Evaluation of current coal burst control techniques and development of a coal burst management framework

Tunnelling and Underground Space Technology 81 (2018) 129–143

Contents lists available at ScienceDirect

Tunnelling and Underground Space Technology journal homepage: www.elsevier.com/locate/tust

Evaluation of current coal burst control techniques and development of a coal burst management framework

T

⁎

Chunchen Weia, Chengguo Zhanga, , Ismet Canbulata, Anye Caob, Linming Doub a b

School of Minerals and Energy Resources Engineering, University of New South Wales, Sydney, NSW 2052, Australia Key Laboratory of Deep Coal Resource Mining, Ministry of Education, China University of Mining and Technology, Xuzhou, Jiangsu 221116, China

A R T I C LE I N FO

A B S T R A C T

Keywords: Coal burst Preventative controls Mitigating controls Coal burst management framework

Coal burst has been increasingly attracting attention in Australian coal mines recently as they go deeper. Coal burst is well known for its catastrophic destruction, complex mechanisms and difficulty of control in the mining industry. This paper summarises the control measures used globally for this dynamic failure, and shows how to develop site specific control management plans. Firstly, relevant terminology used in dynamic rock failure events in the international underground mining industry is discussed. Preventative controls and mitigating controls are then presented and discussed. Current coal burst controls include general management strategy, mine design, preconditioning and destressing as risk mitigation, and ground support strategies. Optimum layout methods, mitigating strategies, including latest ground support techniques and destressing techniques, are reviewed in this paper. A framework of coal burst management is then proposed, including three critical stages: identification of coal burst profile, development of a management plan and management of coal burst.

1. Introduction

type of dynamic rock or coal failure are similar.

As a major type of rock or coal dynamic failure, coal burst is one of the most catastrophic events for underground excavations, especially for those at greater depth. Since the first coal burst in Britain in 1738 (Dou and He, 2001), burst events have been reported worldwide and severely threaten mine safety and productivity. The first official documented coal burst in Australia (Hebblewhite and Galvin, 2017), a rib burst, occurred at Austar mine in New South Wales in 2014. Table 1 summarises international coal burst occurrences. This paper extends previous reviews on coal burst contributing factors, mechanisms, monitoring and controls (Bräuner, 1994; Zhang et al., 2017), with a focus on coal burst control strategies. Four major groups of coal burst controls are discussed in this paper: general management strategy, preconditioning and destressing as risk mitigation, mine design, and support technologies and strategies. There is still a lack of understanding of appropriate and effective control techniques and their implementation under different mining and geological conditions. Therefore, this paper summarises and compares the control measures based on a worldwide database, identifies knowledge gaps and proposes a general systematic control strategy for coal bursts. Although the paper focuses on coal bursts in underground mines, it can also inform burst control strategies in other underground excavations such as tunnelling since the fundamentals of control of this

2. Coal burst terminologies, occurred conditions and monitoring systems

⁎

Coal burst is a dynamic form of rock failure and usually happens with audible sound and large deformation of roadways. Strain energy stored in the surrounding rock mass is suddenly released at the same time (Jiang et al., 2014; Galvin, 2016; Mark, 2016). There are other terms used to describe seismic activities, such as pressure burst, strain burst, outburst, tremors, bump and bounce, as described below. The term pressure burst is synonymous with coal burst, but coal burst, together with pillar burst, refers specifically to a pressure burst event that expels coal into excavation, as opposed to rock from roof or floor (Hebblewhite and Galvin, 2017). A strain burst is a form of coal burst, but with lower magnitude of energy release. Coal mine tremors, bump and bounce mainly refer to a sudden shake scenario with significant audible sound, but have no or minimal rock mass ejection (Jiang et al., 2014; Galvin, 2016). The term coal burst refers to seismic activities with ejection of materials but no or minimal gas pressure, which is the main energy source for outburst. The key outcome of a coal burst event is that it causes damage to the excavation and can result in personnel injuries or equipment destruction. The Mine Safety and Health Administration (MSHA) (Whyatt et al., 2002) defines coal burst

Corresponding author. E-mail address: [email protected] (C. Zhang).

https://doi.org/10.1016/j.tust.2018.07.008 Received 7 July 2017; Received in revised form 23 May 2018; Accepted 7 July 2018 0886-7798/ © 2018 Elsevier Ltd. All rights reserved.

Tunnelling and Underground Space Technology 81 (2018) 129–143

C. Wei et al.

Table 1 Summary of coal bursts in different countries (Zhang et al., 2017). Country

Years

No. of burst events

Fatalities

References

USA

1905–2014

492

132

China

1949–2015

∼2000

300 during 2006–2013

Iannacchione and Tadolini (2008a), Iannacchione and Tadolini (2016), Mark and Gauna (2016) Dou and He (2001), Dou et al. (2014), Jiang et al. (2014)

Poland

1977–2015

Czechoslovakia and Czech Republic Germany

1930–2015

109 (> 60% of mines experienced coal bursts) 467

72

Ptacek (2017)

1973–1992

50

27

Bräuner (1994)

Patyńska and Kabiesz (2014), Makówka (2016)

monitoring methods are based on geophysics, such as the electromagnetic emission method (EME), acoustic emission (AE) method, microseismic method and seismic computed tomography (CT) detection, and focus on monitoring the fracturing process in rock masses. Coal burst monitoring is usually conducted at two levels: regional monitoring and localised monitoring. Reginal monitoring focuses on the risk classification of coal mine. It is usually achieved by combining with comprehensive index methods and multi-factor coupling methods (Dou et al., 2014; Zhang et al., 2017). Localised monitoring methods are then performed in specific and targeted areas to achieve more accurate, reliable and real time monitoring and risk classification. Regional monitoring usually utilities microseismic monitoring and seismic wave tomography. Electromagnetic emission and acoustic emission monitoring are usually used for localised monitoring. These methods can lead to relatively large errors caused by the complicated underground conditions, such as underground water and a complex electromagnetic environment (Qu et al., 2011). In these conditions, test drilling/borehole drilling and roof displacement measurements are conducted for more accurate monitoring and forecasting. Currently, the most widely used monitoring methods in China are electromagnetic emission methods, acoustic emission methods, microseismic methods, borehole stress observation and test drilling methods. These methods have been implemented in high coal burst prone mines (Jiang et al., 2014) and some have been verified as successful monitoring by case studies (Dou et al., 2003; Jiang et al., 2006). Due to the complexity of coal burst, precursor signals and multiple monitoring parameters have been extensively studied and recognised as the future research direction for forecasting methods (Jiang et al., 2014).

as (1) causes persons to be withdrawn, (2) impairs ventilation, (3) impedes passage, or (4) disrupts mining activity for more than one hour. Coal burst occurs under various conditions. The previous literature review has discussed a range of contributing factors in great detail (Zhang et al., 2017). The major parameters are briefly summarised in this section, as the presence of these factors influences coal burst control strategies. The major contributing factors can be classified into two categories: (1) Coal seam and surrounding strata conditions: one of the most important factors causing coal burst is the existence of massive roof and floor layers (Bräuner, 1994; Iannacchione and Zelanko, 1995; Karfakis and Wu, 1995; Mark and Gauna, 2016). Coal seam in this condition is referred to as a “bump sandwich” (Mark, 2016). The massive roof layers would highly likely lead to an irregular periodic weighting, which results in burst-potential seismic activities (Iannacchione et al., 2005). There are also other localised parameters contributing to coal burst at various levels, such as cover depth and variable thickness of the coal seam, which are assessed based on field studies and numerical simulations (Osterwald et al., 1993; Bukowska, 2006; Dou et al., 2009). (2) Geological discontinuities: folded and faulted areas are always highly stressed and vulnerable to coal bursts (Holland, 1958; Gay, 1993; Iannacchione and Tadolini, 2008a; Alber et al., 2009). Other types of geological structures, such as dykes and sandstone channels (Salamon, 1983; Galvin, 2016), also have a pronounced impact on coal burst occurrences. The occurrence of coal burst is highly complex due to varying geological, geotechnical and mining conditions. For instance, data from China (2004–2014) showed that 38% of coal bursts occurred in weak coal seams and some even occurred in coal mines with no significant burst history (Jiang et al., 2014). Mark (2016) stated that burst proneness has no strong relationship with the composition of coal. Hence, in most conditions, one type of factor does not fully contribute to a burst event; therefore, prediction and control of coal burst should use analysis of various potential contributing factors according to the site specific conditions. These parameters are used as indexes to quantify the coal burst proneness, such as the uniaxial compressive strength (UCS or RC), elastic strain energy (WET), bursting energy (KE), dynamic failure duration (DT) and energy release rate (ERR) (Heasley, 1991; Linkov, 1996; Mitri et al., 1999; Mazaira and Konicek, 2015; Cai et al., 2016). Coal burst monitoring aims to understand the signatures of seismic activities, stress changes and geological conditions at a specific location. Monitoring techniques can be classified into two main groups (Jiang et al., 2014). The first technique aims to monitor the process of deformation of the excavations and stress redistributions. Usually, stress and displacement monitoring instruments are used, such as extensometers, tell-tales, borehole stress observation systems and load monitoring system of hydraulic roof supports. The second group of

