Journal of Rock Mechanics and Geotechnical Engineering 13 (2021) 457e478

Contents lists available at ScienceDirect

Journal of Rock Mechanics and

Geotechnical Engineering
journal homepage: www.jrmge.cn

Review

Numerical modeling for rockbursts: A state-of-the-art review

Jun Wang a, Derek B. Apel a, *, Yuanyuan Pu b, Robert Hall a, Chong Wei a,
Mohammadali Sepehri a
a
School of Mining and Petroleum Engineering, University of Alberta, Edmonton, Alberta, T6G 2R3, Canada
b
State Key Laboratory of Coal Mine Disaster Dynamics and Control, Chongqing University, Chongqing, 400044, China

a r t i c l e i n f o a b s t r a c t

Article history: As the depth of excavation increases, rockburst becomes one of the most serious geological hazards
Received 2 March 2020 damaging equipment and facilities and even causing fatalities in mining and civil engineering. This has
Received in revised form forced researchers worldwide to identify different methods to investigate rockburst-related problems.
22 June 2020
However, some problems, such as the mechanisms and the prediction of rockbursts, continue to be
Accepted 9 September 2020
studied because rockburst is a very complicated phenomenon influenced by the uncertainty and
Available online 24 December 2020
complexity in geological conditions, in situ stresses, induced stresses, etc. Numerical modeling is a
widely used method for investigating rockbursts. To date, great achievements have been made owing to
Keywords:
Rockburst
the rapid development of information technology (IT) and computer equipment. Hence, it is necessary
Numerical modeling and meaningful to conduct a review of the current state of the studies for rockburst numerical modeling.
Rockburst mechanism In this paper, the categories and the origin of different numerical approaches employed in modeling
Dynamic modeling rockbursts are reviewed and the current usage of various numerical modeling approaches is investigated
by a literature research. Later, a state-of-the-art review is implemented to investigate the application of
numerical modeling in the mechanism study, and prediction and prevention of rockbursts. The main
achievements and problems are highlighted. Finally, this paper discusses the limitations and the future
research of numerical modeling for rockbursts. An approach is proposed to provide researchers with a
systematic and reasonable numerical modeling framework.
Ó 2021 Institute of Rock and Soil Mechanics, Chinese Academy of Sciences. Production and hosting by
Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/
licenses/by-nc-nd/4.0/).

1. Introduction
With the increasing demand for mineral resources and the
continuous consumption of shallow mineral resources, the mining
depth in the world is going deeper (Feng et al., 2019a). Some
geotechnical engineering projects (e.g. tunnels and underground
laboratories) are constructed in very deep areas with complicated
ground conditions (Jiang et al., 2010; Zhang et al., 2012a). The ul-
timately high in situ stress and complex geological environment
cause a series of geological hazards such as rockburst, rock mass
caving, large deformation of excavations, and high ground tem-
perature (Palei and Das, 2009; Fall et al., 2010; Xie et al., 2015; Yu
et al., 2015; Ranjith et al., 2017; Shi et al., 2020). Rockburst is a
dynamic rock failure phenomenon usually marked by sudden
ejection of rock blocks from an underground opening. The ejection
* Corresponding author.
E-mail address: dapel@ualberta.ca (D.B. Apel).
Peer review under responsibility of Institute of Rock and Soil Mechanics, Chi-
nese Academy of Sciences.
of the rock material is violent and associated with a rapid release of
energy. Due to its unpredictability and high intensity, rockburst is
one of the most hazardous geological disasters. It damages equip-
ment, facilities and even causes fatalities, as shown in Fig. 1.
Generally, rockbursts can be classified into three types: strain-
burst, pillar burst, and fault-slip burst according to the source
mechanism (Hedley, 1992; Kaiser and Cai, 2012; Cai, 2013).
Strainburst is the most common type of rockburst in all under-
ground excavations (Zhang et al., 2012a; Cai, 2013). It occurs due to
the concentration of excavation-induced tangential stress and the
existence of a relatively “soft” loading environment in the rock
mass surrounding the fractured rock (Kaiser and Cai, 2012). The
rock can fail locally at excavation boundaries in an unstable and
violent manner. Pillar burst is a violent failure in the pillar core or
the complete collapse of a pillar. This type of failure will occur
suddenly when the accumulated elastic strain energy reaches a
critical level, which means that the released energy is higher than
the dissipated energy. A large volume of failed rocks are released
and the magnitude is usually larger than that of a strainburst
(Ortlepp and Stacey, 1994). Fault-slip burst is due to the slip along

https://doi.org/10.1016/j.jrmge.2020.09.011
1674-7755 Ó 2021 Institute of Rock and Soil Mechanics, Chinese Academy of Sciences. Production and hosting by Elsevier B.V. This is an open access article under the CC BY-
NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
458 J. Wang et al. / Journal of Rock Mechanics and Geotechnical Engineering 13 (2021) 457e478
Fig. 1. Examples of rockbursts at a diamond mine in Canada.
pre-existing faults or along newly generated shear ruptures. When
the shear stress along a fault or a shear rupture exceeds the shear
strength, the fault or shear rupture will slip. As a result, a large
amount of seismic energy will be released suddenly, with high
ground vibrations or ground motions, and may trigger other types
of rockbursts. Rockburst events have been reported in all mining
countries (e.g. South Africa, Canada, Australia, USA, China, India,
Poland, and Chile) since the first such events were recorded in
South Africa and India at the turn of the 20th century (Hedley, 1987;
Blake and Hedley, 2003). Fig. 2 shows a historical rockburst map of
more than 1100 events from near 50 areas in the world during the
period of 1995e2019. On March 9, 2005, a mining-related seismic
event with a Richter magnitude of 5.3 occurred in the Klerksdorp
District in South Africa, killing two miners and injuring 58 people
(Durrheim, 2010). On November 28, 2009, a serious rockburst with
a Richter magnitude of 2 occurred in a drainage tunnel in the
Jinping II Hydropower Station in China, resulting in seven deaths
and the destruction of a tunnel boring machine (TBM) (Zhang et al.,
2012a). On May 31, 2015, a rockburst event at the Neelum-Jhelum
Hydropower Project in Pakistan killed three workers and
destroyed a TBM (Naji et al., 2018). The large quantity of rockburst
events and the severity of those incidents are incontrovertible ev-
idence that rockburst is a universal and serious problem that re-
quires immediate attention.
At present, great efforts have been devoted to the investigation
of rockbursts. Most research is focused on the mechanisms,
contributing factors, risk evaluation and prediction, and prevention
and mitigation of rockbursts. The methodologies of studying
rockbursts can be generally summarized into five types: empirical,
analytical, experimental, data-based, and numerical, as shown in
Fig. 3. Manouchehrian and Cai (2018) stated that there is not yet an
effective way to control rockbursts because the phenomenon is so
complex and is influenced by many factors (e.g. complexity and
uncertainty in geological conditions, in situ stresses, induced
stresses and triggering conditions (Gao et al., 2019a)). Empirical
methods are generally site-specific and therefore difficult to
generalize. Both analytical and experimental methods employ
many idealized assumptions which are incapable of matching the
real field circumstances. For instance, experimental testing condi-
tions are usually designed to represent field conditions, but the
rock behavior in experimental conditions may be very different
from that in field conditions owing to the different properties of
intact rocks and rock masses (Manouchehrian, 2016). Data-based
methods are reasonable and accurate only if a sized and acces-
sible database is provided (Zhou et al., 2018a; Pu et al., 2020).
Over the last five decades, with rapid developments in computer
software and hardware, significant progress has been made in using
numerical modeling to simulate physical phenomena in rock me-
chanics and rock engineering at various scales. Compared with other
methods such as physical simulation and field tests, the numerical
modeling method has the advantages of low cost, safety, timesaving,
and flexibility. Also, it can provide more information. Salamon (1993)
stated that different aspects of the rockburst problem (e.g. the rela-
tionship between mining activities and the related seismicity, source
mechanism, and the effects of seismic waves on mining excavations)
would benefit from numerical modeling. To date, numerical simula-
tion methods have been widely used to evaluate the mechanical
response of complex rock masses and to study rock mass deformation
and failure mechanisms. Since the 1970s, these methods have also
provided a common and even necessary way to study rockbursts.
Therefore, it is necessary and meaningful to conduct a review of the
current state of rockburst numerical modeling and to analyze the
limitations and prospects in this research field.
In this study, the categories and the origin of different numerical
approaches employed in modeling rockbursts are reviewed and the
current usage of various numerical approaches is investigated by a
literature research. Also, a state-of-the-art review is implemented
to investigate the application of numerical modeling in the mech-
anism study, and prediction and prevention of rockbursts. The main
achievements and problems are highlighted. Finally, this paper
discusses the limitations and future research directions of this
topic. An approach of numerical modeling for rockbursts is also
proposed in order to provide researchers with a systematic and
reasonable numerical modeling framework.
 J. Wang et al. / Journal of Rock Mechanics and Geotechnical Engineering 13 (2021) 457e478
Fig. 2. A historical rockburst map for the period of 1995e2019 (updated after Bennett and Marshall, 2001).
Fig. 3. A summary of methodologies for studying rockbursts (after Manouchehrian,
2016).
2. Numerical approaches for modeling rockbursts
2.1. Category of numerical approaches
With the rapid development of information technology (IT) and
computer equipment, the availability of high-powered computer
equipment such as supercomputers, cloud computing, and a large
number of numerical approaches as well as a great variety of
commercial or academic codes has opened avenues for the analysis
and evaluation of complex problems in rock mechanics and rock
engineering. Nevertheless, it is critical for researchers and engi-
neers to discriminate amongst different numerical approaches and
codes before using numerical modeling to tackle rock mechanics
problems (Wagner, 2019). According to Jing and Hudson (2002) and
Jing (2003), numerical approaches in rock mechanics can be
459
classified into continuum, discontinuum and hybrid methods
(Table 1). The main subtypes and related commercial or research
codes at present are also listed in Table 1. This review is particularly
focused on numerical modeling for rockbursts. The detailed dis-
cussion of the concepts, principles, advances, and development of
various numerical approaches used in rock mechanics and rock
engineering has been excellently reviewed by Jager and Ryder
(1999), Jing and Hudson (2002), Jing (2003), Brady and Brown
(2004), Nikoli
c et al. (2016) among others.
2.2. Origin of numerical approaches employed in modeling
rockbursts
Employing numerical approaches to investigate rock mechanics
problems has been a common practice dating back to the 1960s
(Salamon, 1964; Jing and Hudson, 2002; Jing, 2003), while nu-
merical modeling to study rockburst-related problems was intro-
duced a little later, in the 1970s.
The hazard of spalling is an important consideration when rock
faces are subjected to dynamic loading induced by blasting or
rockbursts near a permanent mine opening. Miranda (1972) first
developed a finite difference computer code to simulate the elastic
pulse propagation problem in the Split-Hopkinson pressure bar
(SHPB) technique. He found that the computed numerical gage data
agreed well with the derived strain and strain rateetime curves of
the specimen. In the same year, Blake (1972a) acted as a pioneer in
using the finite element method (FEM) model to study pillar bursts.
He found that the areas of high stress concentration could be used
as a sign to predict rockburst locations. Brady (1979) developed the
boundary element method (BEM) with a proposed complete plane
strain concept and first used it to study the unstable failure of
pillars (pillar crush). He also put forward a cutting-edge idea:
modeling surrounding rock as an elastic continuum by exploiting
the efficient BEM and treating pillars as inclusions within which
more complex constitutive equations were proposed, for example,
the finite difference method (FDM) might be used in pillars to
model the complex behavior of rock materials. Maybe this is the
origin of the hybrid method concept. Board and Voegele (1981)
employed a hybrid and a digital computer model utilizing the
displacement discontinuity method (BEM) to study stress and
displacement changes. Then they used the calculated energy
460 J. Wang et al. / Journal of Rock Mechanics and Geotechnical Engineering 13 (2021) 457e478

Table 1
Overview of numerical approaches and codes in rock mechanics.

