primarily within the range of 182.65°–267.94°, indicating that the stress direction changes slightly. The direction of the maximum principal stress is basically perpendicular to the synclinal axial, especially for the No. 1 and No. 2 measure stations.

Rock burst distributions

The arrangement of working faces where rock bursts occurred in W1 Area is shown in Fig. 3a, b, and the M1 Area is shown in Fig. 3c. During the exploitation, eleven rock burst events occurred in the W1 area, including five in the W-No. 1 working face, two in the E-No. 1 working face, two in the W-No. 4 working face, and two in the W-No. 5 working face. Four rock burst events occurred in the M1 Area, including one in the E-No. 4 working face and three in the E-No. 3 working face.

Fig. 3

The arrangement of working faces. a Middle-south of W1 Area. b North of W1 Area. c M1 Area

West No. 1 mining area of the +250 level

The W-No. 1 working face

“2.28” and “3.6” events

On February 28, the first rock burst (referred to as the “2.28” event, and the other rock burst event will be called in this way, too) occurred during tunneling the headgate of the W-No. 1 working face ,and the driving distance was only 25 m. Just one week later on March 6, this event was followed by a more intense rock burst (referred to as the “3.6” event). During this event, the driving distance was about 46 m, as shown in Fig. 4a.

Fig. 4

Careful descriptions of the W-No. 1 working face and every rock burst. a The layout of W-No. 1 working face. b “2.28” event. c “3.6” event. d “6.28” event. e “8.26” event. f “8.29” event

The “2.28” event caused a 1.1 m wide coal seam of both side walls loosening from the heading face, and the influenced length was about 10 m. From 10 to 15 m, the loosening width was 0.4 m because the bolt-net support system worked, as shown in Figs. 4b and 5a. Meanwhile, the roof of the roadway throat showed a separation phenomenon. The “3.6” event, however, caused serious rib spalling at the range of 6–10 m, and the spalling width was about 0.3–1.0 m, as shown in Figs. 4 and 5b. Also in this area, one bolt was pulled out, and others were loosened to different degrees.

Fig. 5

Photos of roadway’s damage after rock bursts in the W-No. 1 working face. a “2.28” event. b “3.6” event. c “6.28” event. d “6.28” event

These two events both occurred in the triangle area isolated by the F _1-1 fault. The “2.28” event happened when the heading face was about 22 m away from the F _1-1 fault, and the “3.6” event occurred at the location just ahead of F _1-1 fault. Obviously, they are related to the slip of the F _1-1 fault induced by excavation activities. Thus, they both belong to a “fault slip” burst.

“6.28” event

During the early recovery, on June 28, 2009, when stoping the position where it was about 140 m away from the uphill roadway, a rock burst event occurred in the headgate, as shown in Fig. 4a, d. The failure length of the headgate was about 17 m (From the north end of the longwall face to the east), including a 0.8–1.2 m wide south rib spalling, and Fig. 5c gives the damage photograph. The damage photograph is shown in Fig. 5d, where the section of 5.5–13.8 m was destroyed seriously with a scraper conveyor buried and three hydraulic props pulled down. In addition, this event also impacted the longwall face. The length was about 16.5 m from the north end to south, mainly including a rib spalling of 0.4–1.0 m in width, but the range of 16.5–33 m just had a sign of coal slagging without any damage.

The “6.28” event occurred when stoping the area where the thickness of the coal seam fluctuates from 3.5 to 2.2 m. Stoping from the thick to the thin is not advantageous for burst control, because a high stress concentration area could be formed more easily (Zhao et al. 2016). There also existed a F _1-2 fault just across this section, which caused the formation of an isolated triangle area, as shown in Fig. 4d. This triangle area might be activated by stoping activities. According to the high damage level, both of the geological factors are more likely to make contributions to the occurrence of the “6.28” event. Thus, the “6.28” event might be called “coal seam thickness change + fault slip” burst.