3. Coal burst control strategies Coal burst control is an important element of overall coal burst management, as it directly relates to mine safety and productivity. Researchers and operators have studied control measures for rock or coal burst for decades. Current control techniques can be classified into two groups: preventative controls and mitigating controls. The preventative controls are usually implemented at the start of underground mines to avoid occurrence of coal bursts by optimising the mine design, while mitigating controls are applied as risk mitigation measures to minimise the risks of coal bursts. 3.1. Preventative controls Preventative controls, or mine design optimisation to prevent coal bursts, include mine layout design, pillar design, and protective seams in multiple seam mining. These control measures aim to avoid high static stress concentration and reduce the magnitude of dynamic events induced primarily by strata breaking. Therefore, during mining activities, the accumulated strain energy would distribute more evenly around the excavations after the implementation of preventative controls. In this section, gateroad design, critical pillar design and other layout designs are discussed. 130

Tunnelling and Underground Space Technology 81 (2018) 129–143

C. Wei et al.

Fig. 1. Yield-chain-abutment gate entry designs used in mines with high methane emission rates and coal burst potential, modified from Iannacchione and Tadolini (2008a).

stable“ designs (Fig. 2b) place a row of yield pillars next to the tailgate to reduce coal burst risks in gateroads. However, this layout can be ineffective in mitigating the burst risks in the maingate. Therefore, a two-entry yield pillar system (Fig. 2c) is utilised in relatively deep mines (up to 600 m). Where depth is greater than 600 m, a barrier pillar that is 90–180 m wide is used outside the yield pillars to resist abutment pressure as shown in Fig. 2d (Iannacchione and Tadolini, 2008b). It is of note that some United States mines do not allow mining activities greater than 900 m depth (Mark, 2016), especially where the coal seams are located between massive and competent strata layers or there is a history of coal bursts. With respect to pillar design, Koehler et al. (1996) defined a critical pillar as “one that is too large to either yield non-violently or yield before the roof and floor sustain permanent damage but is too small to support full longwall abutment loads”. Fig. 3 shows that yield pillars (with width-to-height ratios usually less than 4) seldom burst because they yield under relatively low abutment loads before a large amount of strain energy is stored (DeMarco et al., 1995). As for the critical pillars, the width-to-height ratios are usually greater than 7 but less than 15, and they have the highest potential to build up a large amount of strain energy, which is then released in a violent manner (Campoli et al., 1989; Mucho et al., 1993; DeMarco, 1994). The pillar has an elastic core to store a large portion of the strain energy when the width-toheight ratio is greater than 4 (Wang et al., 2016c). However, a width-toheight ratio of 4 or 5 has a better resistance to dynamic loading than a

3.1.1. Gate entry and pillar design In longwall mining, gate entry design is considered to be one of the major preventative controls. Gate entry pillars have two functions: (i) to provide enough fresh air to guarantee the ventilation of headings and the longwall face; and (ii) to resist the abutment loads. Therefore, appropriate mine layout design can allow the gate entry pillars to deform and yield under the abutment loading thereby reducing the stress concentration at the longwall face. Figs. 1 and 2 present an example of gateroad system design in United States coal mines. Three rows of pillars (Fig. 1a) are conventionally used in coal mines to provide more fresh air intake than the two rows of pillars. The two rows of pillars (Fig. 2a) help to control spontanous combustion in a bleederless longwall system with U-shaped ventilation in western United States mines (Iannacchione and Tadolini, 2008a). In Fig. 1b, “yield-stable-stable” pillar gate entry design mainly protects operators in tailgate entries. The yield pillars next to the tailgate are able to converge enough to redistribute the abutment load on to adjacent larger pillars or unmined ground. In Fig. 1c, yield pillars are used on either side of a stable pillar. In relatively deeper coal mines, the “yield-stable-yield” pillar design (Fig. 1d) was implemented to resist the higher abutment load (Campoli et al., 1990; Hendon, 1998). In this design, the previous chain pillars are replaced by abutment pillars which have a larger pillar size. The layouts shown in Fig. 1 are adapted at moderate cover depths, i.e., 300–600 m. In the two row pillar system (i.e., triple gate road entries), the “yield-

131

Tunnelling and Underground Space Technology 81 (2018) 129–143

C. Wei et al.

Fig. 2. Gate entry designs used in mines with spontaneous combustion and burst potential, modified from Iannacchione and Tadolini (2008a).

and high magnitude of strain energy is alleviated after the removal of large pillars (Zhang et al., 2017). Other than that, roadway stagger layout has been conducted successfully in coal mines to reduce the high burst risks of the gate entries along the goaf side (Li et al., 2016; Wang et al., 2017). 3.1.2. Other mine design considerations There are also other aspects that need to be considered for mine designs in different mining systems. In multiple-seam mining, inappropriate layout is a contributing factor to coal burst (Mark et al., 2007). Mining beneath remnant pillars, such as bleeders, chain pillars, barriers and stop lines of upper coal seam, has a high coal burst risk, especially when combined with depth and massive and competent roof and floor strata. Therefore, these conditions should be avoided (Mark et al., 2007). The protective coal seam mining concept aims to reduce the high static stress concentration in burst prone coal seams by premining the adjacent seams, which have minimal coal burst proneness. Mine operators should avoid remnant pillars left in the protective coal seams, and a monitoring program should be implemented if large pillars have to be left to avoid for various reasons, e.g., geological structures. In other forms of underground excavations, the principle of layout design is to avoid stress concentration. The best direction of roadway is to allow the ratio of σn (normal stress of the ribs of roadway) to σv (vertical stress) to be equal or approximately 1 to avoid stress concentrations. Pan et al. (2012) suggests that:

Fig. 3. The transition from successful yield pillar systems, through unsuccessful critical designs, to successful abutment pillar system (Koehler et al., 1996).

ratio of 2, according to both numerical simulations and practical experiments conducted in China (Wang et al., 2016c). For abutment pillars, the ratios are usually larger than 20; therefore, they can have the capacity to resist high levels of abutment loads and have a large stable pillar core. But the edges of the abutment pillars can still have burst potential, especially when the longwall face is relatively wide (e.g., > 300 m) with a high level of stress concentration (Iannacchione and Tadolini, 2008a). There are other gate entry designs and pillar designs to reduce the coal burst risk. The pillarless mining method is widely used in China and Europe. This method avoids high strain energy to be stored in the ‘elastic core’. In this method, concrete are positioned along the side of longwall panel to build artificial roadways. High stress concentration

1. When σHmax (the maximum horizontal principal stress) > σHmin (the minimum horizontal principal stress) > σv, the best option is σn = σHmin so that the ratio (σn/σv) is closest to 1, which means the direction of roadways should be parallel to the direction of σHmax; 2. When σHmin < σv < σHmax, the roadway direction should have an 132

Tunnelling and Underground Space Technology 81 (2018) 129–143

C. Wei et al.

angle α to the direction of σHmax, where the ratio (σn/σv) equals to 1. The angle α is calculated by Pan et al. (2012), as it is given below:

α=

kinetic energy on the support system. In practice, this increase can be achieved by softening the laminated layers by water infusion (Pan et al., 2013). Energy absorption becomes more effective when adding the same amount of water to the overlying strata rather than to the support system. Other than that, the peak particle velocity (PPV) has been widely used to calculate the energy absorption capacity for support system (Kaiser, 1996; Kaiser and Cai, 2012; Zhang et al., 2015). As for a seismic activity, the total energy of the ejected coal or rock mass mainly consists of kinetic energy and potential energy, as it is expressed in Eq. (1) (Kaiser and Cai, 2012). The energy absorption capacity of the support system should be larger than the total energy. However, some factors, such as the interaction of ground motion and support system, the evolution of PPV around the excavation, are extremely complex. Accurate energy absorption capacity for support system is still need much more research (Potvin and Wesseloo, 2013).