Numerical approach Subtype Commercial/academic code Institution/author (year)

Continuum method Finite element method (FEM) ABAQUS Dassault Systèmes

ADINA ADINA R&D, Inc.
ANSYS ANSYS, Inc.
GEO5 Fine Software
LS-DYNA LSTC
Midas GTS NX MIDAS IT
PLAXIS2D, PLAXIS3D Plaxis
RFPA2D, RFPA3D Mechsoft
RS2 (Phase2), RS3 Rocscience
Finite difference method (FDM) FLAC, FLAC3D Itasca Consulting Group, Inc.
Boundary element method (BEM) Examine Rocscience
Map3D Non-Linear Map3D
Discontinuum method Discrete element method (DEM) PFC2D, PFC3D Itasca Consulting Group, Inc.
UDEC, 3DEC Itasca Consulting Group, Inc.
Discontinuous deformation analysis (DDA) DDA codes Goodman and Shi (1985)
Discrete fracture network (DFN) FracMan Golder
NAPSAC AEA Technology
Hybrid method BEM/DEM DEM_SRS þ BEDA þ FNET þ BEFA Wei (1992), Wei and Hudson (1998)
BEM/FEM BEM/FEM codes Zienkiewicz et al. (1977)
DEM/FEM CA3 Fakhimi (2009)
ELFEN Rockfield
IRAZU Geomechanica
NMM Shi (1991)
Y2D Munjiza et al. (2004)
Y-Geo Mahabadi et al. (2012)
DEM/FDM PFC2D/FLAC, PFC3D/FLAC3D Itasca Consulting Group, Inc.

release rate (ERR) to examine the effect of mining in an underhand
fashion on the potential for sill pillar bursting. The results sug-
gested that the ERR was lower when a larger sill pillar remained,
but its magnitude was still enough to cause a pillar burst. This
shows that numerical simulation is a useful and potential tool to
evaluate rockburst risks and to improve mining design. Hart (1980)
used the explicit FDM code, STEALTH, and the BEM code, MINAP, to
evaluate rock failure and dynamic instabilities (rockbursts) based
on the ERR. He concluded that for the elastic simulation, the results
from the two codes were close and the BEM model was much easier
to formulate and less expensive to run. However, the results also
suggested that stored energy simulated by MINAP using the elastic
constitutive model was smaller than that using STEALTH employing
a nonlinear constitutive model, due to the lack of plastic defor-
mation stage. This shows how important it is to select a reasonable
constitutive model for geotechnical modeling. Zubelewicz and
Mroz (1983) combined the FEM and a dynamic approach to study
rockburst processes in a rock pillar, excavation faces, and tunnels.
They also confirmed that the dynamic mode of failure can be
numerically studied for any specified configuration and initial
conditions.
The usage of the discontinuum method for modeling rockbursts
is later than that of the continuum method, since more latter re-
searchers began to notice the significance of discontinuities such as
joints, which control and influence the mechanical behaviors of
rocks, as well as the requirement of the large displacement of rocks,
while the continuum method is based on the hypothesis of
continuous small deformation. The discrete element method (DEM)
is a discontinuum modeling approach that can, in a straightforward
manner, model large displacements, disintegrations, rotations, and
general nonlinear constitutive behaviors for both rock masses and
joints. After the DEM was developed and established (Cundall, 1971,
1988; Cundall and Strack, 1979), Lemos et al. (1987), as the first
researchers, used it to investigate the features of jointed rocks
subjected to dynamic loading such as rockbursts. They stated that
the DEM provided a useful tool for understanding a wide variety of
dynamic problems in jointed rock masses. The discrete fracture
network (DFN) is a special discrete method that is the most useful
for the study of flow in fractured rock masses (Jing and Hudson,
2002; Jing, 2003). Arndt et al. (2009) used the DFN model to
conduct dynamic tests of a very heavy support system in order to
investigate seismic hazards to ground support. Sun et al. (2007)
investigated rockbursts in a circular tunnel under unloading con-
ditions using realistic failure process analysis (RFPA) program and
discontinuous deformation analysis (DDA). In order to simulate the
gestation process of rockbursts, the failure patterns of surrounding
rocks of a circular tunnel were first studied by RFPA, taking into
consideration the non-homogeneity of rock materials and different
in situ stresses. The crack lines and some potential cracks in the
RFPA model were then imported into the DDA model to explicitly
simulate the dynamic processes of a rockburst. The results sug-
gested that the loss of stability of the surface rock mass was an
omen or the beginning of a rockburst. However, Sun et al. (2007)
also pointed out that the rock masses had to be discretized artifi-
cially in the DDA model and that the released strain energy was
ignored in the DDA model. Hence, an improved simulation of
rockbursts needs to be further investigated.
Jing (2003) reported that the main influence area of damage is
concentrated near the excavation face and that linear material
behavior is exhibited in the far-field region without fractures.
Therefore, the discrete method is more suitable for near-field
analysis, while the continuous method is more suitable for far-
field analysis. Thus, it is a good way to combine those two nu-
merical methods, which simultaneously makes full use of their
respective advantages and avoids their disadvantages. Daehnke
(1999) analyzed the dynamic fracturing due to the interaction of
primary (P) and secondary (S) waves with stopes by implementing
a parametric study through the FEM/DEM code ELFEN. The results
verify that ELFEN can accurately simulate stress wave interactions
and dynamic fracturing in underground excavations. After the
FDM/DEM coupled method was proposed and developed (Itasca
Consulting Group, Inc., 2000), Hazzard and Young (2002) adopted
it to model microseismic (MS) activities surrounding a deep tunnel.
The target area was modeled by an assemblage of particles from
particle flow code (PFC) and was then coupled with the fast
Lagrangian analysis of continua (FLAC) program. Although the
 J. Wang et al. / Journal of Rock Mechanics and Geotechnical Engineering 13 (2021) 457e478
model is clearly a large simplification of realities, the simulation
results, such as seismic locations, magnitudes, and mechanisms
obtained from the numerical model compared with seismic data in
the field, give further confidence that the FDM/DEM coupled model
can behave in a realistic fashion. Later, Cai et al. (2007) employed
the FLAC/PFC coupled numerical model to investigate excavation-
induced acoustic emission (AE) activities in large-scale under-
ground excavations. In the modeling, PFC was used to simulate AE
sensors around surrounding rocks and the remaining rock masses
were modeled by FLAC. The simulated AE activities were in good
agreement with field monitoring results, which confirms that the
coupled numerical method as an advanced tool can be applied to
the interpretation of monitoring data and stability evaluation of
large-scale underground excavations. SPECFEM2D is a powerful
software package that can simulate the propagation of acoustic and
elastic waves in a variety of media such as fluids, elastomers, vis-
coelastics, and anisotropic and porous objects, but it does not have
the ability to analyze excavation-induced stress. By combining
FLAC/SPECFEM2D, Wang and Cai (2017) proposed a coupled nu-
merical method for a nonlinear velocity model to study the exca-
vation effect on the ground motion in an excavation boundary. They
found that the amplification effect at the excavation surface agreed
well with underground field observations and when the simulated
excavation rock was of fair quality, the excavation boundary of the
stope had a stronger ground motion and wider seismic response.
The coupled numerical method is helpful to better estimate ground
motion parameters in dynamic load support design and can provide
reasonable ground motion evaluation parameters in an inversion
analysis of rockburst damage. Fig. 4 shows the development pro-
cess of various numerical methods employed in the history of
modeling rockbursts.
2.3. Usage investigation of numerical modeling for rockbursts
In order to understand the usage of various numerical ap-
proaches to model rockbursts, we used “TS¼(rockburst* OR rock
burst*) AND TS ¼ numerical” as query words in the Web of Science
database to carry out an investigation about literature published
during the last 20 years. It is very hard to search all published
literature about this topic due to copyright and access-related
Fig. 4. History of numerical methods employed in modeling rockbursts.
461
issues. However, after reviewing the published literature one by
one in this well-accepted and authoritative database and having
relatively large amounts of data, we were confident that our
investigation basically reflected the usage situations of various
numerical approaches to model rockbursts. The search results are
shown in Figs. 5e7, and the variation laws are summarized as
follows:
(1) Fig. 5 shows that since 1999, the amount of literature in this
field has increased. From 1999 to 2008, the growth was not
very apparent, and the number of published articles
increased only from three to seven. After 2009, the number
of published articles grew rapidly although they may suffer
decreases in a few years. After 2016, the number of publi-
cations was nearly 10 times that of 1999. Moreover, most
researchers choose the continuum method for addressing
rockburst-related problems. After 2014, more and more re-
searchers begin to use the discontinuum and hybrid
methods.
(2) As shown in Fig. 6, the most common numerical approach is
the continuum method (77.36%) and the second most com-
mon is the discontinuum method (19.54%). The hybrid
method accounts for only 3.1%. For the continuum method,
FEM and FDM are the most popular numerical methods
(94.1% in total). Furthermore, in the FEM, the widely used
numerical programs are ABAQUS (13.27%), ANSYS (11.22%),
and RFPA2D (22.45%). Suit3D (14.29%) and Map3D (35.71%)
are the most popular codes in the BEM, while FLAC (20.72%)
and FLAC3D (76.89%) account for the largest proportion in
the FDM. For the discontinuum method, DEM is the most
popular (93.33% in total). The reason that there is no DFN in
the pie chart is that DFN is the built-in function in some
programs such as ABAQUS, FLAC3D, and 3DEC. Thus we did
not show it in a single segment. In the DEM, the most popular
codes are UDEC (54.46%), 3DEC (9.86%), and PFC2D (17.86%),
respectively. For the hybrid method, the DEM/FEM (47.37%) is
widely used, in which FLFEN (55.56%) is the most common
numerical program.
(3) In reality, rockburst types are various according to different
researchers (Zhou et al., 2018a). Therefore, we classified the
462 J. Wang et al. / Journal of Rock Mechanics and Geotechnical Engineering 13 (2021) 457e478
Fig. 5. Literature about numerical modeling of rockbursts in the last 20 years (the search results in 2019 are incomplete).
published literature as static and dynamic simulations,
respectively, based on loading conditions and seismic waves.
As shown in Fig. 7a, from 1999 to 2008, most numerical
modeling of rockbursts was conducted by static simulation
and then there was rapid growth after 2009, especially after
2015. In 2016e2018, the proportion of dynamic simulation
was more than 30% in each year. Overall, from Fig. 7b we can
see that most researchers employ static simulation to
investigate rockbursts (71.34%).
3. A state-of-the-art review of the application of numerical
modeling for rockbursts
3.1. Numerical modeling for rockburst mechanism
The initial studies of rockburst mechanisms through numerical
modeling were based on some specific engineering cases, in which
the continuum method, elastic constitutive relationship, and two-
dimensional (2D) model were very often used owing to the lower
computation cost and modeling ability at that time, although many
researchers noticed that using elastoplastic constitutive relation-
ships and three-dimensional (3D) models was a better choice. In
this stage, researchers tried to reveal the rockburst mechanism by
modeling the changes of maximum principal stress, strain energy,
and deformation after mining or excavating activities (Blake, 1972a;
Miranda, 1972; Brady, 1979; Board and Voegele, 1981; Napier and
Stephansen, 1987; Bardet, 1987, 1990).
After that, with the development of computing capacities and
progress in numerical programs for modeling complex geometry
and plastic nonlinearities, more researchers tended to utilize elas-
toplastic constitutive relationships and/or 3D models (Blake, 1972a;
Miranda, 1972), and representative studies are listed in Table 2.
Rockburst is a type of rock failure phenomena caused by the initi-
ation, growth, and expansion of micro-fractures, which then form a
macroscopic fracture (Wang et al., 2006). After confirming that AE
parameters are related to the damage variable of rocks, Wang et al.
(2003) used RFPA2D to simulate pillar bursts and found that the
simulation results can reflect the macroscopic failure evolution
process induced by microscopic fractures, and the spatio-temporal
distribution characteristics of AE events. Using RFPA2D, Wang et al.
(2003) also studied the effects of rock heterogeneity on rockburst
potentials. Then, Wang et al. (2006) employed RFPA2D to study the
progressive failure process and associated MS behavior of rock
pillars. The simulated results verified that the stiffness of the roof
and floor play an important role in controlling the unstable failure
or collapse of rock pillars. Zhu et al. (2010) used RFPA-Dynamics to
simulate the rockbursts triggered by the dynamic disturbance
around a deep underground opening and confirmed that the dy-
namic disturbance is one of the most important contributing fac-
tors inducing rockbursts.
Most of the aforementioned work is limited to the study of rock
failures based on the small deformation rule. In fact, the gestation
and development of rockburst is a process from static failure to
dynamic failure, which is transformed from a continuous small
deformation into a discontinuous large deformation in a very short
time. It is difficult to simulate the discontinuous deformation
behavior of rock masses based on continuum methods such as FEM,
BEM, and FDM. To this end, scholars tended to use DEM and other
hybrid numerical methods suitable for simulating discontinuous
deformation to reproduce rockburst phenomena. After years of
development, the DEM has made great progress in both theories
and applied research: (1) from rigid elements to deformable ele-
ments; (2) from 2D to 3D modeling; (3) from the simulation of
static problems to the simulation of dynamic problems; and (4)
from the single mechanical simulation to the simulation of multi-
phase media and multi-field coupling problems. The general
block DEM considers the rock mass to be composed of discrete rock
masses and joint faces between rock blocks. The rock mass can
move, rotate, and deform, and the joint faces can be compressed,
slid, and separated to more realistically simulate the continuous
and discontinuous deformations of jointed rock masses (Jiang,
2017). Therefore, the DEM and DEM-related hybrid methods are
undoubtedly a good choice for the numerical modeling of the
continuous-discontinuous deformation behavior of the rockburst
gestation and failure processes. The corresponding studies are also
summarized in Table 2. Recently, Vazaios (2018) and Vazaios et al.
(2019) used FDEM (FEM/DEM) models through IRAZU to investi-
gate the effect of pre-existing joints on strainburst phenomena in a
 J. Wang et al. / Journal of Rock Mechanics and Geotechnical Engineering 13 (2021) 457e478 463

Fig. 6. Distribution of different numerical methods in modeling rockbursts in the last 20 years (the search results in 2019 are incomplete).

