Tunnelling and Underground Space Technology 70 (2017) 409–421

Contents lists available at ScienceDirect

Tunnelling and Underground Space Technology

journal homepage: www.elsevier.com/locate/tust

Propagation and prediction of blasting vibration on slope in an open pit MARK

during underground mining
⁎
Nan Jianga, Chuanbo Zhoua, , Shiwei Lub, Zhen Zhanga
a
Faculty of Engineering, China University of Geosciences, No. 388, Lumo Road, Wuhan 430074, Hubei, China
b
School of Urban Construction, Yangtze University, No. 1, Nanhuan Road, Jingzhou 434023, China

A R T I C L E I N F O A B S T R A C T

Keywords: In order to assess the influence of underground mining blasting on the stability of open pit slopes and control the
Underground mining potential risk, it is crucial to investigate the propagation and attenuation of blasting vibration in rock slopes. In
Rock slope this paper, we aimed to investigate the attenuation of blasting vibration on open pit slopes subjected to un-
Peak particle velocity derground mining activities using Daye Iron Mine as an example. To this end, we first analyzed the char-
Propagation
acteristics of blast loadings using the dynamic finite element method. We then established a three dimensional
(3D) numerical model for open pit subjected to underground mining and proved its reliability using the field
monitoring data. Next, we calculated and analyzed the distribution characteristics of peak particle velocity (PPV)
by inputting the obtained blast loadings into the numerical model and discussed the impacts of blasting vibration
on open pit slopes subjected to underground mining. To better and more conveniently assess and predict the
impacts of blasting vibration, we further established a mathematical model to describe the attenuation of PPV on
open pit slopes subjected to underground mining blasting through theoretical analysis and proposed a PPV
predicting model for slopes in Daye Iron Mine based on the numerical simulations of underground mining
blasting at different elevations.

1. Introduction
With the continuous surface mining of mineral resources, open pit
slopes become higher and steeper, potentially leading to instability. In
addition, with depletion of near-surface mineral resources, surface
mining becomes more costly. Obviously, to ensure mining safety and
efficiency, underground mining is an alternative choice. However, in
the process of transition from open pit to underground mining, blast-
induced seismic wave propagates in open pit slopes. Such a disturbance
is often the critical factor triggering the instability of high and steep
open pit slopes (Ak et al., 2009; Deb and Jha, 2010). In order to ef-
fectively assess the seismic effect of blasting during underground
mining on open pit slopes and control the potential risk of instability, it
is crucial to investigate the impacts of blasting vibration on slope and
understand the propagation and attenuation of blasting vibration in
rock slope.
At present, although many scholars have extensively studied the
impact of blasting vibration on rock slopes, most of them focused on the
influence of open pit excavation and mining blasting on rock slopes
(Guo et al., 2004; Li and Zhang, 2007; Afeni and Osasan, 2009;
Baczynski, 2010; Deb et al., 2011; Choi et al., 2013; Hu et al., 2014;
Saadat et al., 2014). For example, Kesinal et al. discussed the influence
of blast-induced acceleration on slope stability at a limestone quarry by
field monitoring and theoretical analysis (Kesimal et al., 2008); Thote
and Venkat experimentally investigated the slope damages due to blast
loadings of different frequencies and vibration velocities (Thote and
Ramana, 2013). In recent years, with the development of computer
technology, numerical simulations are frequently adopted to study the
influence of blast on rock mass and rock slopes (Zhong et al., 2010;
Sangroya and Choudhury, 2013; Zhang et al., 2011; Deng et al., 2015).
Yan et al. analyzed the variation law of slope stability coefficient in-
fluenced by blasting vibration using the software GEO-SLOPE (Yan
et al., 2014). Yang et al. adopted the software FLAC to analyze the
dynamic stability of opencast slopes affected by blast loading (Yang
et al., 2011). Zheng et al. and Azizabadi et al. analyzed the vibration of
slope subjected to excavation blasting using the software UDEC (Zheng
et al., 2014; Azizabadi et al., 2014). Liu et al. conducted discontinuous
analysis of the stability of blast-induced slopes based on the DDA nu-
merical method (Liu et al., 2010). Li et al. adopted the software FINAL
to analyze the dynamic response of prestressed anchorcables subjected
to blast loading on slope (Li et al., 2007). Moreover, many scholars
utilized ANSYS/LSDYNA, a nonlinear finite element software, to

⁎
Corresponding author.
E-mail addresses: happyjohn@foxmail.com (N. Jiang), cbzhou@cug.edu.cn (C. Zhou), lushiwei364@163.com (S. Lu), zhangzhen9168@163.com (Z. Zhang).