1 σ + σHmin−2σv arccos Hmax 2 σHmax −σHmin

3. When σv > σHmax, the best option is σn = σHmax, making the ratio (σn/σv) closest to 1, which means the roadway direction should be perpendicular to the direction of σHmax. Moreover, the shape and direction of an excavation determines the stress redistribution around the excavation. For instance, a circular shape excavation usually has a more homogenous stress distribution than a square shape excavation (Kaiser and Cai, 2012). 3.2. Mitigating controls Burst risks still exist in some cases even preventative controls are implemented. Then mitigating measures should be conducted to minimise the impact of coal burst including operational controls and administrative controls. Administrative controls refer to measures that limit the exposure of operators to high burst risk areas (Varley and Whyatt, 2008), such as autonomous mining, positioning remote-control equipment, and allowing only a minimum number of operators in an active mining area. The following sections mainly focus on operational controls. Various techniques have been used in coal mines for burst risk mitigation, including shotfiring, hydraulic fracturing, water infusion, stress relief boreholes, bump cutter and gas explosive charges. The purpose of these controls is to introduce a weak point to relieve the stress or soften the system in a controlled manner. The underlying principles of destress drilling in inducing impact damage in a confined state are not yet fully understood.

E=

1 2 mv + mgd 2

(1)

where E is the total energy; m is the mass of the ejected coal or rock; v is the corresponding peak particle velocity (PPV); g is the gravitational acceleration; d is the displacement of the ejected coal/rock. 3.2.1.2. Yielding support. In a coal burst event, it is unfeasible to avoid deformations in roadways but it is possible to reduce the high stress concentration to an acceptable level. A yielding support is better suited than a stiff support in burst risk areas (Mazaira and Konicek, 2015). Yielding support was introduced in South Africa by Ortlepp (1968). The test verified that the yielding support system was able to withstand the dynamic load induced by blasting that destroyed the adjacent conventional support system (Ortlepp, 1992). The destruction occurred because the dynamic load was higher than the loading capacity of conventional stiff support elements (Jager et al., 1990; Ortlepp and Stacey, 1994). However, the artificial blasting was a limited approach as it could not realistically simulate the rock burst process. Practical cases were tested in Australia which proved that the yielding support was more effective than the conventional system in burst risk areas (Turner and Player, 2000). A yielding support system can absorb the dynamic energy by its large strain capacity, and extend the energy absorption capacity by the pre-designed slipping out process on the second deform stage (Cai, 2013), as seen in Fig. 4. This function can be achieved by special cable bolts, such as conebolts and friction bolts (their large deformation capacity is shown in Fig. 5), and the energyabsorption bolts described below. A special roof bolt was developed by Wang et al. (2016b) to improve support effectiveness based on the above energy absorption theory. As illustrated in Fig. 6, this roof bolt has three parts (from left to right): bolt and pre-stressed cable, energy absorption sleeve with built-in hexagon pipe to connect the bolt and cable, and energy absorption devices at the bolt tail. The work process is described in Fig. 4: the pretensioning and grouting assist to resist the static loading; and when dynamic failure occurs, the energy absorption sleeve is extended (maximum displacement is 200 mm). This bolt is able to absorb approximately 28 kJ of released strain energy and resist 140 KN of static load. When the energy absorption capacity of the bolt is exceeded, the bolt would produce large axial displacement (60 mm of largest compressed displacement and resistant load is 150 KN) to protect the bolt and cable from cracking. McCreath and Kaiser (1992); Ortlepp (1992) suggest that an energy-absorption bolts designed for seismic events should have a capacity to deform 200–300 mm to compromise the large deformation in high stress environments. The appropriate combination of stiff and soft support is critical for a support system in coal burst prone conditions. Pan et al. (2013) suggested that the impact of dynamic loading from overlying strata decreases in an exponential fashion with the help of stiff support. The ground stability is enhanced by increasing the stiffness of the support

3.2.1. Ground support system Due to the unpredictability of coal burst occurrences, ground support is usually the final line of protection to ensure safety in high risk zones (Cai, 2013). Conventional support systems mainly focus on the rock mass deformation caused by gravity (Kaiser and Cai, 2012), including roof falling, bulking, floor heave and superficial unravelling, which is different from the primary purpose of a support system for coal burst. Implementing ground support for coal burst control aims to resist sudden and dynamic failure of rock mass and/or avoid large deformation in excavations (Cai, 2013; Jiang et al., 2014). The following section summarises the critical characteristics and considerations for the ground support system in burst prone areas. 3.2.1.1. Energy absorption. One of the critical considerations for coal burst support is to ensure that the support systems can accommodate the levels of energy release in a coal burst. Therefore, as a first step, the expected energy levels and the energy absorption capacity need to be identified. Wang and Pan (2015) stated that the energy absorption should rely more on the rock mass itself than on the support elements. This is similar to the ground support reinforcement theory which states that strengthening the surrounding rock mass by itself is more important than by support materials. There are many examples of “yielding support elements” used in deep metalliferous mines, particularly in Australia, Canada and South Africa. Analytical and numerical approaches are used to help quantify the energy absorption of the support elements under different conditions. Wang and Pan (2015) analysed a dynamic model with a group of laminated layers and support elements, and studied the most efficient way to improve energy absorption capability of the support elements. They found that an increase in water inflow (referred to as of viscous damping) between two neighbouring laminated layers helps reduce the 133

Tunnelling and Underground Space Technology 81 (2018) 129–143

C. Wei et al.

Fig. 4. The advantage of yielding support in burst-prone conditions (Cai, 2013).

and yielding elements together, and also be able to absorb the released energy instantly when dynamic failure occurs. Another similar but simple support system was suggested by Lu and Pan (2012). It has an energy absorption steel bracket, which is positioned between the rock mass and the steel beams. The capacity to reduce shock wave pressure by the energy absorption mechanism was quantified by experimental and theoretical calculations. The experiment results showed that 50% of shock wave pressure can be reduced and more than 70% of energy induced by the dynamic experiment can be absorbed using this system. Apart from rock bolts, the combined surface elements, such as shotcrete reinforced by fibre/mesh/strap, also help to increase the support performance (Ortlepp and Stacey, 1994; Cai, 2013). They effectively redistribute the dynamic loading more evenly and allow the rock bolts to withstand more loading, as verified in practical cases described by Milev (2005). Nevertheless, the bolt-mesh linkage and the retaining elements themselves are often the weakest part of the system so they need extra attention. 3.2.1.3. Support density. The support layouts need to be optimised in burst prone areas. To avoid strain burst, it is important to enhance the critical buckling strength of the excavation by increasing support density: firstly, enlarging the size of the face plates and installing more bolts to reduce the spacing, then implementing shotcrete with fibre/mesh to strengthen the support system (Ortlepp and Stacey, 1994). However, with the increase of support density, the stored strain energy increases and this may result in a more destructive burst event, especially when geological structures are close to the installation position (Kaiser and Cai, 2012; Cai, 2013; Wang et al., 2016a). Wang et al. (2016a) attempted to quantify the relationship between the bolt spacing, dynamic loading and the released energy and showed that the spacing of bolts has a significant influence on plastic rock mass but little impact on brittle rock mass.

Fig. 5. Load displacement graph of various rock bolts (Cai, 2013).

3.2.1.4. Pre-tension. The influence of pre-tensioning of tendons in coal burst control has been discussed extensively. Some statements suggest that pre-tensioning should not be applied in burst risk areas (Stillborg, 1986; Mazaira and Konicek, 2015). However, under certain conditions, high pre-tension is suggested to protect the surrounding intact rock mass from fracturing and to maintain the strata in a compressed stress state rather than buckling, tensioning and shearing (Kang et al., 2015). Further investigations need to be conducted to study the influence of pre-tensioning on support effectiveness under dynamic loading. Previous studies revealed that the failure of the support system and the deformation of excavations is not due to insufficient support capacity but because of the inappropriate application of support principles under dynamic conditions (Pan et al., 2013).

Fig. 6. Working process of the energy-absorption bolts device, translated from Wang et al. (2016b).

system in static loading environments but not necessarily in a dynamic loading environment. Comparatively, support effectiveness can be improved significantly by adding yielding elements. Based on this theory, a new energy-absorption hydraulic support was designed (Fig. 7), to be tested in roadways of Yima Coal Mine in China (Pan et al., 2013; Pan et al., 2014). This system consists of an arc top beam, a straight bottom beam, side link-bars and three hydraulic props with total load resistance of 6000 KN. Beams and props are used to increase the stiffness of the support system. The energy absorption elements are able to have a large amount of yield deformation during dynamic load tests. Thus, it is designed to appropriately combine stiff

3.2.2. Destressing techniques Destressing techniques aim to release the strain energy stored in 134

Tunnelling and Underground Space Technology 81 (2018) 129–143

C. Wei et al.

Fig. 7. New energy-absorption hydraulic support and its structure sketch map, translated from Pan et al. (2013).