“8.26” and “8.29” events

In the final exploitation process, two rock burst events occurred in the uphill roadway on August 26 and 29, 2009, respectively, as shown in Fig. 4a. Distances between the stoping face and uphill roadway were 35 m for the “8.26” event and 28 m for the “8.29” event, respectively. For the “8.26” event, the damage length was about 55 m from the headgate throat to the south, as shown in Fig. 4e. In this area, the roof broke, the west rib spalled, and the floor heaved, but not very seriously.

The “8.29” event, however, caused more serious damage to the uphill roadway. Starting from the headgate throat to the south, nearly a 104 m length of roadway was influenced. Figure 4f gives the detailed description. In the section of 26–40 and 60–76 m, the rib spalled slightly; while in sections of 40–60 and 76–104 m, the whole west rib was pulled out, causing the roadway( height was only 0.8–1.2 m) and that a few winches to be buried. Despite the extensive damage, it shows a really interesting phenomenon that the headgate had no damage. The possible reason is that the high stress concentration of the two walls of the headgate might have been released.

In addition to the mining-induced stress, which could not be completely avoided, apparently there are two main factors causing the two events, including coal pillar and fault. The F _1-1 fault might make more contribution to the “8.26” event because the damage section was close to the F _1-1 fault. But for the “8.29” event, it might be related to the coal pillar more possibly because the pillar was only 20–28 m in width and the serious damage section was far away from the fault. Meanwhile, the F _1-1 fault might have slipped twice in the “2.28” and “8.26” events, and the high stress concentration had been mitigated more or less. This might be the reason that the headgate had no damage during the “8.29” event. In general, both the “8.26” and “8.29” events are a kind of “coal pillar + fault slip” burst during the final exploitation process.

More attention needs to be paid to another interesting phenomenon. The interval between the “8.26” event and “8.29” event is only three days, and the intensity of the former is relatively weak. In other words, when in the late stoping process, the deformation or damage of the roadway may be a warning sign of an intense rock burst event.

The E-No. 1 working face

“5.24” event

On May 24, 2008, when the driving distance was only 25 m, the “5.24” event occurred in the headgate of the E-No. 1 working face, as shown in Fig. 6a, b. In this event, the heading face was not damaged. The damage section started at the station 8–18 m away from the heading face, and the width of the rib spalling was 0.5–1.0 m. From the headgate throat to the south about 10 m, the uphill roadway had various kinds of rib spalling phenomena without any damage. What is more, when the headgate was finished, both side walls of the headgate heaved about 200 mm. Figure 6b shows that headgate was excavated between the F _2-1 fault and F _2-2 fault; in other words, it was tunneled in the stress concentration area. The main damage section was near the F _2-1 fault. Obviously, the “5.24” event is a kind of “fault slip” burst.

Fig. 6

Careful descriptions of the E-No. 1 working face and every rock burst. a The layout of E-No. 1 working face. b “5.24” event. c “11.10” event

“11.10” event

On November 10, 2009, an intense rock burst occurred during the stoping process. The longwall face, tailgate and uphill roadway were all damaged, as shown in Fig. 6a, c. When the “11.10” event happened, the distance between the tailgate and uphill roadway was only 41 m. Between the middle tailgate throat and tailgate throat, the whole uphill roadway was destroyed: the coal ribs were pulled out about 1.2–1.5 m in width, the two belt conveyors were both buried, some hydraulic props were broken, and the roadway’s height was less than 0.8 m, as shown in Fig. 7a. Meanwhile, from the tailgate throat to the longwall face, the height of the tailgate was only about 1.0–1.2 m, a few hydraulic props were broken and the scraper conveyor was buried, as shown in Fig. 7b. Although the longwall face suffered the minimum damage, the rib spalling was about 1 m in width, as shown in Fig. 7c.