http://dx.doi.org/10.1016/j.tust.2017.09.005
Received 17 February 2017; Received in revised form 24 August 2017; Accepted 5 September 2017
0886-7798/ © 2017 Elsevier Ltd. All rights reserved.
N. Jiang et al.
analyze the effects and characteristics of blast loading induced by de-
tonation of explosives (Li et al., 2011; Xie et al., 2016; Yi et al., 2017;
Xie et al., 2017).
In this paper, we focused on the effects of blast during underground
mining on the open pit slopes of Daye Iron Mine in China (Section 2).
First, we calculated the characteristics of blast loading in underground
mining by employing the dynamic finite element software ANSYS/LS-
DYNA (Section 3). Second, we established a 3D numerical model of
open pit and verified its reliability using the field monitoring data
(Section 4). Third, we used the model to analyze the impacts of blast
vibration on pit slopes based on the distribution characteristics of peak
particle velocity (PPV) on slopes (Section 5). At last, based on these
primary results, we established a mathematical model to describe the
attenuation of PPV on slopes subjected to underground mining blast
through theoretical analysis and proposed a PPV predicting model for
the open pit slopes of Daye Iron Mine during underground mining
(Section 6).
2. General information of the open pit to underground mining
mine
2.1. General information of the open pit slopes
Daye Iron Mine is located in Tieshan District, Huangshi City, Hubei
Province, PR China, about 90 km west to Wuhan City, 25 km east to the
downtown of Huangshi City and 15 km southeast to Daye City, as
shown in Fig. 1. Throughout 100 years of mining activities, a deep open
pit of 2400 m long longitudinally and 1000 m wide latitudinally has
been formed, as shown in Fig. 2. The east open pit of Daye Iron Mine
was selected as the study area, as shown in Fig. 3. Its north slope is
170 m–270 m above the mean sea level (AMSL) and composed of
diorite, and its south slope is 86 m–200 m AMSL and composed of
Fig. 1. Location of Daye Iron Mine in Hubei Province, China.
410
Tunnelling and Underground Space Technology 70 (2017) 409–421
marble. The slopes’ angle is 38°–43°, and the slopes’ height is
230 m–430 m. Magnetite distributes in the contact zone between
diorite and marble. The pit bottom is filled to ± 0 m AMSL with cru-
shed stones. There are many natural faults in the study area. Among
them, Fault F9 and Fault F25 have great effects on the slope stability. A-
slope is located in the east of the north slope and its stability is mainly
controlled by Fault F9. In history, A-slope was locally destabilized for
many times and has become a potential threat. Meanwhile, under the
impacts of underground blasting operation and weathering, the partial
collapse of surficial rock mass of the open pit slope always happens,
destroying the pit slopes’ benches (Yao, 2008).
2.2. Underground mining method
Non-pillar sublevel caving method is adopted for the underground
mining of Daye Iron Mine. A total of 14 boreholes with diameter of
80mm and various length ranging from 7 m to 17.4 m were distributed
in a fan-shape with angle between adjacent boreholes varying between
6° and 9° for blasting (Fig. 4). For each mining blast, the 14 boreholes
were detonated simultaneously to create a tunnel of 3.6 m wide and
3 m high with an arch shaped wall using the No.2 rock emulsion ex-
plosive of China.
3. Characteristics of blast loading in underground mining
3.1. Numerical model and parameters
In order to obtain the blast loading in underground mining, a 3D
numerical model of 30 m high, 16.4 m wide and 4.8 m long in the axial
direction of the tunnel was established based on the distribution of
boreholes in the tunnel (Fig. 5). These boreholes were distributed on
the top of the tunnel within a fan-shaped vertical plane, 1.6 m away
N. Jiang et al. Tunnelling and Underground Space Technology 70 (2017) 409–421

Fig. 2. Actual photo of open pit in Daye Iron Mine.