Fig. 8. Stress-strain curves of intact and drilled coal samples, after Zhu et al. (2015). Fig. 9. Effects of destress blasting (Roux et al., 1957).

coal and rock mass by inducing fracturing and/or allowing the mass to yield without storing large amounts of energy. These techniques include destress blasting, destress drilling, hydraulic fracturing and water infusion. The related mechanisms are illustrated in Fig. 8, where the intact and drilled coal samples help to understand the destressing process. The blue solid line is the pre-peak stress-strain curve of an intact coal sample and the blue dashed line represents the post-peak line in curve 1. A second sample was drilled at point C (peak stress) until the sample stress decreased to an arbitrary point A. Then, reload the sample from point A to point B (peak stress of curve 2). Drilling of the sample reduced the peak strength from C to B. The hatched area shows the released energy during the destressing process.

The blasting boreholes can be classified into two types (Saharan and Mitri, 2011; Mazaira and Konicek, 2015): (i) boreholes within the coal seam, which aims to soften coal properties and (ii) boreholes in the host rock mass, which aims to pre-cut the competent roof layers for better caving. As regulations in many countries restrict the amount of explosives used in the coal (such as 1000 g per hole), blasting within rock mass is more frequently used. Konicek et al. (2011a) reported that in the Upper Silesian Coal Basin, more than 2000 boreholes within the rock mass were blasted using more than 3.5 million tonnes of explosives. The critical design parameters in rock mass blasting have been identified as the size and position of boreholes, the amount of explosives in the boreholes and the geological structures in and around the excavations. Konicek et al. (2012) provided some practical parameters as follow. The position of boreholes in maingates and tailgates are usually 30–100 m outbye of retreat longwall face. The suggested parallel spacing of boreholes is a distance of 5 m away from the face, increasing to 12 m when the boreholes are in close proximity to the longwall face. The bottom of the boreholes is usually 30 m above the coal seam. The explosives are plastic explosives within cartridges which account for 45–85% of the whole length of drill and pneumatic sand stemming. The charges in each stage are fired simultaneously without delays and the waiting time after blasting is 45–60 min. A typical layout of destress blasting holes is presented in Fig. 10. Table 2 presents the blasting parameters used in Poland and China. The length of boreholes mainly depends on the mining height, the rock strata conditions, the borehole angles, etc. In China, according to Dou (2017), it is typically 3–5 times the mining height in gateroads in front of longwall face. As for developments, it is 0.7–1 times the width of high stressed zones around excavations. The length of explosives accounts for more than half of the drilling holes with 1–4.5 kg/m of

3.2.2.1. Destress blasting. Destress blasting is a strain energy release method that is usually applied in high stress concentration environments by fracturing the targeted coal or rock zones. It is one of the most widely used mitigating methods, which was first introduced in deep gold mines in South Africa in the 1950s. Since then it has been used effectively in Europe, especially in the Upper Silesian Coal Basin of the Czech Republic. The purposes of destress blasting are to minimise the roof guttering, floor heave and rock or coal bursts in difficult geological and geotechnical conditions in underground mines (Kexin, 1995; Konicek et al., 2011b; Mazaira and Konicek, 2015). Although it is one of the oldest preconditioning and proactive measures, the mechanisms are still not well understood (Konicek et al., 2011b). This method is used to move the peak stress forward to the interior rock mass (Kexin, 1995), as shown in Fig. 9. Konicek et al. (2011a) stated that destress blasting in the Upper Silesian Coal Basin had two objectives: softening the high stiffness rock layers and decreasing their resulting elastic modulus; and releasing the stress of the targeted area. 135

Tunnelling and Underground Space Technology 81 (2018) 129–143

C. Wei et al.

Fig. 10. A typical layout of destress blasting used in Poland and Czech Republic (Konicek et al., 2011b).

emulsion explosives. The waiting time (after the blast) is more than 30 min and the waiting place is more than 300 m away from the blasting site. Destress blasting is only used in low gas mines (He et al., 2012). The blasting operators suffer from a large amount of misfires and security problems. The targeted rock mass is sometimes vulnerable to over explosion. Therefore, this technique should be used with caution and combined with other mitigating methods.

Table 2 Typical blasting parameters used in Upper Silesian Coal Basin (Konicek et al., 2011a, b) and China. Parameters

Borehole diameter

Inclination

Length of borehole/ average

Spacing

USCB (Czech) China

75–105 mm 42–50 mm

10–30° –

(30–120) 70 m (3–5) H (mining height) in working face (0.7–1) W (width of stressed zone) in gateroads

5–12 m 5–20 m

3.2.2.2. Destress drilling. Destress drilling has been widely used as a preconditioning technique for coal burst control. Its mechanisms are similar to destress blasting which aims to soften the targeted area and move the high stress concentration further away to prevent coal burst.

18304 Tailgate

LEGEND:

Fig. 11. Layout of large diameter destress drilling. 136

Tunnelling and Underground Space Technology 81 (2018) 129–143

C. Wei et al.

easily flow along the discontinuities rather than the designated direction. The purpose of the fracturing is to provide artificial discontinuities to crack and soften the roof in a controlled fashion so that the sudden break of roof strata is avoided. Two types of artificial fracturing are created in this technique (Du et al., 2012): the first type aims to let the roof break along the directional fracture, and the second type is to induce a horizontal cracking to reduce the caving thickness of the overlying strata. There are two types of positions of initial slot that help achieve the purpose of the technique: in the main roof but close to the immediate roof; and two parallel and horizontal slots connected to each other by the hydraulic fracturing in the middle of the main roof (Qin and Dou, 2014). The effectiveness of hydraulic fracturing have been verified using seismic monitoring in field tests, which showed that the frequency and magnitude of released energy decrease in areas where this technique is used (He et al., 2012). However, key parameters in hydraulic fracturing need to be further studied including the influence of the initial slot angle, the range of broken roof strata after fracturing and the direction of the injected liquid when propagating.

Table 3 Parameters of large diameter drilling used in China (Dou, 2017). Parameters

Diameter/mm

Depth/m

Spacing/m

Drilling (China)

100–150

15–20

1–1.5

The effectiveness of destress drilling on mitigating coal burst risks has been proved by field monitoring, numerical simulations and experiments (Liu et al., 2014). As indicated in a laboratory testing of drilled rock samples (Liu et al., 2012), the uniaxial compressive strength and elastic modulus of rock samples are reduced by 27.3% and 19.6%, respectively, after drilling; and the acoustic emission is lowered and cracked pieces smaller with the increase of the number of drilling holes in the rock samples. Destress drilling is regularly used at a number of mines in China to mitigate the coal burst risks by using large diameter (≥100 mm) drill holes. A typical layout of the drilling is shown in Fig. 11. Currently, the drilling parameters (diameter, spacing and depth of hole) are determined by risk assessment, numerical modelling and empirical experience in most cases. For example, as shown in the following equation, energy dissipation index (XE) is used to help quantify the drilling parameters. In their study, Zhu et al. (2015) suggested 40%, 29% and 23% of XE respectively in high, medium and low coal burst risk areas. Energy dissipation index is also used to determine the diameter and spacing of drilling holes. Common drilling parameters used in China are shown in Table 3. The diameter of drilling holes is usually 100–150 mm, which is based on coal burst risks and associated stress build-up. The depth of borehole should allow the drilling hole to reach the peak stress zone in the ribs.