deep hard rock excavation. To conduct a parametric study, three
numerical model configurations were created. The first configura-
tion did not include any structures (massive rock masses), and the
two other configurations included stochastic joints by integrating a
different number of DFNs. In those models, the initiation, propa-
gation, and coalescence of fractures and abrupt rock ejection
occurring in the excavation under high-magnitude stresses were
able to be explicitly simulated (part of results are shown in Fig. 8).
Fig. 8 shows that rock blocks located around the excavation
boundary possess higher velocities than those far away from the
excavation wall, which is the reality (Ortlepp and Stacey, 1994; Qiu
et al., 2014). Additionally, larger ejected volumes are generated
with the increasing number of pre-existing joints which also
govern the shape and size of ejected rock blocks. The FDEM models
have made it possible to simulate the change from a continuous
deformation to a discontinuous deformation to reproduce the
physical process of rockburst phenomena, which further high-
lighted the application of the hybrid method in rockburst numer-
ical modeling.
In summary, the employment of various numerical methods,
codes, and constitutive models at different dimensions and scales
has greatly enriched people’s understanding of the complex source
mechanism and damage mechanism of rockbursts. Compared with
continuum methods, the DEMs, especially DEM-related hybrid
methods, have achieved the explicit simulation of rockburst pro-
cess, providing an effective tool for researchers to reveal the nature
of rockbursts. This might motivate more researchers to use dis-
continuum and hybrid methods in this research field. However, it is
not true because these methods are not flawless. To eliminate the
effects of mesh size on the fracture position within rocks, a very fine
464 J. Wang et al. / Journal of Rock Mechanics and Geotechnical Engineering 13 (2021) 457e478
Fig. 7. Literature about static and dynamic simulations of rockbursts in the last 20 years (the search results in 2019 are incomplete).
mesh resolution is usually needed, which sacrifices the computa-
tional efficiency of the model. Moreover, the calculation time will
be greatly extended when the blocks or elements undergo a large
deformation and are detached from each other (Gao, 2019). For
instance, Vazaios et al. (2019) reported that the total run time of the
third model configuration is around 100e144 h. This is an impor-
tant reason why discontinuum and hybrid methods are less used
than continuum methods (see Fig. 6). It should be noted that we
cannot easily say which numerical method is better to simulate
rockburst mechanisms since every method has its own advantages
and limitations (Table 3). The selection of numerical methods
should be based on the match between the capabilities of numer-
ical codes and rockburst mechanisms, and the specific engineering
situation. To provide some references for researchers and engi-
neers, a simple guideline for choosing numerical methods is sum-
marized based on this review and authors’ experience over the
years (Fig. 9).
3.2. Numerical modeling for the evaluation and prediction of
rockbursts
Researchers agree that it is very difficult to accurately predict
the occurrence time of rockbursts due to the randomness and
complexity of the mechanism of rockbursts (Qian, 2014; Zhou et al.,
2018a). However, the occurrence of rockbursts is mainly deter-
mined by the change in ground stress caused by the excavation of
deep rock masses. The geological survey technology, ground stress
detection technology, rock mechanics theories and methods, and
long-term development of numerical simulation have made it
possible to perform qualitative and quantitative prediction of the
location and intensity of rockbursts (Qian, 2014). Scholars believe
that the era of quantitative prediction of rockbursts has arrived, and
such quantitative progress needs to be achieved through a combi-
nation of numerical simulation and on-site observation (Stacey,
2013; Cai, 2016).
In the past few decades, scholars have proposed a number of
prediction and evaluation indices or indicators of rockburst po-
tential based on a variety of rockburst theories, and rockburst
phenomena from the aspects of strength, stiffness, energy, stability,
fracture, damage, etc (Li et al., 2013). To learn about the complete
and detailed prediction and evaluation indices or indicators of
rockbursts, readers are recommended to references (Zhang and Fu,
2008; He et al., 2012; Li et al., 2013; Qian, 2014; Mazaira and
Konicek, 2015; Feng et al., 2017b; Leveille et al., 2017; Zhou et al.,
2018a; Afraei et al., 2019; Jiang et al., 2019; Pu et al., 2019). At
present, the proposed prediction and evaluation indicators of
rockbursts can be generally classified into two categories: (1) in-
dicators based on the stress/strength criterion, such as tangential
stress, axial stress, uniaxial compressive strength (UCS), and major
principal stress, in which the representative indices are sq/sc
(Russenes, 1974), sc/s1 (Barton et al., 1974), sL/sc (Turchaninov et al.,
1972), sc/sq (Hoek and Brown, 1980), sc/s1 (Tao, 1988), excess shear
stress (ESS) (Ryder, 1988), sc/st (Peng et al., 1996) (s1es3)/sc (Castro
et al., 2012), etc.; and (2) indicators based on the energy criterion,
such as elastic strain energy and elastoplastic deformation energy,
in which the representative indices are strain energy storage co-
efficient (Wsp/Wst) (Kidybin  ski, 1981), the ratio of kinetic energy to
released energy (Wk/Wr) (Hedley, 1992), ERR (Salamon, 1984;
Kaiser et al., 1996), burst potential index (BPI) (Mitri et al., 1999),
local energy release rate (LERR) (Wiles, 2002; Jiang et al., 2010),
loading system stiffness (LSS) (Wiles, 2002), strain energy density
(SED) (Wattimena et al., 2012), etc. Due to the complexity of the
geological and construction conditions, and mining- or excavation-
induced effects, it is very difficult to predict and estimate rock-
bursts based on analytical and experimental methods, because
many idealized assumptions are employed in these methods. By
contrast, numerical modeling can simulate the elastoplastic,
nonlinear, and post-yield behavior of rock masses and the effects of
in situ stresses and geological features on mining or excavating
works, which enables researchers to understand the “real world” in
underground engineering. Thus, using numerical modeling with
the prediction and evaluation indices is a useful tool for researchers
and engineers to estimate rockbursts.
The studies about numerical modeling and prediction of rock-
bursts are summarized in Table 4. As shown in Table 4, different
indicators (stress/strength-based, or energy-based) have been
widely employed to assess and predict rockburst potentials in
various engineering projects. In addition to basic variables such as
Table 2
Summary of the applications of numerical modeling for rockburst mechanisms.

Numerical Rockburst Study type Numerical Geometry Constitutive Description Reference

approach type code model

Continuum Pillar burst Case study FLAC 2D Strain- A cohesion-softening model was utilized to simulate laboratory specimens and then was used to evaluate the Whyatt and Board (1991)
method softening shear-fracture rockburst potential for underhand longwall and overhand cut-and-fill stopes in deep mining
Fault-slip Parametric FLAC3D 3D Barton’s shear The Barton’s shear strength model was implemented in FLAC3D using Cþþ programming language while at the Sainoki (2014), Sainoki and
burst study strength, Saeb same time considering the change from static to dynamic friction law. The results suggested that the roughness Mitri (2014a, b, 2015)
and Amadei’s of the fault surface should be considered when estimating the magnitude or the energy of the fault-slip. The
model Saeb and Amadei (1992)’s model was also employed to simulate the effect of the fault surface asperities on the
intensity of seismic waves arising from the fault-slip
Strainburst Parametric ABAQUS 2D Strain- Considering the effect of material heterogeneity on rock failure processes, a heterogeneous model was Manouchehrian (2016),
and case softening developed in ABAQUS using Python scripts to simulate rockbursts near fault zones in deep tunnels. The Manouchehrian and Cai
study simulated results corresponded well with the post-event observation in the 1128 rockburst event in the (2018)

J. Wang et al. / Journal of Rock Mechanics and Geotechnical Engineering 13 (2021) 457e478
Jinping II Hydropower Station, China
Fault-slip Case study FLAC3D 3D Mohr- Recognizing that the mechanism of intense rockburst induced by nearby small-scale shear zones in deep civil Naji et al. (2018)
burst Coulomb tunnels was unclear, the FLAC3D explicit code was used to simulate the shear zone in a twin-tunnel system
Strainburst Parametric FLAC3D 3D Mohr- Numerical simulation was adopted to study the failure laws of the surrounding rocks in the process of Guo et al. (2019)
study Coulomb rockbursts in roadways with different ratios between principal stresses
Discontinuum Pillar burst Case study 3DEC 3D Mohr- Two numerical models were built using 3DEC to estimate the pillar stress before failure and to study the Dehghan et al. (2013a)
or hybrid Coulomb rockburst mechanism in order to investigate a pillar burst event that occurred in a chromite mine
method Strainburst Parametric FEM/DEM 2D Mohr- The FDEM (FEM/DEM) method was employed to study the effect of unloading rates on granite bursts Li et al. (2015)
study Coulomb (strainbursts) at a laboratory scale. A loading/unloading stage was used in the modeling to reproduce the
excavation-induced rockburst. The method could accurately and quantitatively manifest the growth,
coalescence, and nucleation of microcracks at the whole process of the granite burst
Fault-slip Case study FEM/DEM 3D Mohr- In order to investigate the effect of fault surface roughness on fault-slip rockbursts, Grisi et al. (2016) Grisi et al. (2016)
burst Coulomb investigated the failure mechanism of a gypsum pillar by comparing the continuum FEM with the hybrid FEM/
DEM. They reported that the FEM/DEM can effectively describe the mechanical response of the pillar through
fracture propagation, but it requires much more computational cost with increasing errors compared with the
continuum FEM
Strainburst Parametric CA3 3D Contact-bond CA3 was used to study strainbursts where the rock was modeled by discrete particles and the frame structure Hosseini (2016)
study model was simulated by finite elements. The main contributing factors of the rockburst severity were finally
determined
Strainburst Parametric ELFEN 2D Mohr- Feng et al. (2017a) investigated the failure mechanisms of the surrounding rock near a circular hole, taking into Feng et al. (2017a)
study Coulomb with consideration the effects of the structural surface through the ELFEN code. The mechanical failure
rotating characteristics were studied, such as the crack growth behaviors around the circular hole under the conditions
crack of different locations, lengths, and dip angles of pre-existing flaws
Strainburst Parametric PFC2D 2D Parallel-bond To investigate the dynamic stress concentration and energy evolution of a circular tunnel under blasting Li et al. (2018)
study model loading, Li et al. (2018) established a numerical model to analyze the distribution of tangential stress and
dissipated strain energy around the tunnel under various in situ stresses and different waveforms
Strainburst Parametric UDEC 2D Elastic model A novel distinct-element bonded block method was proposed to simulate strainbursts. In this method, a seismic Gao et al. (2019a, b)
study event induced by dynamic rock cracking is simulated by a strength reduction approach to initiate a dynamic
disturbance in a rock pillar, which can produce P- and S-waves with related seismic wave properties such as
velocity, the ratio of Vs/Vp, and frequency that are needed for evaluating strainbursts. Compared with most
existing methods, this proposed method overcomes the difficulty of determining the seismic wave properties of
the dynamic disturbance input
Strainburst Parametric ELFEN 2D Mohr- Different factors such as the dip angle, location, and frictional coefficient of the structural planes and the lateral Feng et al. (2019b)
study Coulomb with pressure coefficient were simulated to analyze their effects on the rock failure process and rockburst risks of a
rotating circular tunnel under excavation unloading
crack

Note: In discontinuum or hybrid method, the contact- and parallel-bond models are used for contacts, and other constitutive models are adopted for blocks or elements.