from the back side of the model and 3.2 m away from the front side
(Fig. 5a). The back side of the model was a free surface and used as the
mining working plane and the other sides were applied with non-re-
flecting boundaries (Fig. 5b). A 3D-solid-164 element was always used
to simulate 3D explicit structural solid. The element contained 8 nodes,
each has degrees of freedom on displacement, velocity and acceleration
in x-, y- and z-directions. Meanwhile, the element can only be used in
dynamic explicit analysis. Overall, in the numerical simulation, the 3D
explicit simulation was performed using the software ANSYS/LS-DYNA
and the 3D-solid-164 element as the element type.
The ore body in the region of interest was simplified as a homo-
geneous rock material, and cracks and structural surface in ore body
were not taken into consideration. The constitutive relationship of rock
under blast loading in ANSYS/LSDYNA was numerically simulated
using the elastic-plastic constitutive model of rock and the
MAT_PLASTIC_KINEMATIC tab (Hallquist, 2007). This model was
suited to model isotropic and kinematic hardening plasticity with the
option of including rate effects. The physical and mechanical para-
meters of the ore body were determined based on laboratory tests
(Miao, 2013) and are listed in Table 1.
During the blasting operation, the detonation of explosives is a rapid
chemical reaction process. Many different equation of state (EOS) have
been proposed for the description of detonation products. In ANSYS/
LSDYNA, the Jones–Wilkins–Lee (JWL) EOS is a high energy combus-
tion model, and can be used to describe the chemical reaction process.
However, JWL EOS has capacity to predict a large range of pressure

Fig. 3. Top view of east open pit in Daye Iron Mine.

411
N. Jiang et al. Tunnelling and Underground Space Technology 70 (2017) 409–421

Fig. 4. Schematic diagram of the cross-section view of boreholes and

the underground tunnel.

Fig. 5. Numerical model for calculation of blast loading.

412
N. Jiang et al. Tunnelling and Underground Space Technology 70 (2017) 409–421

Table 1 8 ms after successive detonation and decreased dramatically after-

Physical and mechanical parameters of magnetite. wards. After 16 ms, the pressure vibration was attenuated to a level-off
value close to zero. On the other hand, the size of fracture zone was
Mass density Young’s Poisson’s ratio Yield Tangent
(g/cm3) modulus (GPa) strength modulus (GPa) about 10–15 times of the boreholes’ diameter according to the relevant
(MPa) research (Hustrulid, 1999; Esen et al., 2003). In our model, the ele-
ments on the boundaries were located outside the fracture zone.
4.12 14.180 0.27 100 1.50
Therefore, the pressure on the boundaries can be regarded as the blast
induced seismic wave when analyzing the seismic effect far from the
Table 2
explosion source (Lu et al., 2011; Chen et al., 2014). In this study, the
Material model and EOS parameters of the No.2 rock emulsion explosive. time history of pressures on the boundaries was considered as the input
blast loading of underground mining in numerical simulations in the
Density Detonation A (GPa) B (GPa) R1 R2 ω E (GPa) next section.
(g/cm3) velocity (cm·μs−1)

1.09 0.4 214.4 18.2 4.2 0.9 0.15 4.192 4. Numerical modeling of mine area and verification

4.1. Numerical modeling and parameter selection

caused by explosion, in which its parameters can be obtained experi-
mentally (Yi et al., 2017). In this study JWL EOS was chosen to describe
the explosion, and is expressed as follows:
ω ⎞ −R1 V ω ⎞ −R2 V ωE
P = A ⎛1−⎜ e ⎟ + B ⎛1− ⎜ e + ⎟
⎝ R1 V ⎠ ⎝ R2 V ⎠ V (1)
where P and V are the relative specific volume and the pressure of the
detonation products, respectively, E is the internal energy per unit
volume and A, B, R1, R2 and ω are material constants.
In the current study, No.2 rock emulsion explosive was chosen, and
450 kg explosive was charged at the bottom of each borehole. However,
due to the lack of measurement instruments, these JWL EOS parameters
except the density were all referred to Li et al. (2011) and Xie et al.
(2016). The density of explosive was determined by field test. All the
EOS parameters are listed in Table 2.
3.2. Characteristics of blast loading
In order to characterize the blast loading, the pressures of the re-
presentative elements on non-reflecting boundaries (Fig. 5b) with time
are recorded during the simulation. Because the model is symmetrical
on the Y-Z plane, only elements on the right hand side are discussed.
Fig. 6 shows the time history of pressure of the representative elements.
Form the figure it can be found that the pressure of all re-
presentative elements peaked from 1.9 MPa to 43.05 MPa at 2 ms to
The rock volume influenced by the underground blasting, which is
about 2.5 times of the underground stope, was fully considered in our
simulation and both the north and south slopes were fully included.
The research area of the 3D numerical model is shown in Fig. 3. The
model was ore body-centered and 900 m, 800 m and 680 m in x- y- and
z-directions, respectively (Fig. 7). In the model, Fault F25 was located
in the south slope and Fault F9 was located in the north slope. The top
surface was a free surface and all other eternal surfaces were applied
with non-reflecting boundary.
Based on the inversion analysis of physical parameters (Miao,
2013), the physical and mechanical parameters of different parts in the
numerical model are listed in Table 3, and the magnetite was in ac-
cordance with that in Section 3.1.
In the numerical simulation, the position of underground mining
blasting was selected based on the on-site blasting operation on May 22,
2013 and located at −108 m AMSL in the middle of the No.17 tunnel,
which was about 46 m long. As described in Section 3.1, a cube had
been excavated previously, in which the blast loading was applied to a
4.8 m long region. Fig. 8 shows the planes to apply the blast loading.
4.2. Numerical calculations and verification
To verify the numerical simulations, 4 filed monitoring points were
set up at the foot of pit slopes located on the stable benches, as shown in
Figs. 2 and 3, to record blast vibration using the TC-4850 self-recording