3.2.2.4. Water infusion. Water infusion is an effective control in mitigating coal burst risks by increasing the moisture content of coal (Wu and Wang, 1989; Dubinski, 1994; Calleja and Porter, 2016). It has been considered the first choice of mitigating measures at many mines (Song et al., 2006), due to its simplicity, low cost and multiple additional functions, such as dust reduction, cooling and rock mass softening (Zhang et al., 2003). Water infusion can mitigate coal bursts by softening the coal mass and lubricating existing fractures, joints and cleats. Experiments showed that increasing moisture content by 3% decreased the uniaxial compressive strength of coal by 32% (Wu and Wang, 1989). Cohesion, internal friction angle, elastic modulus and pre-peak strain energy also decreased to some extent (Su et al., 2014; Calleja and Porter, 2016) and post-peak modulus became relatively smoother and flatter (Song et al., 2006). Xia et al. (2015) used an alkaline solution to weaken coal samples, which indicated that the infusion of this solution was more effective than water-only infusion. However, it should be noted that the softening effectiveness is strongly related to the time taken to moisten the coal seam. Pre-existing free gas in boreholes and cracks is gradually replaced by water in the moisture process (Song et al., 2006). Based on this replacement theory, the time and amount of water needed to be injected to thin and thick coal seams can be derived theoretically (Song et al., 2006). Su et al. (2014), based on their practical coal sample experiment tests in Qianqiu Coal Mine in China, found that the mechanical properties and burst proneness indexes decreased significantly after 7–10 days of saturation, but these parameters stayed relatively stable afterwards. It is suggested by many authors that it is necessary to conduct a pre-water infusion test in mines, considering the time effect and infusing position of this technique. Water infusion can also change the stress redistribution to reduce burst risks. A series of numerical simulations (Wu and Wang, 1989; Shen and Deng, 2013) indicated that water infusion can reduce the peak abutment pressures, and moved it to the interior of the coal pillars. The width of the yielding pillar zone increased and the level of stress concentration reduced. Moreover, energy is released more steadily and evenly in both time and space during dynamic events after water infusion (Wu and Wang, 1989). The effectiveness of water infusion is greatly influenced by the permeability and the stress state of the targeted coal and rock mass (Dubinski, 1994; Klishin, 2006). This method is ineffective when the targeted zone is too competent, impermeable and highly stressed. In addition, the time taken to moisten the coal is another shortcoming because it may need a few days or even several months to get the best moisture effectiveness (Calleja and Porter, 2016), although the new

XE = ΔS /(ΔS + ΔS1) where XE represents the energy dissipation index ΔS is the released strain energy in destressing process, ΔS1 is the residual strain energy. Inappropriate design of drilling parameters can result in unexpected problems. For example, small diameters and large borehole spacing are easy to operate but may not be effective sufficiently to mitigate the risks. Drilling with large diameters and small spacing can, on the other hand, cause: (1) large deformation or even destruction of excavations (Liu et al., 2012; Zhu et al., 2015), using destress drilling holes of 125 mm in diameter, 1 m spacing and 25 m borehole depth; and (2) triggering of larger coal bursts (Iannacchione and Zelanko, 1995), using 61 cm diameter drill holes. In addition to the determination of appropriate design parameters, drilling in an extremely weak or extremely hard coal seams and relationship between the drill hole and the ground support densities in roadways need to be studied further (Zhu et al., 2015). 3.2.2.3. Hydraulic fracturing. Hydraulic fracturing technique is traditionally implemented to fracture coal seams and host rock, in order to enhance permeability, improve top coal cavability and mitigate coal burst (Huang et al., 2007). He et al. (2012) proposed a directional hydraulic fracturing technique, which is mostly used for fracturing the competent overlying strata to release the stored energy in the roof. Typical layouts of directional hydraulic fracturing technique are illustrated in Fig. 12: firstly, an initial slot is cut out in the borehole by a special purpose slot cutter (Fig. 13a), and then high pressure liquid is injected into the initial slot to conduct a directional fracturing as presented in Fig. 13. The high pressure liquid has 40 MPa maximum pressure capacity and more than 80 L/min of flow rate, and it is injected along the desired direction in a short time period of time (approximately 30 min). The boreholes are suggested to have a depth of 10–20 m and diameter of 46–48 mm. The perimeter of fractured zone can be extended to 15–25 m (Qin and Dou, 2014). It should be noted that this technique is conducted in competent roof layers instead of layers with existing discontinuities, as the high pressure liquid can 137

Tunnelling and Underground Space Technology 81 (2018) 129–143

C. Wei et al.

(a) Planar picture of drilling holes layout

(b) B-B profile of drilling holes layout

Fig. 12. Layout of directional hydraulic fracturing holes (He et al., 2012).

alkaline solution can improve the speed of the moistening process and the effectiveness of softening (Xia et al., 2015). In China, the parameters used in this technique are: the increased moisture content should be more than 2% (where the moisture content of the original coal seam is higher than 5%) or 3% (where the moisture content of the original coal seam is less than 5%); spacing of water drill holes should be 10–20 m in longwall faces and 1.5 m in development in roadways (Dou, 2017). Therefore, water infusion is able to reduce burst risks but in limited conditions where (Yan and Ning, 2000; Kang et al., 2004; Qi and Dou, 2008; Shen and Deng, 2013): (1) coal seams are porous and permeable; (2) coal seams are relatively stable to ensure the completion of drilling and integrity of boreholes; (3) there are no geological discontinuities in the vicinity of the targeted area; and (4) the roof and floor are not the rocks that are softened in the moisture process.

Fig. 14. Layout of test drilling.

3.3. Test drilling for control measures

to 3 m is slumped or crushed in their drilling holes. But in some other cases, the data from 1 to 3 m is also used. In this method, the volume of cuttings per unit, particle size and the position of maximum volume reflect the stress state of the coal pillar. For example, Chen et al. (2013) proposed a risk classification method based on those three parameters. They proposed the risk classification that is presented in Table 4, where M is the total volume of material that can burst into the roadway, and the value of peak-volume position refers to the length of drilling when the cuttings volume per metre reaches the maximum. Qu et al. (2011) attempted to quantify the relationship between the

Test drilling is a monitoring method that is widely applied to identify high burst risk areas by using small diameter bores (usually 42–50 mm) (Bräuner, 1994; Varley and Whyatt, 2008; Qu et al., 2011; Chen et al., 2013). Both the drilling density and diameter are smaller than in destress drilling. The boreholes are usually drilled along the longwall face and gate entries. In China (Chen et al., 2013), the length and spacing of drilling holes are 10 m, and the position of the borehole is 1.0–1.3 m above the floor. A typical layout of test drilling is shown in Fig. 14. The cuttings volume should be accounted properly in practice. Chen et al. (2013) began the volume test from 4 to 10 m, as the area of 1

(a) Initial borehole slot cutting machine

(b) Essence of directional fracturing

Fig. 13. Process of directional fracturing (He et al., 2012). 138

Tunnelling and Underground Space Technology 81 (2018) 129–143

C. Wei et al.

monitoring systems should be used (Jiang et al., 2014). Intensive destressing measures help to remove peak stress concentration on the inner side of coal pillars or barriers. Intensive monitoring and assessment includes real time online monitoring, real time online risk classification and risk level assessment following the destress work. Extra ground support elements, such as energy-absorption cables, conebolts and friction bolts, are critical in mitigating high risk zones. Personal protective equipment (PPE), such as body shields and helmets, should also be assessed for different risk zones. This paper has evaluated a number of control techniques, but there are also other measures that are implemented at individual mines. A summary of these controls is presented in Table 5.

Table 4 Risk classification based on test drilling around the world (Chen et al., 2013; Rob, 2016).

Cuttings volume

Low risk

Moderate risk

High risk

China Germany

– –

Poland

–

2.5–3.0 kg/m > 6 L/m within 3M > 6 L/m within 3.4 M 6–8 m > 30%

– > 15 L/m within 3 M > 6 L/m within 1.5 M 4–5 m –

Peak-volume position (China) Percentage of particle size larger than 3 mm (China)

>8 m < 30%

cuttings volume, the stress state of boreholes and the abutment pressure through field tests and numerical simulations. Based on their results, they proposed a real time online risk classification that has been applied in some coal mines in China. However, the effectiveness of the test drilling method is poor when: (1) the stressed area is too dangerous to operate; (2) periodic and dynamic abutment loading can impact pillars among the test intervals (Varley and Whyatt, 2008); and (3) coal bursts occur away from the drilling area (Boler et al., 1997; Jiang et al., 2011). In these conditions, this technique should be used with other monitoring techniques to achieve the risk classification of the targeted area.

4.2. Coal burst management plan A three-stage coal burst management framework is proposed in this section. As coal burst management is highly complex and there is no approach that is applicable to all operations, the purpose of this framework is to emphasise the critical considerations in a coal burst management plan for site specific applications; it is not to develop a generic management plan. The three stages include: (1) Identification of coal burst profile (2) Development of coal burst management plan (3) Management of coal burst.