465
466 J. Wang et al. / Journal of Rock Mechanics and Geotechnical Engineering 13 (2021) 457e478
Fig. 8. Effects of different pre-existing joints on strainburst phenomena (from Vazaios et al., 2019).
stress and strength, some more complex rockburst indicators,
either empirical or user-defined, can be easily obtained by pro-
cessing the basic variables using the built-in or common pro-
graming languages (e.g. Fish language in FLAC3D, Cþþ, Python,
etc.) in most numerical programs. The use of these indicators
combined with numerical methods has become an easy, useful, and
applicable approach to predict rockburst risks qualitatively and
quantitatively for research and engineering projects. However, it
should be noted that different prediction and evaluation indicators
of rockbursts have their best application conditions and scopes.
Because most of these indicators are based on some specific cases
from different regions in the world, of which the geological con-
ditions, rock properties, geometry of engineering, and excavating-
induced effects are very different. For instance, Mitri et al. (1999)
pointed out that ERR values in a Canada mine are much lower
than those leading to rockburst problems in South African mines.
The rockburst indicators cannot be directly transposed to the
particular situation from one to another region. Furthermore,
although many researchers reported that the effectiveness of
rockburst indices had been verified with on-site monitoring results,
these indices cannot be popularized casually even in the same re-
gion due to the small number of rockburst cases and the random-
ness and complexity of rockburst mechanism. Therefore, a strict
calibration procedure should be implemented before using in-
dicators to predict and estimate rockburst tendency. A methodol-
ogy is proposed to effectively evaluate rockburst potentials using
numerical modeling combined with laboratory tests and feedbacks
from the field. A flowchart of the proposed methodology is shown
in Fig. 12. The first step is to evaluate the rockburst proneness of
rocks based on indicators at a laboratory scale. If the rock does not
possess rockburst proneness, rockbursts will not occur during rock
mass engineering in this kind of rock (Cai, 2016; Leveille et al., 2017;
Zhou et al., 2018a). The second step is to choose indicators (engi-
neering scale) initially and to predict rockburst tendency using
numerical simulation. The selection of rockburst indicators can
refer to a summary made by Zhou et al. (2018a): FAI (failure
approaching index, Zhang et al., 2011), BSR (brittle shear ratio, same
as (s1es3)/sc, Castro et al., 2012), ERR (Salamon, 1984; Kaiser et al.,
1996), BPI (Mitri et al., 1999), and LERR (Wiles, 2002; Jiang et al.,
2010) can be used for evaluating strainbursts; and ESS (Ryder,
1988), BPR (bursting potential ratio, Simon, 1999) and OBI (out-
of-balance index, Simon, 1999) are suitable for assessing fault-slip
bursts. However, Zhou et al. (2018a) did not mention what in-
dicators could be adopted to estimate pillar burst risks. Some re-
searchers (Wiles, 2002; Castro et al., 2012; Kusui et al., 2016;
Vennes and Mitri, 2017; Li et al., 2019) suggested that the ERR
(Salamon, 1984; Kaiser et al., 1996), LSS (Wiles, 2002), sc/sq (Hoek
and Brown, 1980), BSR (Castro et al., 2012), and SED (Wattimena
et al., 2012) could be employed for the assessment of pillar burst
potentials. Then the selected indicators can be calculated in nu-
merical programs, thereby letting us either determine or update
the indicators initially by comparing the occurrence area and in-
tensity of a rockburst case in the field. In the third step, the
reasonable indicators are finally obtained after the calibration with
more rockburst cases or a rockburst database of this region.
Following these three steps, engineers will be confident to use
some indicators as rockburst criteria to predict risks quantitatively
and to take appropriate measures for rockburst prevention.
 J. Wang et al. / Journal of Rock Mechanics and Geotechnical Engineering 13 (2021) 457e478
Table 3
Advantages and limitations of numerical methods in modeling rockburst mechanisms.
Numerical method Advantage
Continuum method (1) Capability of 3D modeling;
(BEM) (2) Allow the rapid assessment of stress concentration, deforma-
tion, and designs
Continuum method (1) Capability of 3D modeling;
(FEM, FDM) (2) Easily handle material heterogeneity and geometric
nonlinearity;
(3) Simulate complex behavior of rock/rock masses with various
constitutive models;
(4) Allow dynamic loading
Discontinuum (1) Capability of 3D modeling;
method (DEM) (2) Easy to model jointed rock systems;
(3) Able to model complex behavior of rock/rock masses with
various constitutive models;
(4) Allow the large deformation and detachment of blocks;
(5) Allow dynamic loading
Hybrid method (1) Able to model the extension of existing fractures and creation
(FEM/DEM) of new fractures in intact rocks;
(2) 3D modeling is possible;
(3) Allow dynamic loading
3.3. Numerical modeling for the prevention and mitigation of
rockbursts
After the burst-prone zones are predicted and evaluated, pre-
vention and mitigation measures should be carried out. Generally,
there are three approaches or steps for the prevention and miti-
gation of rockbursts, as shown in Fig. 13. Estimating the effects of
those measures in fields is very dangerous, time-consuming, and
costly. For instance, distress drilling is a very common technique to
mitigate rockburst risks in the field. The design parameters of
boreholes such as diameter, length, position, and pattern layout are
normally determined according to engineers’ specialization and
experience. However, the design parameters depend on many
factors, e.g. the size and shape of excavations, mechanical proper-
ties of rocks, and in situ stress. Thus, engineers need to spend
tremendous time and money costs to obtain relatively ideal design
parameters considering many influencing factors of distress dril-
ling. Besides, conducting experimental schemes of engineers in the
field will always make personnel be exposed in a dangerous envi-
ronment. In contrast, numerical modeling is a cheap, fast, safe, and
effective tool for the evaluation of those three types of measures,
Fig. 9. Selection of numerical methods for modeling rockburst mechanisms.
467
Disadvantage
(1) Normally elastic analysis only;
(2) Calculation time increases exponentially with number of elements (Sainoki,
2014)
(1) Input limitations, e.g. some critical input parameters are difficult to
determine;
(2) Interfaces can model simple structures, but are not suitable for highly
jointed-blocky media (Coggan et al., 2012);
(3) Rich experience in using numerical analysis is essential;
(4) Calculation time increases exponentially with number of elements
(1) Limited data on joint properties are available and the calibration of pa-
rameters might be needed;
(2) Rich experience in using numerical analysis is essential;
(3) 2D modeling is usually used due to the great calculation cost of 3D models
(1) Few data are available for contact properties and fracture mechanics
properties;
(2) The calibration of parameters is needed;
(3) Extremely long run times will require the use of parallel processing for large
models (Coggan et al., 2012);
(4) 2D modeling is usually used
especially when optimizing support design and project layout
(Mazaira and Konicek, 2015).
The studies about numerical modeling of rockburst prevention
are summarized in Table 5. As shown in Table 5, numerical
modeling has been widely employed to assess the effects of various
techniques on the prevention and mitigation of rockburst risks. For
the first type of rockburst prevention approach, the 3D modeling
with elastic analysis is a fast and effective method to determine the
areas with stress concentration and energy accumulation in the
design stage, thereby letting engineers choose rational project
layouts, and mining/excavation methods to avoid potential rock-
bursts. In the production stage, some calibrated rockburst in-
dicators can be used to estimate the rockburst tendency so as to
adjust construction schemes or adopt distress and support mea-
sures in time. In contrast, the situation of modeling ground pre-
conditioning and rock support is very complex. At present, the
main approaches to simulate the effects of destress blasting are
reducing the rock properties such as elastic modulus or adding a
stress dissipation factor to model the instantaneous stress drop
(Blake, 1972b; Tang and Mitri, 2001; Sainoki et al., 2017; Vennes
and Mitri, 2017). Although these methods are very simple to use,
no actual blastholes exist in the numerical model, which is indeed a
468 J. Wang et al. / Journal of Rock Mechanics and Geotechnical Engineering 13 (2021) 457e478

Table 4
Summary of the applications of numerical modeling for rockburst prediction.

Classification Rockburst Research Numerical Geometry Constitutive model Description Reference

indicator object code

Stress/ Strength Rock pillar Examine 3D Hoek-Brown The strength factor was selected as an indicator for assessing the Apel (2005)
strength factor ground stability and rockburst potential of a rock pillar after
benching a drift. The predicted results were further confirmed by
the in situ observations (Fig. 10)
FAI Powerhouse FLAC3D 3D Strain-softening Failure approaching index (FAI) was proposed to estimate the Zhang et al.
and tunnel stress-induced risk based on the stress state, yield surface, and (2011)
equivalent plastic shear strain. This index successfully predicted
the potential rockburst zones in facilities in the Jinping II
Hydropower Station, China
Pillar stress Rock pillar 3DEC 3D Mohr-Coulomb Compared with the tributary area theory, the numerical model Dehghan
and strength could reproduce the realistic distribution of load on pillars. The et al. (2013b)
simulated Domino failure mode of pillars was verified with the
observations in a chromite mine
sc/sq Tunnel and ABAQUS 3D Mohr-Coulomb The simulation results suggested that a distinctive ratio of UCS to Kusui et al.
rock pillar excavation-induced maximum tangential stress could be (2016)
determined for different stages of progressive failure such as
rock spalling at tunnel surfaces and pillar crushing
Damage risk Roadway FLAC 2D Mohr-Coulomb A damage risk index based on Mohr-Coulomb criterion was Wang et al.
index proposed to study the rockburst risks and dynamic response (2017)
characteristics of roadway surrounding rocks under dynamic
loading. The potential rockburst zones and severity could be
predicted with the proposed index
Brittle Tunnel FEM 3D A self-developed After developing a brittle-yield criterion combining damage Diederichs
fracture brittle-yield model initiation and spalling limit (Diederichs, 2007), Diederichs (2018)
zone (2018) successfully predicted the development of the burst
potential of a deep tunnel with respect to advance through
incorporating the criterion in a 3D nonlinear finite element
analysis
(sq þ sL)/sc Tunnel ABAQUS 3D Elastic model An analytical method was developed to investigate the Yang et al.
inhomogeneous stress concentration in interbedded strong and (2018)
weak layers which was confirmed by FEM modeling. Numerical
results also suggested that rockburst risks change with a
mutative orientation of rock layers
MS events Tunnel PFC3D 3D Bonded particle model Ma et al. (2019) proposed a brittle fracture model and a Ma et al.
(simulated numerical simulation method of hard rock fracture based on MS (2019)
fractures) information in PFC3D. The simulation results agreed well with
field data in tunnel engineering, showing that the proposed
method is an effective tool for interpreting and detecting early-
warning signs of rockburst hazards
Energy ERR, BPI Stope FEM 2D Elastic A new indicator, BPI, was proposed based on the energy storage Mitri et al.
rate and critical energy density. The ERR and BPI were adopted to (1999)
predict rockburst potentials in a cut-and-fill stope
LERR Tunnel FLAC3D 3D Elastic-brittle-plastic Jiang et al. (2010) proposed a new LERR index considering the Jiang et al.
model difference in energy stored in rock masses before and after (2010)
brittle failure based on an elastic-brittle-plastic model. The
simulation results suggested that the new LERR could
successfully predict the intensity and failure depth of rockbursts
in deep tunnels
URLERI Tunnel FLAC3D 2D A self-developed A new energy indicator of rockbursts, the unit time relative local Yang et al.
elastoplastic model energy release index (URLERI), was proposed and used for (2015)
simulating the rockburst case in the Jinping II Hydropower
Station, China. The simulated results agreed well with the field
monitoring results
Accumulated Mining level FLAC3D 3D e The accumulated disturbance energy in surrounding rocks was Cai (2016)
disturbance calculated in FLAC3D, and then was employed to predict the
energy magnitude of rockbursts in each mining level
sq/sc, BPI Stope and ABAQUS 3D Mohr-Coulomb (for A mine-scale model was established to capture the realistic Sepehri
drift rock mass) and elastic distribution of in situ stress and mining-induced stress. sq/sc and (2016),
model (for backfill) BPI were adopted as indicators to evaluate the rockburst risks of Sepehri et al.
stopes and drifts in an underground diamond mine (Fig. 11) (2020)
Excess Tunnel UDEC 2D Strain-softening A DEM/BEM hybrid model was built to study seismicity and Khademian
energy unstable failures in circular excavations using a proposed et al. (2016)
concept of “excess energy”. The results confirmed that a
lower horizontal to vertical stress ratio resulted in more
unstable excess energy of rock masses
Elastic Stope ABAQUS 2D Elastic model The elastic deformation energy density was employed to obtain Cieslik et al.
deformation the factors and conditions that determine the location of the (2017)
energy sources and the mechanism of mining-induced tremors. The
density shear strain energy could be treated as an index of the potential
threat of mining-induced tremors
s1, SED Stope FLAC3D 3D Mohr-Coulomb The rockburst proneness of stopes with different mining Ma et al.
sequences was analyzed through simulating the variations of (2018)
 J. Wang et al. / Journal of Rock Mechanics and Geotechnical Engineering 13 (2021) 457e478
Table 4 (continued )
Classification Rockburst Research Numerical Geometry Constitutive model
indicator object code
SED Rock pillar FLAC3D 3D Mohr-Coulomb
Fig. 10. Redistribution of the strength factor (a) before and (b) after floor benching (after Apel, 2005).
type of equivalent approach. Moreover, specific zones with poten-
tial blasting-induced damages need to be assumed first to assign
parameters, which is not real and increases extra efforts. By
analyzing a pressure profile of ANFO-type explosive and the
propagation of detonation, Sainoki et al. (2017) proposed an inno-
vative method that can simulate a time-varying blast pressure.
Using this method, the blasting-induced damage can be simulated
more precisely. Nevertheless, the volume dilation of rocks induced
by the creation and development of fractures was neglected,
because the continuum method was used in the previous work. A
Trigon method developed by Gao (2013) can simulate the obvious
dilation phenomenon caused by the generation and propagation of
cracks, which can be further adopted to evaluate the blasting-
induced damage and the dissipated energy induced by rock frac-
turing. The numerical modeling of water infusion is less studied
owing to the infrequent usage of water infusion for hard rock en-
gineering. Currently, numerical simulation is mainly utilized to
examine the effects of water infusion on mitigating coal burst risks
(Li et al., 2005; Fan et al., 2012; Song et al., 2014; Liu et al., 2017;
Zhou et al., 2018b). Numerical modeling of rock support systems is
probably one of the most important and extended applications of
numerical methods to underground excavations, and its impor-
tance is even greater when designing support systems in burst-
prone grounds (Mazaira and Konicek, 2015). Numerical analyses
in this field mainly focus on the behavior of support elements such
as rockbolts and steel arches under dynamic loading, or stress
waves produced during rockbursts. The main problem is that re-
searchers need to develop reasonable models to reproduce the
realistic response of support elements, requiring many efforts to
469
Description Reference
maximum principal stress and elastic strain energy
accumulation in surrounding rocks
After conducting a series of SHPB tests on rock specimens, Li Li et al.
et al. (2019) obtained the formula of stored strain energy by (2019)
rejecting the fractured dissipative energy based on the fitting of
experimental data. The formula was programmed into FLAC3D
as a variable to analyze the strain energy and burst potential of
rock pillars
calibrate the models with experimental data or in situ observations.
Besides, engineering experience shows that the failure of rein-
forcement elements often occurs in the rockboltegrout interface
rather than the grouterock interface (Marambio et al., 2018).
However, the input parameters for modeling rockboltegrout in-
terfaces are usually set based on researchers’ experience, which
need to be further investigated to capture the full mechanical
behavior of rockbolt supporting.
4. Limitations and future research
4.1. Dynamic modeling of rockbursts
As mentioned above, rockburst is a kind of dynamic disasters
that develop from the static accumulation of mining-induced stress
and strain energy. However, only a small portion of research
(28.66%, as shown in Fig. 7b) used dynamic modeling to study
rockburst-related problems. Additionally, most existing indices or
indicators of rockbursts that can be used for numerical calculations
are proposed based on static simulation results, which can only
make static and qualitative evaluations of rockburst risks, and
cannot reasonably reflect the dynamic process of actual rockbursts
(Yang et al., 2015). Therefore, the dynamic modeling of rockbursts is
necessary. Some research work has already been implemented to
investigate the effects of dynamic load or stress wave on rockbursts.
But some questions remain:
(1) Dynamic mechanical properties of rocks and discontinuities.
The reality and accuracy of numerical modeling results
470 J. Wang et al. / Journal of Rock Mechanics and Geotechnical Engineering 13 (2021) 457e478
Fig. 11. Estimation of rockburst tendency with sq/sc (Ts) and BPI in a kimberlite pipe (after Sepehri, 2016).
depend on the mechanical properties of rocks and disconti-
nuities. Only if the mechanical properties of rocks and dis-
continuities are scientific and reasonable can the numerical
modeling results accurately reflect the actual deformation
and damage of rocks or rock masses on site. It is well known
that the mechanical properties of rocks and discontinuities
are related to loading conditions and loading rates (strain
rates). Up to now, a large amount of research has been con-
ducted to investigate the dynamic mechanical parameters of
rocks in the laboratory (Li et al., 1993, 1994; Demirdag et al.,
2010; Liu et al., 2012; Wang et al., 2016). Unfortunately, most
present research that uses dynamic modeling adopts the
static mechanical properties of rocks to study rockbursts. On
the other hand, although a great deal of research has been
done on the mechanical properties of rocks under dynamic
loading in the laboratory, the study of dynamic mechanical
properties of discontinuities is rare, and thus the dynamic
mechanical properties of discontinuities are even less
frequently used for the numerical simulation of rockbursts.
For instance, Gao et al. (2019a, b) stated that the joint
constitutive model in UDEC was obtained under the condi-
tions of static loading. Whether this model can be used to
simulate the dynamic response of jointed rock masses is
worthy of further study. Therefore, the employment of the
dynamic mechanical parameters of rocks and discontinuities
as well as their constitutive relationships considering the
loading rate (strain rate) in numerical modeling could be a
research topic for rockburst-related problems in the future.
(2) The heterogeneity of rocks and rock masses under dynamic
loading. Manouchehrian and Cai (2018) pointed out that using
homogenous material models may not be able to capture the
realistic failure processes and patterns even if the overall
stressestrain curve properly reflects the prescribed mechan-
ical properties. To date, the material heterogeneity has been
introduced into some numerical models to investigate the
influence of rock heterogeneity on dynamic rock failure and
associated energy release. However, the widely used numer-
ical models are only 2D in laboratory scale, and the simulation
results are not accurate and reliable. Hence, the simulation of
the effects of rock heterogeneity under dynamic loading on
rockbursts with 3D models at an engineering scale needs to be
further studied. In addition, the distribution of rock hetero-
geneity is usually modeled with a normal distribution or
Weibull distribution, while the real distribution of the het-
erogeneity of rocks is generally stochastic. Fortunately,
recently developed 3D scanning and 3D printing technology
can capture the stochastic distribution of rock heterogeneity,
which will be a good choice for addressing this problem when
combined with numerical modeling.
(3) Self-initiated dynamic disturbance. Researchers usually
employ an assumed dynamic disturbance such as a sine P-
waveform on the model boundary or inside the model to
trigger unstable failures, which does not represent real
 J. Wang et al. / Journal of Rock Mechanics and Geotechnical Engineering 13 (2021) 457e478
Fig. 12. A methodology to evaluate rockburst potentials.
situations. Gao et al. (2019a, b) proposed a novel method:
simulating a seismic event induced by dynamic rock frac-
turing through a strength reduction approach to initiate a
dynamic disturbance in a rock pillar, which can produce P-
and S-waves with related seismic wave properties such as
velocity, the ratio of Vs/Vp, and frequency that are needed for
evaluating strainbursts. Compared with most existing
methods, this proposed method overcomes the difficulty of
determining seismic wave properties of the dynamic
disturbance input, which makes a significant step forward in
Fig. 13. Methods to reduce damaging effects of excessive stress in underground mining
(after Mitri, 2000).
471
the simulation of rockburst processes. Nonetheless, the
location of the dynamic disturbance in the rock pillar is also
manually determined, while in the field, the rock fracturing
occurs randomly in many positions surrounding excavations
and then produces a series of seismic waves or dynamic
disturbances that may trigger rockbursts. Thus, one needs to
address the problem that makes the rock fracturing induce
seismic waves or dynamic disturbances spontaneously
instead of assuming the positions of dynamic disturbances
artificially. To handle this problem, theoretical and laboratory
research is necessary to investigate the relationship between
the properties of dynamic disturbances and rock fracturing.
But the numerical fulfillment of that relationship in numer-
ical modeling is quite sophisticated and a large amount of
programming work needs to be done.
4.2. Modeling of yielding support elements
Up to now, yielding supports or energy-absorption supports have
been well developed owing to the tireless efforts of scholars. Veri-
fications have shown that many types of yielding rockbolts, e.g. cone
bolt (Ortlepp, 1992), Swellex rockbolt (Wijk and Skogberg, 1982),
Roofex (Charette and Plouffe, 2007), D-bolt, Garford bolt (Varden
et al., 2008), Yield-Lok (Wu et al., 2011), and constant-resistance
and large-deformation bolt (CRLDB) (He et al., 2012) can control
rockbursts effectively. Unfortunately, the development of yielding
support elements or yielding support models in numerical programs
is lagging significantly behind the development process of those
yielding supports in engineering practice. For instance, the con-
ventional support elements (e.g. rockbolts) in some numerical
programs are not able to handle high loads and withstand large
deformations to control rockbursts. Thus, it is imperative to develop
appropriate support elements or support models in numerical
programs that can accurately simulate yielding support systems to
estimate their performance during rockbursts. In addition, the
properties of contact between rockbolts and grout, and the inter-
action or coupling effects between different support fashions (e.g.
rockbolts and shotcrete) under dynamic loading are also expected to
be further investigated to assess the full behavior of integrated
yielding support systems in burst-prone areas.
4.3. Modeling existing discontinuities and new fractures in 3D
models
Stead et al. (2004) and Coggan and Stead (2005) pointed out that
the continuum and discontinuum methods are limited by their
inability to capture the interaction between existing discontinuities
and new fractures created in intact rocks. Fortunately, recently
developed hybrid methods (e.g. DEM/FEM, DEM/FDM) provide the
solution for this problem and some studies are currently underway
(Feng et al., 2017a; Vazaios, 2018; Vazaios et al., 2019). However, the
current hybrid numerical models are usually 2D. It is reasonable that
simulating a long tunnel based on the assumption of the plane strain
condition, but it is invalid if the research objective is a rock pillar or a
stope or more complicated excavations when the 2D model is
applied. More importantly, the discontinuity consists of two rough
surfaces, or two rough surfaces sandwiched with soft rocks or clays.
It has both width and length, and thus it is neither accurate nor
realistic to model it only as a line in a 2D model. Elmo et al. (2005)
proposed that the out-of-plane continuity would lead to an easier
creation of blocks, and in a 3D pillar model there would be greater
freedom for failures to occur due to the lack of confinement coming
out of the 2D plane, which means that the 2D model tends to result in
an overestimation of pillar strengths. Thus, the 3D modeling will
472 J. Wang et al. / Journal of Rock Mechanics and Geotechnical Engineering 13 (2021) 457e478