Fig. 6. Time history of pressure of the representative elements.

413
N. Jiang et al. Tunnelling and Underground Space Technology 70 (2017) 409–421

Fig. 7. Complex numerical model of Daye Iron Mine.

Table 3 ranging from 50.63 Hz to 73.91 Hz. This phenomenon is probably due
Physical and mechanical parameters of different part in numerical model. to the influence of possible cracks in rock mass on the attenuation of
PPVs (Deng et al., 2012; Zhu et al., 2013; Deng et al., 2014). The above
Type Mass Young’s Poisson’s ratio Yield Tangent
density modulus strength modulus
analysis shows that the numerical model and the selected parameters
(g/cm3) (GPa) (MPa) (GPa) are reasonable and the numerical simulation can reproduce the dis-
tribution of PPVs on open pit slopes due to underground mining blast.
Diorite 2.71 18.370 0.23 75 1.00
Marble 2.71 16.220 0.23 75 1.00
Backfilling 1.94 2.687 0.20 25 0.50 5. Distribution of blasting vibration on pit slopes
Fault 1.77 2.492 0.20 50 0.50
5.1. Distribution of PPVs on the south slope

instrument developed and manufactured by Zhongke (Chendu) Instru-
ments Company Limited. The instrument contains a trivector sensor
with measuring vibration ranging from 0.001 cm/s to 35.4 cm/s and
measuring frequency ranging from 0 to 1000 Hz. Fig. 9 shows the set-
ting of field monitoring points. Corresponding to these field monitoring
points, 4 nodes were selected at the same positions in numerical model
to record the blast vibrations, as shown in Fig. 7. Coordinates of these
nodes in numerical model and the corresponding field monitoring
points are listed in Table 4.
For comparison with field monitoring data, vibration velocities at 4
numerical simulation points were retrieved. Among them, the vibration
velocities in x-, y- and z-direction at No.102078 simulation point were
plotted in Fig. 10. It took about 50 ms for blasting induced stress waves
to propagate from explosion source to the position of Node No.102078.
The blast vibrations lasted about 100 ms and reached their peak velo-
cities of 0.936 cm/s, 0.843 cm/s and 0.827 cm/s at x-, y- and z-direc-
tions, respectively, within the time range of 50 ms–75 ms.
Blast vibration velocities in x-, y- and z-directions at #1 field
monitoring point were plotted in Fig. 11. It can be obtained from
Figs. 10 and 11 that the calculated vibration velocities and field mon-
itored vibration velocities showed essentially the same changing ten-
dency, indicating that the numerical simulations could truthfully reflect
the changing characteristics of vibration velocity.
Main frequencies of the simulated data and field monitoring data
were acquired through spectral analysis. Table 5 shows the PPVs and
main frequencies. It can be obtained from Table 5 that PPVs of field
monitoring data were slightly smaller than those of the simulated data.
The main frequencies of the simulated data ranged from 63.69 Hz to
72.53 Hz and were very close to those of the field monitoring data,
The property and geological structure of slope rock mass influence
its dynamic response to blast vibration (Tang, 2007). The north slope is
mainly composed of diorite and the south slope is mainly composed of
marble. The differences in their geometry result in different vibration
patterns between the south and the north slopes. In order to investigate
the vibration pattern, a series of nodes on the south slope and north
slope were selected as the monitoring points. The IDs of those on the
south slope were No.123375, No.123666, No.127838, No.127512,
No.126885 and No.127122 (Fig. 12).
Altitude effect exists when a blast seismic wave propagates in slopes
(Tang, 2007). The simulated PPVs at all monitoring points are shown in
Fig. 13. The maximum PPVs were 0.697 cm/s, 0.588 cm/s and
0.604 cm/s for x-, y- and z-components, respectively, and all occurred
at No.123375 monitoring point. Meanwhile, PPVs in all directions de-
creased with the elevation of monitoring points on the south slope in-
creasing, which was contradict to the phenomenon that PPVs increased
with the elevation increasing should the altitude amplification effect
exert, indicating that the altitude amplification effect was not obvious.
At the same time, the Fast Fourier Transform (FFT) spectrum analysis
on particle velocity curves showed that the main frequency decreased
from 38.75 Hz to 23.14 Hz with the elevation of monitoring points in-
creasing, suggesting that the high-frequency component of blast seismic
waves decayed and separated from the low-frequency component with
the increase of the horizontal distance and elevation difference between
the explosion source and the monitoring point, and the low-frequency
component propagated farther. Based on the above analysis, we found
that the PPVs at all monitoring points on the south slope were small,
showing the maximum of only 0.697 cm/s and the frequencies of
38.75 Hz–23.14 Hz. Previous studies have reported that the natural