Fig. 7

Photos of roadway’s damage after the “11.10” event in the E-No. 1 working face. a Uphill roadway, b Tailgate, c Longwall face

In Fig. 6c, it was clearly found that the “11.10” event happened when stoping the isolated coal pillar. In addition, other two coal pillars existed next to the coal pillar, one in the west (10 m width) and the other in the south (4–10 m width). There were no large geological structures in this area. The distance between the F _2-2 fault and the impact region was more than 50 m. In the “11.10” event, the coal pillar plays a dominant role. There are three obvious items of evidence for supporting this viewpoint. Firstly, the damage length of the headgate was only 10 m, and the damage intensity was far weaker in the “5.24” event (“fault slip” burst) than that in the “11.10” event. Secondly, the uphill roadway between the middle tailgate throat and the headgate throat and the middle tailgate, which were closer to the fault F _2-1 and F _2-2, were not damaged or influenced in the “11.10” event. Thirdly, the damage intensity of the uphill roadway and the tailgate next to the coal pillars were rather stronger than the longwall face. Therefore, the “11.10” event is a kind of “coal pillar” burst. In this event, the geometric irregularity of coal pillar also makes a contribution because the coal pillar is prismatic, existing as two acute angle areas where the abutment pressure is more easily overlapping between the two roadways. Yu et al. (2016) has summarized the factors affecting coal pillar stability in detail (Note: the detailed analysis of factors affecting the “coal pillar” burst is not the main topic in this paper, thus it will not be added).

The W-No. 4 working face

During the stoping process, two rock burst events occurred in the tailgate of the W-No. 4 working face, as shown in Fig. 8a. They occurred on December 4 and 18, 2008 (“12.4” and “12.18” events), respectively.

Fig. 8

Careful descriptions of the W-No. 4 working face and every rock burst. a The layout of W-No. 4 working face. b “12.4” event. c “12.18” event

The “12.4” event happened when the stoping face was about 213 m away from the uphill roadway, as shown in Fig. 8b. This event caused a rib spalling 13 m in length on the south side wall of tailgate, starting from 5 m away in front of the longwall face. It led to some bolt failures and a number of pre-support props half buried, as shown in Fig. 9a, b. The location of the “12.4” event was in a triangle area of the coal pillar (high stress concentration area), which was isolated by the F_4-3 fault. Obviously, this event is a kind of “coal pillar + fault slip” burst.

Fig. 9

Photos of roadway’s damage after rock bursts in the E-No. 1 working face. a “12.4” event. b “12.4” event. c “12.18” event. d “12.18” event

The following “12.18” event also started from 5 m away in front of the longwall face, and had 10 m damage in length which was weak, as shown in Figs. 8c and 9c, d. There is an interesting phenomenon that this event only occurred on the north side wall of the tailgate. From the feedback information of underground workers, it can be known that only the south side wall was destressed after the “12.4” event. It is obviously that the high stress concentration inside the south side wall was released, but the north side wall was not. Thus, the “12.18” event occurred. This event might also be a kind of “coal pillar + fault slip” burst.

The W-No. 5 working face

During the stoping process of the W-No. 5 working face, two rock burst events occurred on April 16 and July 8, 2013 (“4.16” and “7.8” events), respectively, as shown in Fig. 10a.

Fig. 10

Careful descriptions of the W-No. 5 working face and every rock burst. a The layout of W-No. 5 working face. b “4.16” event. c “7.8” event

“4.16” event

When the distance between the headgate and the uphill roadway was about 171 m, the “4.16” event occurred. Figure 10b gives the detailed description of the damage. The damage location started from the location 13 m ahead of the longwall face. The range of 13–43 m experienced a 0.8–1.2 m wide rib spalling, and the belt conveyor was deformed and buried (Fig. 11a, b); while the range of 43–64 m only suffered a slight rib heave (Fig. 11c).