Table 5
Summary of the applications of numerical modeling for rockburst prevention.

Rockburst Numerical Geometry Constitutive model Support Description Reference

prevention code element
approach

Alternative mining FLAC3D 3D Elastic model (for ERR and ESS), and A top pilot tunnel drill-and-blast method was proposed Zhang et al.
methods the cohesion and evaluated using numerical simulation with several (2012b)
weakening and frictional prediction indices of rockbursts (FAI, ERR, and ESS). The
strengthening model (for FAI) simulation and field monitoring results showed that the
pretreatment aspect of this method could effectively
reduce the risks of strainburst and fault-slip rockburst
compared with the conventional full-face TBM
excavation method
NFOLD, 3D Strain-softening Numerical modeling was used to assess the effects of Castro et al.
Map3D mining sequence, advance direction and backfill mining (2012)
method on the mitigation of the risks of pillar burst and
fault-slip burst
FLAC3D 3D Mohr-Coulomb Zhang et al. (2018) used FLAC3D to study different sized Zhang et al.
sections of tunnels, considering circumferential and (2018)
radial stresses, and found that it is possible to reduce the
intensity of rockbursts by increasing the size of the
excavation inside the tunnel. The numerical simulation
results have also been verified by the statistical data of
rockbursts in field tests
Ground UDEC 2D Elastic model UDEC was employed to simulate the effects of fluid Lightfoot (1993)
preconditioning injection on the control of fault-slip and to model
preconditioning the stope by blasting. Results
suggested that the injection of high gas into fractured
rocks was more likely to be the best preconditioning
method
FLAC3D 3D Drucker-Prager FEM modeling was used to examine different layouts of Zhu et al. (2009)
destressed holes under different in situ stress
conditions to reduce rockburst risks
FLAC3D 3D Mohr-Coulomb Sainoki et al. (2017) compared the simulation results of Sainoki et al.
the traditional modeling approach with the alternative (2017)
modeling approach. They found that the traditional
modeling approach may lead to an overly optimistic
indication of destress blasting efficiency
FLAC3D 3D Elastic model To investigate the effects of large-scale choked panel Vennes and
destress blasting on stress reduction in a highly- Mitri (2017)
stressed remnant ore pillar, numerical modeling
was adopted by Vennes and Mitri (2017) to study the
stress change and BSR before and after distressing in the
pillar. They concluded that the panel destressing
method can significantly reduce rockburst risks when
mining a highly-stressed ore pillar (see Fig. 14)
Rock support FEM 2D Drucker-Prager (for rock) and von Self- A type of energy-absorption coupling support fashion Lv and Pan
Mises (for steel) developed was proposed by Lv and Pan (2010). The effects of this (2010)
element new support fashion were examined using FEM
modeling by comparing a roadway maintained by this
fashion with the one without it
FLAC3D 3D Mohr-Coulomb (for rock) and Cable element Mortazavi and Alavi (2013) studied the behavior of Mortazavi and
elastic model (for rockbolt) three types of fully grouted rockbolts (with or without Alavi (2013)
head plate, and yielding rockbolt) under dynamic
loading and concluded that the yielding rockbolt was
the best choice for absorbing dynamic stress waves and
controlling rock mass movement
DDA 2D Elastic and strain-hardening Self- Rockbolt models were developed using DDA to Nie et al. (2014)
developed investigate the failure mechanism of an expansion-
element shell-anchored bolt, the Split Set, the fully grouted rebar
and the D-bolt under static and dynamic loads
FLAC 2D Mohr-Coulomb Beam element FLAC simulation was used to investigate the effects of Masny et al.
the dynamic load on the arch steel support in stress (2017)
concentration zones. It was found that the higher the
peak particle velocity (PPV), the higher the HMH stress
(a type of stress developed by Huber, von Mises and
Hencky), and that the initial static stress level plays an
important role in controlling the failure of arch steel
support induced by dynamic loads
FLAC3D 3D Elastic-perfectly plastic Cable element Numerical modeling was applied for a laboratory-scale Marambio et al.
test to investigate the behavior of threadbar in dynamic (2018)
load testing. The simulation results corresponded well
with laboratory observations. It was also found that the
grout played a secondary role in the overall strength of
the system, a phenomenon that needs to be further
studied in dynamic conditions
ABAQUS 3D Strain-softening Sengani (2018)
 J. Wang et al. / Journal of Rock Mechanics and Geotechnical Engineering 13 (2021) 457e478
Table 5 (continued )
Rockburst Numerical Geometry Constitutive model
prevention code
approach
DDA 2D e Self-
provide useful insight into the effects of existing discontinuities and
new fractures on rockbursts. Further work needs to be undertaken to
establish 3D models that incorporate existing discontinuities and
new fractures created in intact rocks.
4.4. A systematic numerical modeling approach for rockbursts
Owing to the complicated mechanisms and the uncertainty of
occurrence, rockburst is different from other rock mechanics
problems such as stable failure modes (e.g. rock spalling) after
tunnel excavation and ore extraction. Many contributing factors for
simulating rockbursts should be considered when modeling this
complex phenomenon. Thus, it is necessary to propose an approach
of numerical modeling for rockbursts in order to provide re-
searchers with a systematic and reasonable numerical modeling
framework. The selection of numerical modeling approaches, nu-
merical programs, numerical modeling sequences, parameters, and
Fig. 14. Stress state in the pillar before and after distressing (after Vennes and Mitri, 2017). Stress in Pa.
473
Support Description Reference
element
Self- Sengani (2018) adopted different support fashions
developed under dynamic conditions and conducted a large-scale
element numerical model for an ultra-deep platinum mine using
the ABAQUS explicit finite element solver in order to
study the performance of in-stope pillars. The results
indicated that extensive scaling and fracturing occur in
most bolt-reinforced in-stope pillars, while in-stope
pillars supported by bolt and shotcrete sustain minor or
no damage during the dynamic movement
A new energy-absorbing rockbolt called deformation- Yokota et al.
developed controlled rockbolt (DC-bolt) was proposed by Yokota (2019)
element et al. (2019). The performance of the DC-bolt was
simulated by the DDA and results suggested that the
DC-bolt possessed both high loading capacity and
deformation capacity compared with the fully grouted
rockbolt
model calibration should be described in this method. The first step
is the preparation which includes problem analysis, selection of
numerical methods, and geometry analysis of the research objec-
tive. The second step includes five procedures: model establish-
ment and meshing, input of in situ stress, constitutive models and
rock mass parameters, initial and boundary conditions, geostatic
step and model calibration, and analysis of simulation results. The
details of this approach are shown in Fig. 15.
5. Conclusions
This paper summarizes the categories and the origin of different
numerical approaches for modeling rockbursts and the current
usage of various numerical approaches is investigated. A state-of-
the-art review of the application of numerical modeling for
studying rockbursts is implemented. This paper also discusses the
limitations and prospects of rockburst numerical modeling. An
474 J. Wang et al. / Journal of Rock Mechanics and Geotechnical Engineering 13 (2021) 457e478

Fig. 15. Flowchart of a systematic numerical modeling approach for rockbursts.