4. Discussion

Fig. 15 presents the steps to identify coal burst risk at a mine site. The BurstRisk classification system (Vardar et al., 2018) can be used to identify the potential burst prone area, which is a coal burst risk ranking method based on back analyses of 24 development cases and 23 longwall cases from Australia, China and the United States to define three risk categories of low, medium and high. Using the BurstRisk system, operators can compare their mine conditions to known cases of coal burst and no coal burst and then assess the proneness of their operation so that appropriate controls can be implemented if there is potential for coal burst. The second stage follows a logical framework to develop the coal burst management plan (Fig. 16). The review of legislation should include the requirements of principal hazard management plans and all associated considerations. BurstRisk-based geotechnical

4.1. Recommendation for coal burst controls To optimise the control effectiveness, two aspects should be considered: (1) the preferred control strategies need to be aligned with the hierarchy of controls, being: elimination, substitution, engineering control, administrative controls and personal protective equipment (Manuele, 2005; Trade and Investment, 2015); (2) real time evaluation of control effectiveness should be conducted to assess the effectiveness and to ensure safety. One of the most widely used real time evaluation techniques is the test drilling method as mentioned in Section 3.3. Moreover, the primary concern of coal burst controls is to balance the destressing intensity, ground support capacity and magnitude of potential coal burst. In high risk areas, intensive destressing techniques, and support and Table 5 Summary of control measures. Development Mitigating controls

Administrative controls

Operational controls

Preventative controls

Prevention controls

Longwall face

only the minimum number of people in the areas where • Allow coal is being mined remote control equipment operators as far from the • Position active mining as practical remote mining • Use re-entry • Delay training • Provide yielding roof/rib support system in critical areas according • Use to the energy absorption rate development rate • Reduce physical barriers and personal protective equipment • Use directional hydraulic fracturing • Use water infusion • Use deep-hole relief blasting (preconditioning) • Use large borehole drilling • Use • Use personal protective equipment pillar design (i.e., intact or yield) • Use mine layout design • Consider pillarless mining • Use mining sequence • Consider • Use protection seams in multiple seam mining

139

only the minimum number of people into the • Allow areas where coal is being mined remote-control equipment operators as far • Position from the active mining as practical automation • Implement re-entry • Delay training • Provide the shear speed • Reduce the web depth • Reduce uni-directional • Cut double cuts at the gate ends • Avoid physical barriers (including belts secured on to the • Use shields) large borehole drilling • Use hydraulic fracturing • Use water infusion • Use • Use personal protective equipment pillar design • Use mine layout design • Consider mining sequence • Consider gateroad system design • Consider pillarless mining • Use • Use protection seams in multiple seam mining

Tunnelling and Underground Space Technology 81 (2018) 129–143

C. Wei et al.

Fig. 15. Stage 1–Identification of coal burst profile.

Fig. 16. Stage 2–Development of coal burst management plan.

consequences of the event. The absence or failure of a critical control would significantly increase the risk despite the existence of the other controls. In addition, a control that prevents more than one unwanted event or mitigates more than one consequence is normally classified as critical (ICMM, 2015). As for the “communication” box, roles and responsibilities for determination and modification of control measures should be defined.

characterisation is to identify appropriate controls and monitoring techniques to be implemented in different risk zones. The review of monitoring techniques is provided in Section 2 and control measures are summarised in Table 5. In the third and last stage, a coal burst management strategy is presented in Fig. 17. Coal burst is a complex phenomenon; despite decades of research, the source and mechanics of bursts are imperfectly understood, and the means to predict and control remain elusive (Mark, 2016). Therefore, every step of a coal burst management plan should be comprehensively risk assessed to ensure the preferred monitoring and control techniques are appropriate and effective. In the box of “develop a critical control management strategy” critical control is defined as a control that is critical to prevent the event or mitigating the

4.3. Knowledge gaps The paper identified several knowledge gaps in the understanding and implementation of coal burst controls, as listed below: 140

Tunnelling and Underground Space Technology 81 (2018) 129–143

C. Wei et al.

Fig. 17. Stage 3–Management of coal burst.

• Every specific coal mine has its own geological and mining condi•

•

•

or similar systems are relatively effective. In Polish coal mines, gateroads and yielding steel arches are used in development. It should be noted that in the overall support strategy, any weaknesses in the support system can lead to the failure of the support system at low energy levels. It is recommended that the concept of a yield support system should be examined, at least to be used in high risk zones and possibly some type of different surface line or surface design system to work in conjunction with a yielding system. A coal burst management framework was proposed that includes three stages of identification of risk, development of a management plan and management of coal burst. Current knowledge gaps were identified and discussed. For implementation of a successful control technique, systematic methods should be developed to recognise and address all the complex contributing factors that may be found in a specific mine. Establishment of control strategies must be based on a monitoring program and risk classification technique. Prior to the implementation of coal burst prevention and control techniques, a clear understanding of the driving geological and geotechnical factors and the underlying mechanisms should be gained. The preferred coal burst control technology should also be in line with the mining experience, regulations, mine design and operational practices. Otherwise, failure of the control technology is inevitable, with impacts from coal bursts on mine safety and productivity.

tions, so there are no universal control techniques that are able to prevent all types of coal bursts. Operators must consider how to combine these measures and use them in a more scientific way. Mechanisms of control techniques need to be further studied. There are sufficient control measures to mitigate the impact of coal bursts but sometimes in an unknown way. For example, the over drilling case in Xinjulong mine in China (Liu et al., 2012) and adverse effects of over blasting (Mazaira and Konicek (2015) reflected that mitigating techniques cannot be used properly without an in-depth understanding of critical corresponding parameters. Some methods referred to in this paper are uniquely used in a specific country, such as the large diameter drilling (100–150 mm) method applied in China. Similarly, the pillar design used in the United States cannot be used directly in underground mines with different gate entry systems. Therefore, future studies should be conducted to optimise those techniques in different geological and geotechnical environments. There is a need to establish a quantitative relationship between risk classification results and selecting methods of control measures. As an example, in Section 3.2.2.2, the diameter of drilling boreholes is determined by combining risk classification results.

5. Conclusion References This paper reviewed and evaluated current coal burst control techniques. The role of mine design, mine layouts, the role of destressing and the role of ground support were discussed as effective coal burst controls. The United States experience suggests that yield bolt systems plus surface constraint (via) mesh, (via) large diameter plates

Alber, M., Fritschen, R., Bischoff, M., Meier, T., 2009. Rock mechanical investigations of seismic events in a deep longwall coal mine. Int. J. Rock Mech. Min. Sci. 46 (2), 408–420. Boler, F., Billington, S., Zipf, R.K., 1997. Seismological and energy balance constraints on