Fig. 11

Photos of roadway’s damage after the “4.16” event in the W-No. 5 working face. a Conveyor deformation and half buried. b Rib spalling. c Slight rib heave

There were no coal pillars or obvious geological structures in the W-No. 5 working face. Figure 12 gives the abutment pressure distribution rule when stoping the W-No. 5 working face, which clearly shows that the influencing range of abutment pressure was about 0–55 m. Compared with Fig. 10b, we can see that the main damage section was just located in this area but mining activities could not be avoided. Two significant external causes in this event are that the W-No. 5 working face was just located in the core of a synclinal fold, and that the stoping direction was sub-parallel to the synclinal axis. In accordance with Fig. 2, it can be seen that the stoping direction was almost perpendicular to the maximum principle stress direction. In addition, the core of synclinal fold is a high in situ stress concentration area (Zhang et al. 2012; Gu et al. 2015). Zhang (2015) and He et al. (2011) have proved this phenomenon through numerical simulation directly and micro-seismic monitoring indirectly (Micro-seismic events and energy were larger when stoping approaching the syncline than that stoping in other areas of the No. 250102 working face, revealing the relief of high in situ stress), respectively, as shown in Fig. 13. Therefore, the “4.16” event could be classified as a “fold” burst.

Fig. 12

Distribution rule of abutment pressure of longwall face during stoping the W-No. 5 working face

Fig. 13

In-situ stress distribution rule and field micro-seismic monitoring results in the core of syncline. a Horizontal stress concentration (Zhang 2015). b vertical stress concentration (Zhang 2015). c Micro-seismic monitoring results during stoping (He et al. 2011)

“7.8” event

The “7.8” event occurred when the headgate was about 130 m away from the uphill roadway, as shown in Fig. 10a, c. In this event, both the headgate and uphill roadway were damaged. From the south end of the longwall face to the east, the north side wall of the headgate had a rib spalling of 0.4–1.2 m in width in the range of 6–30 m, causing the roadway’s height and width to be left at only 1.52 and 1.42 m, respectively. Meanwhile, six hydraulic props were bent. The sections of 0–6 and 30–35 m were also influenced but not seriously. From the headgate throat to the south, the north side wall of the headgate spalled about 0.3–0.6 m in width in the range of 0–55 m, causing the roadway’s height and width to be left at 1.3–1.8 and 1.7–2.5 m, respectively. The damage of the uphill roadway was weak from the headgate throat to the north, the range of 5–35 m suffered rib spalling of 0.3–0.6 m in the width of two walls and some bolt failures, while there was no damage in the range of 0–5 m. Obviously the “7.8” event could also be classified as a “fold” burst. In addition, the coal pillar might also make a contribution to the damage (triangle area in Fig. 10c) near the uphill roadway, but not the main cause of high stress concentration in this working face.

No. 1 mining area of the +150 level

During the tunneling and stoping process, four rock burst events occurred in the No. 1 mining area of +150 level (M1 Area). The first two events happened when the cross heading was tunneled in the E-No. 4 working face on July 20, 2014, and when the headgate was tunneled in the E-No. 3 working face on June 22, 2015. The other two occurred during stoping the E-No. 3 working face on October 24 and November 20, 2015, respectively. The event locations, geological conditions, and mining conditions are displayed in Fig. 14.

Fig. 14

Careful descriptions of the E-No. 3 and E-No. 4 working faces, and every rock burst. a The layout of E-No. 3 and E-No. 4 working face. b “7.20” event. c “6.22” event. d “10.24” event. e “11.20” event

The E-No. 4 working face

The “7.20” event occurred when the distance between the heading face and the headgate (not excavated yet) was about 20 m, as shown in Fig. 14b. The damage position, however, started from 23 m behind the heading face and ended at 63 m. The south side wall and floor heaved along this section. Meanwhile, a large crack sub-parallel to the F_150-1 fault was generated in the middle of the damage section. This event was in the thickness-change area of the coal seam and near the F _150-1 fault. Both the two geological structures could lead to high stress concentration. The “7.20” event is more likely to be the “coal seam thickness change + fault slip” burst.