 J. Wang et al. / Journal of Rock Mechanics and Geotechnical Engineering 13 (2021) 457e478 475

approach of numerical modeling for rockbursts is proposed so as to simulate yielding support systems to estimate their perfor-
provide researchers with a systematic and reasonable numerical mance during rockbursts. In addition, the properties of
modeling framework. The following conclusions are drawn from contact between rockbolts and grout, and the interaction or
this study: coupling effects between different support fashions under
dynamic loading, are also expected to be further investigated
(1) Rockburst-related problems have been investigated by re- to assess the full behavior of integrated yielding support
searchers using continuum, discontinuum and hybrid nu- systems in burst-prone areas.
merical methods and various numerical programs since the (3) Further work also needs to be undertaken through the
1970s. The main tendency of rockburst numerical modeling recently developed hybrid methods to establish 3D models
is to move from continuum methods to discontinuum that incorporate existing discontinuities and new fractures
methods and then to hybrid methods. However, the most created in intact rocks.
common numerical approach is still the continuum method.
It is another tendency that using dynamic simulation to Declaration of competing interest
investigate rockbursts in this field.
(2) The initial studies of rockburst mechanisms through nu- The authors wish to confirm that there are no known conflicts of
merical modeling were based on some specific engineering interest associated with this publication, and there has been no
cases, in which the continuum method, elastic constitutive significant financial support for this work that could have influ-
relationship and 2D model were very often used. With the enced its outcome.
development of computing capacities and progress in nu-
merical programs for modeling complicated geometry and
plastic nonlinearities, more researchers utilized elastoplastic Acknowledgments
constitutive relationships and/or 3D models. Recently, the
development of DEM and other hybrid numerical methods The authors gratefully acknowledge financial support from the
have made it possible to simulate discontinuous deformation China Scholarship Council (Grant No. 201808370185).
to reproduce rockburst phenomena. The selection of nu-
merical methods should be based on the match between the List of symbols
capabilities of numerical codes and rockburst mechanisms,
and the specific engineering situation. sc Uniaxial compressive strength of rock
(3) Rockburst potentials can be predicted qualitatively and sq Maximum tangential stress of surrounding rock
quantitatively using numerical modeling and a number of s1 Maximum principal stress
prediction and evaluation indices or indicators based on st Uniaxial tensile strength of rock
strength and energy criteria. These indicators have their best sL Axial stress of tunnel
application conditions and scopes and cannot be popularized sz Vertical stress
casually. A strict calibration procedure should be imple- g Unit weight
mented before using indicators to predict and estimate n Poisson’s ratio
rockburst tendency. a Tao discriminant index
(4) Numerical modeling is a cheap and effective tool to evaluate D Diameter
the measures used for the prevention and mitigation of Eh Average deformation modulus of the upper part of the
rockbursts. The dilation phenomenon of rocks should be Earth’s crust measured in a horizontal direction
considered to assess the blasting-induced damage and the Es Unloading tangential modulus of rock
dissipated energy induced by rock fracturing when modeling k Ratio of the horizontal stress to the vertical stress
the distress blasting. The input parameters of rockboltegrout L Length
interfaces need to be further investigated to capture the full Ts sq/sc
mechanical behavior of rockbolt supporting. Vs Velocity of secondary wave
Vp Velocity of primary wave
To date, great achievements have been made by numerous re- Wk Kinetic energy
searchers in employing numerical modeling to investigate the Wr Released energy
mechanism, evaluation and prediction, and prevention and miti- Wsp Stored elastic energy
gation of rockbursts. Nevertheless, rockburst is a very complicated Wst Dissipated elastic energy
geological hazard and the research of rockbursts through numerical Z Depth
modeling is always ongoing: B Brittleness coefficient