141

Tunnelling and Underground Space Technology 81 (2018) 129–143

C. Wei et al.

monitoring of rock burst in coal mine. Chinese J. Geophys. – Chinese Ed. 49 (5), 1511–1516. Jiang, Y., Pan, Y., Jiang, F., Dou, L., Ju, Y., 2014. State of the art review on mechanism and prevention of coal bumps in China. J. China Coal Soc. 39 (2), 205–213. Kaiser, P.K., 1996. Canadian Rockburst Support Handbook. Geomechanics Research Centre. Kaiser, P.K., Cai, M., 2012. Design of rock support system under rockburst condition. J. Rock Mech. Geotech. Eng. 4 (3), 215–227. Kang, H., Wu, Y., He, J., 2015. Rock bolting performance and field practice in deep roadway with rock burst. J. China Coal Soc. 40 (10), 2225–2233. Kang, T., Zhang, J., Bai, S., 2004. Theoretical study and application of weakening top coal using water pre-infusion in fully mechanized sublevel caving mining. Chinese J. Rock Mech. Eng. 15, 024. Karfakis, M.G., Wu, X., 1995. Prediction of bump hazards associated with coal mining under competent roofs. In: Proceedings The 35th US Symposium on Rock Mechanics (USRMS), American Rock Mechanics Association. Kexin, D., 1995. Maintenance of roadways in soft rock by roadway-rib destress blasting. China Coal Soc. 20 (3), 311–316. Klishin, S.V., 2006. Destruction of rocks by directional hydraulic fracturing on the basis of models of plasticity with internal variables. In: Proceedings III European Conference on Computational Mechanics Solids, Structures and Coupled Problems in Engineering. Lisbon, Portugal, p. 25. Koehler, J., DeMarco, M., Wuest, W., 1996. Critical pillar concept in yield-pillar-based longwall gate-road design. Min. Eng. 48 (8), 73–78. Konicek, P., Konecny, P., Ptacek, J., 2011a. Destress rock blasting as a rockburst control technique. Proceedings 12th ISRM Congress. International Society for Rock Mechanics. Konicek, P., Saharan, M.R., Mitri, H., 2011b. Destress blasting in coal mining–state-of-theart review. Proc. Eng. 26, 179–194. Konicek, P., Soucek, K., Stas, L., Przeczek, A., 2012. Rockbursts provoked by destress blasting in hard coal longwall mining. In: Proceedings of the 10th International Symposium on Rock Fragmentation by Blasting, pp. 193–202. Li, Z., Dou, L., Cai, W., Wang, G., Ding, Y., Kong, Y., 2016. Roadway stagger layout for effective control of gob-side rock bursts in the longwall mining of a thick coal seam. Rock Mech. Rock Eng. 49 (2), 621–629. Linkov, A.M., 1996. Rockbursts and the instability of rock masses. Int. J. Rock Mech. Min. Sci. Geomech. Abstr. 33 (7), 727–732. Liu, J., Jiang, F., Sun, G., 2014. Mechanism of intensive venting pulverized coal to prevent coal burst and its application. Chinese J. Rock Mech. Eng. 33 (4), 748–754. Liu, J., Jiang, F., Sun, G., Lu, S., Zhang, D., 2012. Selection of advanced hydraulic support in gob-side entry of fully mechanized caving face of deep mine. Chinese J. Rock Mech. Eng. 31 (11), 2232–2239. Lu, X., Pan, Y., 2012. Theoretical analysis and experimental research on rockburst prevention mechanism of rigid-flexible-rigid supporting structure. Chinese J. Rock Mech. Eng. 31 (1), 52–59. Makówka, J., 2016. Rock burst: diagnosing and mitigating, Polish experience for Australian conditions. Focused International Workshop on Coal Burst Experience and Research Direction, UNSW Australia, Sydney. Manuele, F.A., 2005. Risk assessment & hierarchies of control. Prof. Saf. 50 (5), 33–39. Mark, C., 2016. Coal bursts in the deep longwall mines of the United States. Int. J. Coal Sci. Technol. 3 (1), 1–9. Mark, C., Chase, F., Pappas, D., 2007. Multiple-seam mining in the United States: design based on case histories. In: New Technology for Ground Control in Multiple Seam Mining, Pittsburgh. National Institute for Occupational Safety and Health (NIOSH), pp. 15–28. Mark, C., Gauna, M., 2016. Evaluating the risk of coal bursts in underground coal mines. Int. J. Min. Sci. Technol. 26 (1), 47–52. Mazaira, A., Konicek, P., 2015. Intense rockburst impacts in deep underground construction and their prevention. Can. Geotech. J. 52 (10), 1426–1439. McCreath, D., Kaiser, P., 1992. Evaluation of current support practices in burst-prone ground and preliminary guidelines for Canadian hardrock mines. Rock Support in Mining and Underground Construction, Proceedings of the International Symposium on Rock Support, Ontario, Canada, pp. 611–619. Milev, S., 2005. Strong ground motion and site response in deep South African mines. J. South Afr. Inst. Min. Metall. 105 (7), 515–524. Mitri, H., Tang, B., Simon, R., 1999. FE modelling of mining-induced energy release and storage rates. J. South Afr. Inst. Min. Metall. 99 (2), 103–110. Mucho, T.P., Barton, T.M., Compton, C.S., 1993. Room-and-Pillar Mining in Bump-Prone Conditions and Thin Pillar Mining as a Bump Mitigation Technique. United States Department of the Interior, Bureau of Mines. Ortlepp, W., 1968. A yielding rockbolt. Res. Organ. Bull. Chamber Mines South Africa 14, 6–8. Ortlepp, W., 1992. The design of support for the containment of rockburst damage in tunnels-an engineering approach. Rock Support in Mining and Underground Construction, Balkema, Rotterdam, pp. 593–609. Ortlepp, W., Stacey, T., 1994. Rockburst mechanisms in tunnels and shafts. Tunn. Undergr. Space Technol. 9 (1), 59–65. Osterwald, F.W., Dunrud, C.R., Collins, D.S., 1993. Coal mine bumps as related to geologic features in the northern part of the Sunnyside District, Carbon County, Utah. United States Geological Survey, Professional Paper, p. 1514. Pan, J., Ning, Y., Du, T., Zhang, Y., Liu, J., Xia, Y., Wei, X., 2012. The theory and system for preventing rock burst in large-scale areas. J. China Coal Soc. 37 (11), 1803–1809. Pan, Y., Wang, K., Xiao, Y., 2013. Design of anti-scour support based on theory of pendulum-type wave. Chinese J. Rock Mech. Eng. 8, 005. Pan, Y., Xiao, Y., Li, Z., Wang, K., 2014. Study of tunnel support theory of rockburst in coal mine and its application. J. China Coal Soc. 39 (2), 222–228.

the mechanism of a catastrophic bump in the Book Cliffs coal mining district, Utah, USA. Int. J. Rock Mech. Min. Sci. 34 (1), 27–43. Bräuner, G., 1994. Rockbursts in Coal Mines and Their Control. DMT, Essen, Germany, Balkema. Bukowska, M., 2006. The probability of rockburst occurrence in the Upper Silesian Coal Basin area dependent on natural mining conditions. J. Min. Sci. 42 (6), 570–577. Cai, M., 2013. Principles of rock support in burst-prone ground. Tunn. Undergr. Space Technol. 36, 46–56. Cai, W., Dou, L., Si, G., Cao, A., He, J., Liu, S., 2016. A principal component analysis/ fuzzy comprehensive evaluation model for coal burst liability assessment. Int. J. Rock Mech. Min. Sci. 81, 62–69. Calleja, J., Porter, I., 2016. Coalburst control methods. Coal Operators' Conference. University of Wollongong, Australia. Campoli, A.A., Barton, T.M., Van Dyke, F.C., Gauna, M., 1990. Mitigating Destructive Longwall Bumps Through Conventional Gate Entry Design. United States Department of the Interior, Bureau of Mines, United States. Campoli, A.A., Oyler, D.C., Chase, F.E., 1989. Performance of a Novel Bump-Control Pillar-Extracting Technique During Room-and-Pillar Retreat Coal Mining. Bureau of Mines, Pittsburgh, PA (USA). Chen, F., Pan, Y., Li, Z., Wang, A., Xu, M., 2013. Analysis and evaluation of effects of borehole pressure relief measures by drilling cutting method. Chinese J. Geotech. Eng. 35 (S2), 266–270. DeMarco, M., 1994. Yielding-pillar gateroad design considerations for longwall mining. Preprints-Society of Mining Engineers of AIME. DeMarco, M.J., Koehler, J., Maleki, H., 1995. Gate road design considerations for mitigation of coal bumps in western US longwall operations. In: Proceedings Mechanics and Mitigation of Violent Failure in Coal and Hard-Rock Mines. Bureau of Mines Special Publication, pp. 01–95. Dou, L., 2017. Control measurements for coal burst. Personal communication. Dou, L., He, X., 2001. Prevention Theory and Technology of Rockburst. China University of Mining and Technology Press, Xuzhou. Dou, L., He, Y., Zhang, W., 2003. Hazards of rock burst in island coal face and its control. Chinese J. Rock Mech. Eng. 11, 1866–1869. Dou, L., Lu, C., Mou, Z., Gao, M., 2009. Prevention and forecasting of rock burst hazards in coal mines. Int. J. Min. Sci. Technol. 19 (5), 585–591. Dou, L., Mou, Z., Li, Z., Cao, A., Gong, S., 2014. Research progress of monitoring, forecasting, and prevention of rockburst in underground coal mining in China. Int. J. Coal Sci. Technol. 1 (3), 278–288. Du, T., Dou, L., Lan, H., 2012. Numerical study of the prevention mechanisms of directional hydraulic fracturing in coal burst. J. XI'AN Univ. Sci. Technol. 32 (4), 444–449. Dubinski, J., 1994. Geophysical assessment of the hydraulic injection process in coal seams under rockburst hazard. In: Proceedings Geomechanics 93. Proc. Conference, Ostrava, pp. 55–58. Galvin, J.M., 2016. Ground Engineering Principles and Practices for Underground Coal Mining. Springer. Gay, N., 1993. Mining in the vicinity of geological structures – an analysis of mining induced seismicity and associated rockbursts in two South African mines. Proceedings of the 3rd International Symposium on Rockbursts and Seismicity in Mines, Canada, pp. 57–62. He, H., Dou, L., Fan, J., Du, T., Sun, X., 2012. Deep-hole directional fracturing of thick hard roof for rockburst prevention. Tunn. Undergr. Space Technol. 32, 34–43. Heasley, K.A., 1991. An examination of energy calculations applied to coal bump prediction. In: Proceedings The 32nd US Symposium on Rock Mechanics (USRMS), American Rock Mechanics Association. Hebblewhite, B., Galvin, J., 2017. A review of the geomechanics aspects of a double fatality coal burst at Austar Colliery in NSW, Australia in April 2014. Int. J. Min. Sci. Technol. 27 (1), 3–7. Hendon, G., 1998. Gateroad pillar extraction experience at Jim Walter Resources. In: Proceedings the 17th International Conference on Ground Control in Mining. Morgantown, West Virginia, pp. 1–10. Holland, C.T., 1958. Cause and occurrence of coal mine bumps. Trans. SME-AIME 211, 994–1004. Huang, B., Deng, G., Liu, C., 2007. Technology of coal-rock mass hydraulic fracture weakening and its development. Eng. Sci. 9 (4), 83–88. Iannacchione, A., Esterhuizen, G., Bajpayee, T., Swanson, P., Chapman, M., 2005. Characteristics of mining-induced seismicity associated with roof falls and roof caving events. In: Proceedings Alaska Rocks 2005, The 40th US Symposium on Rock Mechanics (USRMS), American Rock Mechanics Association. Iannacchione, A., Tadolini, S.C., 2008a. Coal mine burst prevention controls. In: Proceedings 27th International Conference on Ground Control in Mining, pp. 20–28. Iannacchione, A., Tadolini, S.C., 2016. Occurrence, prediction, and control of coal burst events in the US. Int. J. Min. Sci. Technol. 26 (1), 39–46. Iannacchione, A., Zelanko, J., 1995. Occurrence and remediation of coal mine bumps: a historical review. In: Proceedings Mechanics and Mitigation of Violent Failure in Coal and Hard-Rock Mines, U.S. Bureau of Mines, pp. 27–67. Iannacchione, A.T., Tadolini, S.C., 2008b. Coal mine burst prevention controls. In: Proceedings 27th International Conference on Ground Control in Mining, pp. 20–28. ICMM, 2015. Critical control management implementation guide. International Council on Mining & Metals. Jager, A., Wojno, L., Henderson, N., 1990. New developments in the design and support of tunnels under high stress. Proceedings of Technical Challenges in Deep-Level Mining, South Africa 1155–1172. Jiang, F., Qu, X., Yu, Z., 2011. Real time monitoring and measuring early warning technology and development of mine pressure bumping. Coal Sci. Technol. 39 (2), 59–64. Jiang, F., Yang, S., Cheng, Y., Zhang, X., Mao, Z., Xu, F., 2006. A study on microseismic