The E-No. 3 working face

“6.22” event

When the distance between the heading face and uphill roadway was about 28 m during tunneling the headgate, the “6.22” event occurred. The influenced length was about 9 m, starting from 4 m away behind the heading face. The damage was not very serious with a 0.2–0.5 m width north rib heave, as shown in Fig. 14c. Obviously, the damage section was located near a thickness change area of coal seam, and it might have suffered from the abutment pressure formed due to the uphill roadway. Scholars have reached a consensus that the facies change area of coal seam thickness also has high in situ stress concentration except for faults and folds (Wang et al. 2012). Zhao et al. (2016) and Sun (2003) have proved that in situ stress in the thinning area is larger than that in the thickening area, which means existing stress gradient zones in the variable region of coal thickness, as shown in Fig. 15. Thus, the occurrence of a rock burst is also likely to occur in this kind of tectonic area under the influence of mining activities, including stoping or tunnelling processes. In addition, the impact intensity of the “6.22” event is less serious than that of the “7.20” event because the F _150-1 fault also makes a contribution in the “7.20” event. Therefore, the “6.22” event might be a kind of “coal seam thickness change” burst.

Fig. 15

Distribution rule of in situ stress in the variable region of coal thickness (Zhao et al. 2016)

“10.24” event

The “10.24” event occurred when the stoping face was about 87 m, far away from the uphill roadway, as shown in Fig. 14d. The damage length of advanced support was about 37 m. The two walls and floor heaved, especially the range of 0–12 m had a 0.3–0.5 m width north rib heave, leading to only 2.8 m in the width of the headgate. Figure 14d obviously reveals that the “10.24” event occurred in the section where the coal seam became thinner, especially the most serious damage part was just located in the thickness change area of the coal seam. It has some similar characteristics with the “6.28” event in the W-No. 1 working face. The “10.24” event is a kind of “coal seam thickness change” burst.

“11.20” event

When the distance between the stoping face and uphill roadway was about 42 m, the extremely intense “11.20” event occurred. It caused damage in the longwall face, headgate and uphill roadway, as shown in Fig. 14e. The longwall face had a rib spalling of 25 m in length, starting from the south end side. The north side wall of the headgate spalled about 1.3–2.0 m in width along the range of 8–34 m in front of the longwall face, causing the scraper conveyor to be half buried and hydraulic props moved. The north side wall of the uphill roadway heaved slightly in the range of 0–6 m from the headgate throat to the south. While in the range of 0–15 m to the north, the north side wall suffered a rib spalling of 0.4–0.5 m in width, and a few hydraulic props moved. The distance between the F _150-1 fault and the impact region was more than 100 m. Before the “11.20” event, the “7.20” event was caused by the F _150-1 fault and the variable region of coal thickness changed, but its influencing length was only 40 m. Obviously the F _150-1 fault is not the main reason for the “11.20” event. In Fig. 14e, we can know that the length of the E-No. 3 working face was only left at 42 m, which is less than the influencing range of abutment pressure of the longwall face (similar to Fig. 12). This reason leads to the superposition of abutment pressures of the longwall face and the uphill roadway as illustrated in Fig. 16. Thus, the critical stress for the rock burst could be easily reached. This phenomenon is obviously similar to a “coal pillar” burst. That is to say, the “11.20” event might belong to a “coal pillar” burst.

Fig. 16

Schematic image of superposition of abutment pressures of the longwall face and the uphill roadway

Discussion

The total fifteen rock bursts include five during the tunneling process and ten during the stoping process, and both the occurrence probability and damage intensity of rock bursts during the stoping process are greater than that during the tunneling process. For the ten events occurring during the stoping process, the occurrence frequency is the largest in the gate road, the second largest in the uphill roadway, and the smallest in the longwall face; for the five cases occurring during tunnelling process, four occurred behind a distance of 4–23 m from the heading face.