(1) One possible research topic for rockburst-related problems is References

to use the dynamic mechanical parameters of rocks and
discontinuities as well as their constitutive relationships to
consider the loading rate (strain rate) in numerical modeling.
Using recently developed 3D scanning and 3D printing
technology would be a good way to capture the stochastic
distribution of rock heterogeneity. Theoretical and labora-
tory research is necessary to investigate the relationship
between the properties of dynamic disturbances and rock
fracturing. A large amount of programming work needs to be
done to fulfill that relationship in numerical modeling.
(2) It is imperative to develop appropriate support elements or
support models in numerical programs that can accurately
Afraei, S., Shahriar, K., Madani, S.H., 2019. Developing intelligent classification
models for rock burst prediction after recognizing significant predictor vari-
ables, Section 1: literature review and data preprocessing procedure. Tunn.
Undergr. Space Technol. 83, 324e353.
Apel, D.B., 2005. Design and development of an underground hard-rock mine with
the assistance of rock mechanics software. Min. Eng. 57 (9), 83e88.
Arndt, S.M., Fillery, B.P., Beck, D.A., 2009. Advances in mining simulation. In: Pro-
ceedings of 2009 SIMULIA Customer Conference. Dassault Systèmes.
Bardet, J.P., 1990. Finite element analysis of surface instability in hypo-elastic solids.
Comput. Methods Appl. Mech. Eng. 78 (3), 273e296.
Bardet, J.P., 1987. Numerical assessment of rockbursting, vol. 8618. Agbabian As-
sociates El Segundo Ca, pp. 55e109, 6171.
Barton, N., Lien, R., Lunde, J., 1974. Engineering classification of rock masses for the
design of tunnel support. Rock Mech. Rock Eng. 6 (4), 189e236.
476 J. Wang et al. / Journal of Rock Mechanics and Geotechnical Engineering 13 (2021) 457e478
Bennett, T.J., Marshall, M.E., 2001. Identification of rockbursts and other mining
events using regional signals at international monitoring system stations. In:
Proceedings of the 23rd Seismic Research Review: Worldwide Monitoring of
Nuclear Explosions. Jackson Hole, USA.
Blake, W., Hedley, D.G., 2003. Rockbursts: Case Studies from North American Hard-
Rock Mines. Society for Mining, Metallurgy, and Exploration (SME).
Blake, W., 1972a. Rock-Burst Mechanics. Colorado School of Mines, Golden, USA.
PhD Thesis.
Blake, W., 1972b. Rock-Burst Mechanics. Quarterly of the Colorado School of Mines
67, 1e64.
Board, M.P., Voegele, M.D., 1981. Examination and demonstration of undercut and
fill stoping for ground control in deep vein mining. In: Stephansson, O.,
Jones, M.J. (Eds.), Application of Rock Mechanics to Cut and Fill Mining. Insti-
tution of Mining & Metallurgy (IMM), London, UK.
Brady, B.H., Brown, E.T., 2004. Rock mechanics: For Underground Mining, 3rd ed.
Springer.
Brady, B.H.G., 1979. Boundary Element Methods for Mine Design. Imperial College
of Science and Technology, London, UK. PhD Thesis.
Cai, M., Kaiser, P.K., Morioka, H., Minami, M., Maejima, T., Tasaka, Y., Kurose, H.,
2007. FLAC/PFC coupled numerical simulation of AE in large-scale underground
excavations. Int. J. Rock Mech. Min. Sci. 44 (4), 550e564.
Cai, M., 2016. Prediction and prevention of rockburst in metal mines e a case study
of Sanshandao gold mine. J. Rock Mech. Geotech. Eng. 8 (2), 204e211.
Cai, M., 2013. Principles of rock support in burst-prone ground. Tunn. Undergr.
Space Technol. 36, 46e56.
Castro, L.A., Bewick, R.P., Carter, T.G., 2012. An overview of numerical modelling
applied to deep mining. In: Sousa, L.R., Vargas, E., Fernandes, M.M., Azevedo, R.
(Eds.), Innovative Numerical Modelling in Geomechanics. CRC Press, London,
UK, pp. 393e414.
Charette, F., Plouffe, M., 2007. RoofexÒ e results of laboratory testing of a new
concept of yieldable tendon. In: Proceedings of the 4th International Seminar
on Deep and High Stress Mining. Australian Centre for Geomechanics, Perth,
Australia, pp. 395e404.
Cieslik, J., Burtan, Z., Chlebowski, D., Zorychta, A., 2017. Geomechanical analysis of
location and conditions for mining-induced tremors in LGOM copper mines.
J. Sustain. Min. 16 (3), 94e103.
Coggan, J., Gao, F., Stead, D., Elmo, D., 2012. Numerical modelling of the effects of
weak immediate roof lithology on coal mine roadway stability. Int. J. Coal Geol.
90e91, 100e109.
Coggan, J.S., Stead, D., 2005. Numerical modelling of the effects of weak mudstone
on tunnel roof behaviour. In: Proceedings of the 58th Canadian Geotechnical
Conference. Saskatoon, Canada.
Cundall, P.A., Strack, O.D.L., 1979. The development of constitutive laws for soil
using the distinct element method. In: Numerical Methods in Geomechanics,
Proceedings of the 3rd International Conference on Numerical Methods in
Geomechanics. A.A. Balkema, Rotterdam, the Netherlands, pp. 289e317.
Cundall, P.A., 1971. A computer model for simulating progressive, large-scale
movement in blocky rock systems. In: Proceedings of the Symposium of the
International Society of Rock Mechanics. Nancy, France.
Cundall, P.A., 1988. Formulation of a three-dimensional distinct element model e
Part I. A scheme to detect and represent contacts in a system composed of many
polyhedral blocks. Int. J. Rock Mech. Min. Sci. Geomech. Abstr. 25 (3), 107e116.
Daehnke, A., 1999. Stress wave and fracture propagation in rock. In: 9th ISRM
Congress. A.A. Balkema.
Dehghan, S., Shahriar, K., Maarefvand, P., Goshtasbi, K., 2013a. 3D modeling of rock
burst in pillar No. 19 of Fetr6 chromite mine. Int. J. Rock Mech. Min. Sci. Technol.
23 (2), 231e236.
Dehghan, S., Shahriar, K., Maarefvand, P., Goshtasbi, K., 2013b. 3D numerical
modelling of Domino failure of hard rock pillars in Fetr6 Chromite Mine, Iran,
and comparison with empirical methods. J. Cent. S. Univ. 20 (2), 541e549.
Demirdag, S., Tufekci, K., Kayacan, R., Yavuz, H., Altindag, R., 2010. Dynamic me-
chanical behavior of some carbonate rocks. Int. J. Rock Mech. Min. Sci. 47 (2),
307e312.
Diederichs, M.S., 2018. Early assessment of dynamic rupture hazard for rockburst
risk management in deep tunnel projects. J. S. Afr. Inst. Min. Metall. 118 (3),
193e204.
Diederichs, M.S., 2007. The 2003 Canadian Geotechnical Colloquium: mechanistic
interpretation and practical application of damage and spalling prediction
criteria for deep tunnelling. Can. Geotech. J. 44 (9), 1082e1116.
Durrheim, R., 2010. Mitigating the risk of rockbursts in the deep hard rock mines of
South Africa: 100 years of research. In: Brune, J. (Ed.), Extracting the Science: A
Century of Mining Research. SME, pp. 156e171.
Elmo, D., Coggan, J.S., Pine, R.J., 2005. Characterization of rock mass strength using a
combination of discontinuity mapping and fracture mechanics modelling. In:
Rock Mechanics for Energy, Mineral and Infrastructure Development in the
Northern Regions: Proceedings of 40th US Rock Mechanics Symposium (Alaska
rocks 2005). American Rock Mechanics Association (ARAM), pp. 98e110.
Fakhimi, A., 2009. A hybrid discrete-finite element model for numerical simulation
of geomaterials. Comput. Geotech. 36 (3), 386e395.
Fall, M., Célestin, J.C., Pokharel, M., Touré, M., 2010. A contribution to understanding
the effects of curing temperature on the mechanical properties of mine
cemented tailings backfill. Eng. Geol. 114 (3e4), 397e413.
Fan, J., Dou, L., He, H., Du, T., Zhang, S., Gui, B., Sun, X., 2012. Directional hydraulic
fracturing to control hard-roof rockburst in coal mines. Int. J. Min. Sci. Technol.
22 (2), 177e181.
Feng, F., Li, X., Li, D., Chen, C., 2017a. Numerical analysis of failure characteristics of
circular hole specimen containing pre-existing flaw based on finite/discrete
elements. Rock Soil Mech. 38 (Supp.2), 337e348 (in Chinese).
Feng, F., Li, X., Rostami, J., Li, D., 2019b. Modeling hard rock failure induced by
structural planes around deep circular tunnels. Eng. Fract. Mech. 205, 152e174.
Feng, X.T., Liu, J., Chen, B., Xiao, Y., Feng, G., Zhang, F., 2017b. Monitoring, warning,
and control of rockburst in deep metal mines. Engineering 3 (4), 538e545.
Feng, X.T., Xiao, Y.X., Feng, G.L., Yao, Z.B., Chen, B.R., Yang, C.X., Su, G.X., 2019a. Study
on the development process of rockbursts. Chin. J. Rock Mech. Eng. 38 (4), 649e
673 (in Chinese).
Gao, F., Kaiser, P.K., Stead, D., Eberhardt, E., Elmo, D., 2019a. Numerical simulation of
strainbursts using a novel initiation method. Comput. Geotech. 106, 117e127.
Gao, F., Kaiser, P.K., Stead, D., Eberhardt, E., Elmo, D., 2019b. Strainburst phenomena
and numerical simulation of self-initiated brittle rock failure. Int. J. Rock Mech.
Min. Sci. 116, 52e63.
Gao, F., 2013. Simulation of Failure Mechanisms Around Underground Coal Mine
Openings using Discrete Element Modelling. Simon Fraser University, Vancou-
ver, Canada. PhD Thesis.
Gao, F., 2019. Use of numerical modeling for analyzing rock mechanic problems in
underground coal mine practices. J. Min. Strata Contr. Eng. 1 (1), 013004 https://
doi.org/10.13532/j.jmsce.cn10-1638/td.2019.02.009 (in Chinese).
Goodman, R.E., Shi, G.H., 1985. Block Theory and its Application to Rock Engineer-
ing. Prentice-Hall, Inc., Englewood Cliffs, USA.
Grisi, S., Castellanza, R., Agliardi, F., Crosta, G., Elmo, D., 2016. Failure mechanism of a
gypsum pillar by 2D and 3D FEM/DEM numerical analyses. In: ISRM Interna-
tional Symposium e EUROCK 2016. Taylor & Francis Group.
Guo, X., Zhao, Z., Gao, X., Ma, Z., Ma, N., 2019. The criteria of underground rock
structure failure and its implication on rockburst in roadway: a numerical
method. Shock Vib. 2019, 7509690 https://doi.org/10.1155/2019/7509690.
Hart, R.D., 1980. Geotechnical modelling of a cut and fill mining operation. In:
Proceedings of the Conference on the Application of Rock Mechanics to Cut and
Fill Mining. Lulea. Institute of Mining and Metallurgy, Sweden, pp. 307e315.
Hazzard, J.F., Young, R.P., 2002. Moment tensors and micromechanical models.
Tectonophysics 356 (1e3), 181e197.
He, M., Xia, H., Jia, X., Gong, W., Zhao, F., Liang, K., 2012. Studies on classification,
criteria and control of rockbursts. J. Rock Mech. Geotech. Eng. 4 (2), 97e114.
Hedley, D.G., 1987. Rockburst Mechanics. Canada Centre for Mineral and Energy
Technology, Ottawa, Australia.
Hedley, D.G.F., 1992. Rockburst Handbook for Ontario Hard Rock Mines. Energy,
Mines and Resources, Canada Centre for Mineral and Energy Technology,
Ottawa, Australia. CANMET Special Report SP92-1E.
Hoek, E., Brown, E.T., 1980. Underground Excavation in Rock. Institute of Mining and
Metallurgy, London, UK, p. 527.
Hosseini, O., 2016. Numerical Simulation of Rock Burst of Underground Pillars. New
Mexico Institute of Mining and Technology, Socorro, USA. MSc Thesis.
Itasca Consulting Group, Inc, 2000. FLAC (fast Lagrangian analysis of continua).
Itasca Consulting Group, Inc., Minneapolis, USA.
Jager, A.J., Ryder, J.A., 1999. A Handbook on Rock Engineering Practice for Tabular
Hard Rock Mines. Safety in Mines Research Advisory Committee, Johannesburg,
South Africa, p. 46.
Jiang, F.F., Zhou, H., Liu, C., Sheng, J., 2019. Progress, prediction and prevention of
rockbursts in underground metal mines. Chin. J. Rock Mech. Eng. 38 (5), 956e
972 (in Chinese).
Jiang, J.Q., 2017. Study on Strainburst in Tunnel using True-Triaxial Test and its Ki-
netic Energy Prediction. PhD Thesis. Guangxi University, Nanning, China (in
Chinese).
Jiang, Q., Feng, X.T., Xiang, T.B., Su, G.S., 2010. Rockburst characteristics and nu-
merical simulation based on a new energy index: a case study of a tunnel at
2500 m depth. Bull. Eng. Geol. Environ. 69 (3), 381e388.
Jing, L., Hudson, J.A., 2002. Numerical methods in rock mechanics. Int. J. Rock Mech.
Min. Sci. 39 (4), 409e427.
Jing, L., 2003. A review of techniques, advances and outstanding issues in numerical
modelling for rock mechanics and rock engineering. Int. J. Rock Mech. Min. Sci.
40 (3), 283e353.
Kaiser, P.K., Cai, M., 2012. Design of rock support system under rockburst condition.
J. Rock Mech. Geotech. Eng. 4 (3), 215e227.
Kaiser, P.K., Tannant, D.D., McCreath, D.R., 1996. Canadian Rockburst Support
Handbook. Geomechanics Research Centre, Laurentian University, Sudbury,
Canada, p. 314.
Khademian, Z., Poeck, E., Garvey, R., Ozbay, U., 2016. Studies of seismicity generated
by unstable failures around circular excavations. In: 50th US Rock Mechanics/
Geomechanics Symposium. American Rock Mechanics Association (ARMA).
Kidybin  ski, A., 1981. Bursting liability indices of coal. Int. J. Rock Mech. Min. Sci.
Geomech. Abstr. 18 (4), 295e304.
Kusui, A., Villaescusa, E., Funatsu, T., 2016. Mechanical behaviour of scaled-down
unsupported tunnel walls in hard rock under high stress. Tunn. Undergr.
Space Technol. 60, 30e40.
Lemos, J., Hart, R., Lorig, L., 1987. Dynamic analysis of discontinua using the distinct
element method. In: 6th ISRM Congress. A.A. Balkema, pp. 1079e1084.
Leveille, P., Sepehri, M., Apel, D.B., 2017. Rockbursting potential of kimberlite: a case
study of Diavik diamond mine. Rock Mech. Rock Eng. 50 (12), 3223e3231.
Li, D.J., Qi, H., Feng, J.L., 2015. Granite burst under loading conditions and its char-
acteristics of energy dissipation: implementation by FDEM method and labo-
 J. Wang et al. / Journal of Rock Mechanics and Geotechnical Engineering 13 (2021) 457e478
ratory experiments. In: 13th ISRM International Congress of Rock Mechanics.
International Society for Rock Mechanics and Rock Engineering (ISRM).
Li, G., Zhou, C., Zhang, Y., Xu, H., Gao, Y., Zhang, R., 2013. Review of rockburst in
underground engineering. Adv. Sci. Technol. Water Resour. 33 (3), 77e83 (in
Chinese).
Li, Q., Chen, L., Sui, Z., Wang, L., Dong, J., 2019. Dynamic analysis and criterion
evaluation on rockburst considering the fractured dissipative energy. Adv.
Mech. Eng. 11 (3) https://doi.org/10.1177/1687814019825906.
Li, X., Li, C., Cao, W., Tao, M., 2018. Dynamic stress concentration and energy evo-
lution of deep-buried tunnels under blasting loads. Int. J. Rock Mech. Min. Sci.
104, 131e146.
Li, X.B., Chen, S.R., Gu, D.S., 1994. Dynamic strength of rock under impulse loads
with different stress waveforms and durations. J. Cent.-South Inst. Min. Metall.
25 (3), 301e304 (in Chinese).
Li, X.B., Gu, D.S., Li, H.H., 1993. On the reasonable loading stress waveforms deter-
mined by dynamic stress-strain curves of rocks by SHPB. Explos. Shock Waves
13 (2), 125e130 (in Chinese).
Li, Z.X., Pan, Y.S., Ti, Z.Y., Zhang, Z.H., 2005. Numerical simulation of the high-
pressure water injection into coal seam in Muchengjian coal mine. Chin. J.
Rock Mech. Eng. 24 (11), 1895e1899 (in Chinese).
Lightfoot, N., 1993. The use of numerical modelling in rockburst control. In:
Young, R.P. (Ed.), Rockbursts and seismicity in mines 93: proceedings of the 3rd
international symposium. CRC Press, p. 355.
Liu, E., Huang, R., He, S., 2012. Effects of frequency on the dynamic properties of
intact rock samples subjected to cyclic loading under confining pressure con-
ditions. Rock Mech. Rock Eng. 45 (1), 89e102.
Liu, X., Xu, G., Zhang, C., Kong, B., Qian, J., Zhu, D., Wei, M., 2017. Time effect of water
injection on the mechanical properties of coal and its application in rockburst
prevention in mining. Energies 10 (11), 1783.
Lv, X., Pan, Y., 2010. Numerical analysis on energy-absorption coupling support in
rockburst roadway. Phys. Numer. Simulat. Geotech. Eng. 1, 70.
Ma, C., Li, T., Zhang, H., Jiang, Y., Song, T., 2019. A method for numerical simulation