142

Tunnelling and Underground Space Technology 81 (2018) 129–143

C. Wei et al.

risk classification system. Int. J. Min. Sci. Technol (submitted for publication). Varley, F., Whyatt, J., 2008. Work practices to manage bump prone ground. In: Proceedings of the 27th International Conference on Ground Control in Mining, West Virginia University, Morgantown, pp. 20–28. Wang, A., Pan, Y., Li, Z., Zhao, B., 2016a. Analytical analysis of anti-impact mechanism of bolt supporting for bump-prone roadways. China Saf. Sci. J. 26 (8), 110–115. Wang, A., Pan, Y., Zhao, B., Li, Z., 2016b. Coupling vibration characteristics of rock mass and energy-absorption bolt and its anti-impact mechanism. J. China Coal Soc. 41 (11), 2734–2742. Wang, K., Pan, Y., 2015. An unified theory of energy absorption and anti-impact for surrounding rock and support in rock burst mine. Rock Soil Mech. 36 (9), 2585–2590. Wang, P., Zhao, J., Chugh, Y.P., Wang, Z., 2017. A novel longwall mining layout approach for extraction of deep coal deposits. Minerals 7 (4), 60. Wang, S., Hao, S., Chen, Y., Bai, J., Wang, X., Xu, Y., 2016c. Numerical investigation of coal pillar failure under simultaneous static and dynamic loading. Int. J. Rock Mech. Min. Sci. 84, 59–68. Whyatt, J., Blake, W., Williams, T., White, B., 2002. Years of rockbursting in the Coeurd’Alene District of Northern Idaho, USA: lessons learned and remaining issues. In: Proceedings Presentation at 109th Annual Exhibit and Meeting, Society for Mining, Metallurgy, and Exploration, Phoenix, pp. 25–27. Wu, Y., Wang, S., 1989. Research on mechanisms of water infusion for coal burst. J. China Coal Soc. 2, 69–80. Xia, D., Guo, H., Luo, Y., Ma, J., Chen, S., Wang, Z., 2015. Experimental study on reducing outburst proneness of coal seam via alkaline solution. J. China Coal Soc. 40 (8), 1768–1773. Yan, S., Ning, Y., 2000. The mechanism of hydrobreakage to control hard roof and its test study. J. China Coal Soc. 25 (1), 32–35. Zhang, C., Canbulat, I., Hebblewhite, B., Ward, C.R., 2017. Assessing coal burst phenomena in mining and insights into directions for future research. Int. J. Coal Geol. Zhang, M., Song, W., Pan, Y., 2003. Study on water pouring into coal seam to prevent rock-burst. China Saf. Sci. J. 13 (10), 69–72. Zhang, P., Swan, G., Nordlund, E., 2015. 1D numerical simulation of velocity amplification of P-waves travelling through fractured rock near a free surface. J. South Afr. Inst. Min. Metall. 115 (11), 1121–1126. Zhu, S., Jiang, F., Shi, X., Sun, G., Zhang, Z., Cheng, X., Zhang, H., 2015. Energy dissipation index method for determining rockburst prevention drilling parameters. Rock Soil Mech. 36 (8), 2270–2276.

Patyńska, R., Kabiesz, J., 2014. Crump risk in Upper Silesian Coal Basin mines in the years 1993–2012. Bezpieczeństwo Pracy i Ochrona Środowiska w Górnictwie 5, 3–10. Potvin, Y., Wesseloo, J., 2013. Towards an understanding of dynamic demand on ground support. J. South Afr. Inst. Min. Metall. 113 (12), 913–922. Ptacek, J., 2017. Rockburst in Ostrava-Karvina Coalfield. Proc. Eng. 191, 1144–1151. Qi, Q., Dou, L., 2008. Mechanisms and Control Techniques of Coal Burst. China University of Mining and Technology Press, Xuzhou. Qin, S., Dou, L., 2014. Simulation of hydraulic fracture best slotted position of hard roof. Saf. Coal Mines 45 (5), 151–154. Qu, X., Jiang, F., Yu, Z., Ju, H., 2011. Rockburst monitoring and precaution technology based on equivalent drilling research and its applications. Chinese J. Rock Mech. Eng. 11, 022. Rob, T., 2016. Summary comment on the coal burst management strategies adopted in European coal mines. Coal Burst Work Shop, University of New South Wales, Sydney. Roux, A., Leeman, E., Denkhaus, H., 1957. Destressing: a means of ameliorating rockburst conditions. Part I: The concept of destressing and the results obtained from its applications. J. Southern African Inst. Min. Metall. 57 (1), 66–68. Saharan, M.R., Mitri, H., 2011. Destress blasting as a mines safety tool: some fundamental challenges for successful applications. Proc. Eng. 26, 37–47. Salamon, M., 1983. Rockburst hazard and the fight for its alleviation in South African gold mines. In: Conference on Rockbursts, Prediction and Control. Institute of Mining and Metallurgy, London, pp. 11–52. Shen, C., Deng, G., 2013. Numerical simulation of using high pressure water injection to control rockburst. Saf. Coal Mines 44 (7), 7–9. Song, W., Li, D., Zhang, M., Pan, Y., Sun, C., 2006. Replacement theory of water infusion. Chinese J. Geol. Hazard Control 17 (2), 147–150. Stillborg, E.B., 1986. Professional Users Handbook for Rock Bolting. Trans Tech Publications Ltd, Zurich, Switzerland. Su, C., Zhai, X., Wei, X., 2014. Influence of saturation period on bursting liability indices for coal seam# 2 in Qianqiu Coal Mine. Chinese J. Rock Mech. Eng. 33 (2), 235–242. Trade & Investment, 2015. Safety management systems in mines. New South Wales Government, Australia. < https://www.resourcesandenergy.nsw.gov.au/__data/ assets/pdf_file/0008/543941/NSW-code-of-practice-Safety-managment-systems-inmines.pdf > . Turner, M., Player, J., 2000. Seismicity at Big Bell mine. Proceedings of MassMin. Australasian Institute of Mining and Metallurgy, Brisbane, Australia, pp. 791–797. Vardar, O., Zhang, C., Canbulat, I., Hebblewhite, B., 2018. A semi-quantitative coal burst

143