From the above presentations, descriptions, and analyses, this research reveals that geological structures (such as faults, folds, and thickness change areas of coal seam, etc.) make a critical contribution to the occurrence of rock bursts, which is different from those in deep tunnels focusing on the structural planes (Zhang et al. 2012; Zhou et al. 2015). In addition, the coal pillar is also an important factor, especially in the final exploitation process. Thus, the research on the performance of the coal pillar is also very essential. Fortunately, researchers have done much work to evaluate the performance of coal pillars between working faces (Yu et al. 2016), which can also provide references for determining the critical length of a coal pillar formed during the final stoping process.

Based on the reasons regarding causation, these rock bursts could be generally classified into two major groups, including “single factor” rock burst mainly induced by one cause and “multiple factors” rock burst induced by two or more than two causes. For the “single factor” rock burst, it consists of “fault slip” bursts (“2.28”, “3.6”, and “5.24” events), “fold” bursts (“4.16”, and “7.8” events), “coal seam thickness change” bursts (“6.22”, and “10.24” events), and “coal pillar” bursts (“11.10”, and “11.20” events); for the “multiple factors” rock burst, it consists of “coal pillar + fault slip” bursts (“12.4”, “12.18”, “8.26” and “8.29” events), and “coal seam thickness change + fault slip” bursts (“6.28”, and “7.20” events). Scholars mainly focus on the mechanism of “single factor” rock bursts (Dou et al. 2014; Jiang et al. 2014; Wang et al. 2012); there is little in the literature related to the “multiple factors” rock bursts. However, “multiple factors” rock bursts also account for a large proportion, which is about 40% in the two mining levels in the Muchengjian Mine. Thus, the mechanism of “multiple factors” rock bursts also needs further deep study.

What is more, this classification clearly shows the occurrence mechanism of rock bursts in different geological structures. In addition to the common factors (fault, fold, and coal pillar), equal attention should be paid to the thickness change of a coal seam as for other geological structures. One thing also needs to be noted is that all the “coal seam thickness change” rock bursts occurred when tunneling or stoping from the thick to the thin in this paper. Especially when two or more factors play a role in the occurrence of rock bursts, the warning level must be enhanced. For example, in the “coal seam thickness change + fault slip” burst or “coal pillar + fault slip” burst, the overlying stress generated by mining activities and by the thickness change of the coal seam or coal pillar can lead to fault slip more easily. As for the “fold” burst, the mining area itself is a rock burst-prone region with a high stress concentration. If there are other factors involved in it, the damage intensity would be quite intense.

Conclusions

Based on the information revealed in this paper, some characteristics of these rock burst events have been summarized and a few conclusions have been drawn, as follows:

• The number of rock burst events during the stoping process is larger than that during the tunneling process as well as is the damage intensity. During the stoping process, the occurrence frequency is the largest in the gate road, the second largest in the uphill roadway, and the smallest in the longwall face; during the tunnelling process, rock bursts often occur behind a distance from the heading face.

• The number of rock bursts is twice or more (no matter weak or intense) in every working face except in the E-No. 4 working face. This phenomenon reveals that if one rock burst event occurs, prevention and control measures must be taken. For example, the south side wall of the tailgate in the W-No. 4 working face was destressed after the occurrence of the “12.4 event”, and it had no damage in the following “12.18” event, but the north side wall had.

• The major geological structures, such as faults, folds, and coal seam thickness change areas, play an important role in the rock burst in the Muchengjian Mine. Meanwhile, the coal pillar is another essential factor, especially in the final exploitation process. Mitigation measures must be taken when one of these factors appears. Especially if two or more factors appear, warning and prevention levels must be enhanced. A new phenomenon should be paid equal attention—when stoping from the thick coal seam to the thin, the rock burst could occur more easily.

• In the late stoping process, the deformation or damage of the roadway may be the warning information of an intense rock burst. For instance, the interval between the “8.26” event and the “8.29” event is only three days in the W-No. 1 working face, and the former’s intensity is relatively weak.

• This analysis of the fifteen rock burst events provides an essential reference for understanding the causes and development mechanisms of rock burst. It also lays a critical foundation for exploring control measures of rock bursts and the related technical parameters for the mining or tunneling process under complicated geological conditions.