based on microseismic information and the interpretation of hard rock fracture.
J. Appl. Geophys. 164, 214e224.
Ma, Y., Liu, C., Wu, F., Li, X., 2018. Rockburst characteristics and mechanisms during
steeply inclined thin veins mining: a case study in Zhazixi Antimony Mine,
China. Shock Vib. 2018, 3786047. https://doi.org/10.1155/2018/3786047.
Mahabadi, O.K., Lisjak, A., Munjiza, A., Grasselli, G., Y-Geo, 2012. New combined
finite-discrete element numerical code for geomechanical applications. Int. J.
GeoMech. 12 (6), 676e688.
Manouchehrian, A., Cai, M., 2018. Numerical modeling of rockburst near fault zones
in deep tunnels. Tunn. Undergr. Space Technol. 80, 164e180.
Manouchehrian, S.M.A., 2016. Numerical Modeling of Unstable Rock Failure. Lau-
rentian University, Sudbury, Canada. PhD Thesis.
Marambio, E., Vallejos, J.A., Burgos, L., Gonzalez, C., Castro, L., Saure, J.P., Urzua, J.,
2018. Numerical modelling of dynamic testing for rock reinforcement used in
underground excavations. In: Potvin, Y., Jakubec, J. (Eds.), Proceedings of the 4th
International Symposium on Block and Sublevel Caving. Australian Centre for
Geomechanics, Perth, Australia, pp. 767e780.
Masny, W., Prusek, S., Mutke, G., 2017. Numerical modeling of the dynamic load
changes exerted on the support in the stress concentration zones. Procedia Eng.
191, 894e899.
Mazaira, A., Konicek, P., 2015. Intense rockburst impacts in deep underground
construction and their prevention. Can. Geotech. J. 52 (10), 1426e1439.
Miranda, E.E., 1972. Deformation and Fracture of Concrete Under Uniaxial Impact
Loading. University of Missouri Rolla, Rolla, USA. PhD Thesis.
Mitri, H., Tang, B., Simon, R., 1999. FE modelling of mining-induced energy release
and storage rates. J. S. Afr. Inst. Min. Metall 99 (2), 103e110.
Mitri, H.S., 2000. Practitioner’s Guide to Destress Blasting in Hard Rock Mines.
McGill University, Montréal, Canada. Technical Report.
Mortazavi, A., Alavi, F.T., 2013. A numerical study of the behavior of fully grouted
rockbolts under dynamic loading. Soil Dynam. Earthq. Eng. 54, 66e72.
Munjiza, A., Bangash, T., John, N.W.M., 2004. The combined finite-discrete element
method for structural failure and collapse. Eng. Fract. Mech. 71 (4e6), 469e483.
Naji, A.M., Rehman, H., Emad, M.Z., Yoo, H., 2018. Impact of shear zone on rockburst
in the deep neelum-jehlum hydropower tunnel: a numerical modeling
approach. Energies 11 (8), 1935.
Napier, J.A., Stephansen, S.J., 1987. Analysis of deep-level mine design problems
using the MINSIM-D boundary element program. In: Proceedings of the 20th
International Symposium on the Application of Computers and Mathematics in
the Mineral Industries (APCOM’87), vol. 1: Mining. The Southern African
Institute of Mining and Metallurgy (SAIMM), Johannesburg, South Africa, pp. 3e
19.
Nie, W., Zhao, Z.Y., Ning, Y.J., Guo, W., 2014. Numerical studies on rockbolts
mechanism using 2D discontinuous deformation analysis. Tunn. Undergr. Space
Technol. 41, 223e233.
Nikoli c, M., Roje-Bonacci, T., Ibrahimbegovi c, A., 2016. Overview of the numerical
methods for the modelling of rock mechanics problems. Tehni cki Vjesnik 23 (2),
627e637.
Ortlepp, W.D., Stacey, T.R., 1994. Rockburst mechanisms in tunnels and shafts. Tunn.
Undergr. Space Technol. 9 (1), 59e65.
Ortlepp, W.D., 1992. The design of support for the containment of rockburst damage
in tunnels e an engineering approach. In: Kaiser, P.K., McCreath, D.R. (Eds.),
Rock Support in Mining and Underground Construction. A.A. Balkema, Rotter-
dam, Netherlands, pp. 593e609.
477
Palei, S.K., Das, S.K., 2009. Logistic regression model for prediction of roof fall risks
in bord and pillar workings in coal mines: an approach. Saf. Sci. 47 (1), 88e96.
Peng, Z., Wang, Y.H., Li, T.J., 1996. Griffith theory and rock burst of criterion. Chin. J.
Rock Mech. Eng. 15 (Supp.1), 491e495 (in Chinese).
Pu, Y., Apel, D.B., Hall, R., 2020. Using machine learning approach for microseismic
events recognition in underground excavations: comparison of ten frequently-
used models. Eng. Geol. 268, 105519. https://doi.org/10.1016/
j.enggeo.2020.105519.
Pu, Y., Apel, D.B., Liu, V., Mitri, H., 2019. Machine learning methods for rockburst
prediction-state-of-the-art review. Int. J. Min. Sci. Technol. 29 (4), 565e570.
Qian, Q.H., 2014. Definition, mechanism, classification and quantitative forecast
model for rockburst and pressure bump. Rock Soil Mech. 35 (1), 1e6 (in
Chinese).
Qiu, S., Feng, X., Zhang, C., Xiang, T., 2014. Estimation of rockburst wall-rock velocity
invoked by slab flexure sources in deep tunnels. Can. Geotech. J. 51 (5), 520e
539.
Ranjith, P.G., Zhao, J., Ju, M., De Silva, R.V., Rathnaweera, T.D., Bandara, A.K., 2017.
Opportunities and challenges in deep mining: a brief review. Engineering 3 (4),
546e551.
Russenes, B.F., 1974. Analysis of Rock Spalling for Tunnels in Steep Valley Sides. MSc
Thesis. Norwegian Institute of Technology, Trondheim, Norway (in Norwegian).
Ryder, J.A., 1988. Excess shear stress in the assessment of geologically hazardous
situations. J. S. Afr. Inst. Min. Metall 88 (1), 27e39.
Saeb, S., Amadei, B., 1992. Modelling rock joints under shear and normal loading.
Int. J. Rock Mech. Min. Sci. Geomech. Abstr. 29, 267e278.
Sainoki, A., Emad, M.Z., Mitri, H.S., 2017. Study on the efficiency of destress blasting
in deep mine drift development. Can. Geotech. J. 54 (4), 518e528.
Sainoki, A., Mitri, H.S., 2014a. Dynamic modelling of fault-slip with Barton’s shear
strength model. Int. J. Rock Mech. Min. Sci. 67, 155e163.
Sainoki, A., Mitri, H.S., 2014b. Simulating intense shock pulses due to asperities
during fault-slip. J. Appl. Geophys. 103, 71e81.
Sainoki, A., Mitri, H.S., 2015. Evaluation of fault-slip potential due to shearing of
fault asperities. Can. Geotech. J. 52 (10), 1417e1425.
Sainoki, A., 2014. Dynamic Modelling of Mining-Induced Fault-Slip. McGill Uni-
versity, Montreal, Canada. PhD Thesis.
Salamon, M.D., 1993. Keynote address: some applications of geomechanical
modelling in rockburst and related research. In: Proceedings of the 3rd Inter-
national Symposium on Rockbursts and Seismicity in Mines. A.A. Balkema,
Rotterdam, Netherlands, pp. 297e309.
Salamon, M.D.G., 1964. Elastic analysis of displacements and stresses induced by the
mining of seam or reef deposits e Part II: practical methods of determining
displacement, strain and stress components from a given mining geometry. J. S.
Afr. Inst. Min. Metall 64 (6), 197e218.
Salamon, M.D.G., 1984. Energy considerations in rock mechanics, fundamental re-
sults. J. S. Afr. Inst. Min. Metall 84 (8), 233e246.
Sengani, F., 2018. The performance of bolt-reinforced and shotcreted in-stope pillar
in a rockburst prone areas. Int. J. Min. Geo Eng. 52 (2), 105e117.
Sepehri, M., Apel, D.B., Adeeb, S., Leveille, P., Hall, R.A., 2020. Evaluation of mining-
induced energy and rockburst prediction at a diamond mine in Canada using a
full 3D elastoplastic finite element model. Eng. Geol. 266, 105457. https://
doi.org/10.1016/j.enggeo.2019.105457.
Sepehri, M., 2016. Finite Element Analysis Model for Determination of in-situ and
Mining Induced Stresses as a Function of Two Different Mining Methods used at
Diavik Diamond Mine. University of Alberta, Edmonton, Canada. PhD Thesis.
Shi, G., 1991. Manifold method of material analysis. In: Transaction of the 9th Army
Conference on Applied Mathematics and Computing. Report No. 92-1. US Army
Research office, Minneapolis, USA, pp. 57e76.
Shi, X., Jing, H., Yin, Q., Zhao, Z., Han, G., Gao, Y., 2020. Investigation on physical and
mechanical properties of bedded sandstone after high-temperature exposure.
Bull. Eng. Geol. Environ. 79, 2591e2606.
Simon, R., 1999. Analysis of Fault-Slip Mechanisms in Hard Rock Mining. McGill
University, Montreal, Canada. PhD Thesis.
Song, D., Wang, E., Liu, Z., Liu, X., Shen, R., 2014. Numerical simulation of rock-burst
relief and prevention by water-jet cutting. Int. J. Rock Mech. Min. Sci. 70, 318e
331.
Stacey, T.R., 2013. Dynamic rock failure and its containment-a Gordian knot design
problem. In: Zhao, J., Li, C.L. (Eds.), Rock Dynamics and Applications e State of
the Art: Proceedings of the 1st International Conference on Rock Dynamics and
Applications. CRC Press, pp. 57e70.
Stead, D., Coggan, J.S., Eberhardt, E., 2004. Realistic simulation of rock slope failure
mechanisms: the need to incorporate principles of fracture mechanics. Int. J.
Rock Mech. Min. Sci. 41 (Supp.1), 563e568.
Sun, J.S., Zhu, Q.H., Lu, W.B., 2007. Numerical simulation of rock burst in circular
tunnels under unloading conditions. J. China Univ. Min. Technol. 17 (4), 552e
556.
Tang, B., Mitri, H., 2001. Numerical modelling of rock preconditioning by destress
blasting. Proc. Inst. Civ. Eng. Ground Improv. 5 (2), 57e67.
Tao, Z.Y., 1988. Support design of tunnels subjected to rockbursting. In: ISRM In-
ternational Symposium. A.A. Balkema, pp. 407e411.
Turchaninov, I.A., Markov, G.A., Gzovsky, M.V., Kazikayev, D.M., Frenze, U.K.,
Batugin, S.A., Chabdarova, U.I., et al., 1972. State of stress in the upper part of the
earth’s crust based on direct measurements in mines and on tectonophysical
and seismological studies. Phys. Earth Planet. In. 6 (4), 229e234.
Varden, R., Lachenicht, R., Player, J., Thompson, A., Villaescusa, E., 2008. Develop-
ment and implementation of the Garford dynamic bolt at the kanowna belle
478 J. Wang et al. / Journal of Rock Mechanics and Geotechnical Engineering 13 (2021) 457e478
mine. In: Proceedings of the 10th Underground Operators’ Conference 2008.
The Australasian Institute of Mining and Metallurgy, Melbourne, Australia,
pp. 95e102.
Vazaios, I., Diederichs, M.S., Vlachopoulos, N., 2019. Assessment of strain bursting in
deep tunnelling by using the finite-discrete element method. J. Rock Mech.
Geotech. Eng. 11 (1), 12e37.
Vazaios, I., 2018. Modelling of Hard Rockmasses with Non-persistent Joints to
Assess the Stress Induced Damage of Deep Excavations. Queen’s University,
Kinston, Canada. PhD Thesis.
Vennes, I., Mitri, H., 2017. Geomechanical effects of stress shadow created by large-
scale destress blasting. J. Rock Mech. Geotech. Eng. 9 (6), 1085e1093.
Wagner, H., 2019. Deep mining: a rock engineering challenge. Rock Mech. Rock Eng.
52 (5), 1417e1446.
Wang, P., Xu, J., Liu, S., Wang, H., Liu, S., 2016. Static and dynamic mechanical
properties of sedimentary rock after freeze-thaw or thermal shock weathering.
Eng. Geol. 210, 148e157.
Wang, S.Y., Lam, K.C., Au, S.K., Tang, C.A., Zhu, W.C., Yang, T.H., 2006. Analytical and
numerical study on the pillar rockbursts mechanism. Rock Mech. Rock Eng. 39
(5), 445e467.
Wang, S.Y., Tang, C.A., Xu, T., Tan, Z.H., 2003. Numerical simulation on acoustic
emission during pillar rock burst. Chin. J. Nonferrous Metals 13 (3), 754e759 (in
Chinese).
Wang, X., Cai, M., 2017. Coupled numerical analysis of ground motion near exca-
vation boundaries in underground mines. Rock Soil Mech. 38 (11), 3347e3354,
3370 (in Chinese).
Wang, Z., Dou, L., Li, J., Kang, K., Feng, L., 2017. Numerical investigation of damage
risks of roadway surrounding rocks under oblique incident dynamic loads.
Shock Vib. 2017, 6298372. https://doi.org/10.1155/2017/6298372.
Wattimena, R.K., Sirait, B., Widodo, N.P., Matsui, K., 2012. Evaluation of rockburst
potential in a cut-and-fill mine using energy balance. Int. J. JCRM 8 (1), 19e23.
Wei, L., Hudson, J.A., 1998. A hybrid discrete-continuum approach to model hydro-
mechanical behaviour of jointed rocks. Eng. Geol. 49 (3e4), 317e325.
Wei, L., 1992. Numerical Studies of the Hydromechanical Behaviour of Jointed
Rocks. Imperial College of Science and Technology, London, UK. PhD Thesis.
Whyatt, J.K., Board, M.P., 1991. Numerical Exploration of Shear-Fracture-Related
Rock Bursts using a Strain-Softening Constitutive Law. US Department of the
Interior, Bureau of Mines.
Wijk, G., Skogberg, B., 1982. The Swellex rock bolting system. In: Rock breaking and
mechanical excavation: proceedings of the 14th Canadian Rock Mechanics
Symposium. Vancouver, Canada, p. 20.
Wiles, T., 2002. Loading system stiffness e a parameter to evaluate rockburst po-
tential. In: Proceedings of the 1st International Seminar on Deep and High
Stress Mining. Australian Centre for Geomechanics, Perth, Australia.
Wu, R., Oldsen, J., Campoli, A., 2011. Application of yield-lok bolt for bursting and
convergence grounds in mines. In: Proceedings of the 30th International Con-
ference on Ground Control in Mining. Morgantown, USA.
Xie, H.P., Gao, F., Ju, Y., 2015. Research and development of rock mechanics in
deep ground engineering. Chin. J. Rock Mech. Eng. 34 (11), 2161e2178 (in
Chinese).
Yang, F., Zhou, H., Lu, J., Zhang, C., Hu, D., 2015. An energy criterion in process of
rockburst. Chin. J. Rock Mech. Eng. 34 (Supp.1), 2706e2714 (in Chinese).
Yang, J., Chen, W., Tan, X., Yang, D., 2018. Analytical estimation of stress distribution
in interbedded layers and its implication to rockburst in strong layer. Tunn.
Undergr. Space Technol. 81, 289e295.
Yokota, Y., Zhao, Z., Nie, W., Date, K., Iwano, K., Koizumi, Y., Okada, Y., 2019.
Development of a new deformation-controlled rockbolt: numerical modelling
and laboratory verification. In: Proceedings of the 9th International Symposium
on Ground Support in Mining and Underground Construction. Australian Centre
for Geomechanics, Perth, Australia, pp. 545e556.
Yu, W., Wang, W., Chen, X., Du, S., 2015. Field investigations of high stress soft
surrounding rocks and deformation control. Chin. J. Rock Mech. Geotech. Eng. 7
(4), 421e433.
Zhang, C., Feng, X.T., Zhou, H., Qiu, S., Wu, W., 2012b. A top pilot tunnel pre-
conditioning method for the prevention of extremely intense rockbursts in
deep tunnels excavated by TBMs. Rock Mech. Rock Eng. 45 (3), 289e309.
Zhang, C., Feng, X.T., Zhou, H., Qiu, S., Wu, W., 2012a. Case histories of four
extremely intense rockbursts in deep tunnels. Rock Mech. Rock Eng. 45 (3),
275e288.
Zhang, C.Q., Zhou, H., Feng, X.T., 2011. An index for estimating the stability of brittle
surrounding rock mass: FAI and its engineering application. Rock Mech. Rock
Eng. 44 (4), 401.
Zhang, H., Chen, L., Chen, S., Sun, J., Yang, J., 2018. The spatiotemporal distribution
law of microseismic events and rockburst characteristics of the deeply buried
tunnel group. Energies 11 (12), 3257.
Zhang, J., Fu, B., 2008. Rockburst and its criteria and control. Chin. J. Rock Mech. Eng.
27 (10), 2034e2042 (in Chinese).
Zhou, G., Xu, M., Fan, T., 2018b. Numerical simulation of combined water injection in
deep coal seam and its field application for dust suppression and de-stressing: a
case study at Dongtan Coal Mine, China. Geotech. Geol. Eng. 36 (1), 283e291.
Zhou, J., Li, X., Mitri, H.S., 2018a. Evaluation method of rockburst: state-of-the-art
literature review. Tunn. Undergr. Space Technol. 81, 632e659.
Zhu, Q.H., Lu, W.B., Sun, J.S., Luo, Y., Chen, M., 2009. Prevention of rockburst by guide
holes based on numerical simulations. Min. Sci. Technol. 19 (3), 346e351.
Zhu, W.C., Li, Z.H., Zhu, L., Tang, C.A., 2010. Numerical simulation on rockburst of
underground opening triggered by dynamic disturbance. Tunn. Undergr. Space
Technol. 25 (5), 587e599.
Zienkiewicz, O.C., Kelly, D.W., Bettess, P., 1977. The coupling of the finite element
method and boundary solution procedures. Int. J. Numer. Methods Eng. 11 (2),
355e375.
Zubelewicz, A., Mroz, Z., 1983. Numerical simulation of rock burst processes
treated as problems of dynamic instability. Rock Mech. Rock Eng. 16 (4), 253e
274.