414
N. Jiang et al.
frequency of pit slopes was less than 20 Hz in general (Ming et al.,
2012; Xu et al., 2009; Xu and Peng 2006). In this study, the main fre-
quency of blast seismic waves was slightly greater than the natural
frequency of pit slopes, suggesting that the effect of resonance was
weak. To sum up, the impacts of present blasting vibration on the south
slope were limited.
5.2. Distribution of PPVs on the north slope
The IDs of the selected numerical monitoring points on the north
slope were No.30209, No.30713, No.31071, No.31849, No.31899 and
No.28789 (Fig. 12) and their coordinates were (395.31, 479.66, 24.00),
(400.88, 503.76, 46.00), (403.90, 527.39, 70.00), (403.85, 555.09,
100.00), (414.59, 589.15, 130.00) and (430.94, 679.89, 230.00), re-
spectively.
The simulated PPVs of x-, y- and z-components at all monitoring
points are shown in Fig. 14. It can be found from Fig. 11 that the
maximum PPVs of x, y and z-components were 0.747 cm/s, 0.638 cm/s
and 0.704 cm/s, respectively, and occurred at No.30209 monitoring
Fig. 9. The photo of the field monitoring points.
415
Tunnelling and Underground Space Technology 70 (2017) 409–421
Fig. 8. Position of planes to apply blast loading.
point located at the foot of the north slope. The above analysis sug-
gested that PPVs in all directions decreased with the elevation of the
monitoring points on the north slope increasing, and the altitude am-
plification effect was not obvious. In addition, the main frequency of
vibration decreased from 52.53 Hz to 25.89 Hz with the elevation in-
creasing. The results indicated the PPVs on the north slope were small
and the impacts of present underground mining blasting on the north
slope were limited too.
5.3. Distribution of PPVs on A-slope
In Daye Iron Mine, the benches on A-slope have been damaged se-
verely. Therefore, we performed numerical simulation to investigate its
dynamic response and evaluate its stability and selected a series of
numerical monitoring points with IDs of No.104492, No.104176,
No.104255, No.104357, No.104438 and No.104465 and corresponding
coordinates of (774.11, 481.42, 16.10), (759.27, 503.43, 40.19),
(739.65, 530.39, 70.00), (726.70, 552.75, 92.76), (714.08, 596.60,
125.12) and (694.42, 643.00, 168.62), respectively, as shown in Fig. 7.
N. Jiang et al. Tunnelling and Underground Space Technology 70 (2017) 409–421

Table 4 and Suhadolc, 1998; Song et al., 2009; Tang and Li 2011; Chen et al.,
Coordinates of nodes and the corresponding field monitoring points. 2011). Considering the engineering characteristics of open pit subjected
to underground mining, the distance between explosion source and
ID of field monitoring ID of node in numerical Coordinate
point model ground surface alters with mining depth changing, and the buried depth
of explosion source would exert some influence on the propagation of
1 102078 (481.12, 466.12, 0) blasting seismic wave in open pit slopes (Lu, 2005; Wu and Hao, 2005).
2 13486 (425.15, 347.07, 0)
Table 6 summarizes the main factors affecting the propagation of
3 13188 (524.12, 338.31, −12.28)
4 7077 (550.01, 450.42, −12.28) blasting seismic wave in open pit slopes.
PPV in slope rock mass (v) can be written as:
v = Φ(Q,μ,c,ρ,r ,H ,D,a,f ,t ) (2)

where variables Q, r, c are independent and satisfy the following for-

To understand the law of PPV changing with elevation and evaluate
mulas according to Buckingham’s Pi-Theorem in dimension analysis
the influence of blast vibration on A-slope, the x-, y- and z-component
(Langhaar, 1951):
PPVs as well as their resultant PPVs at each monitoring point on A-slope
v μ ρ
were calculated, as shown in Fig. 15. ⎧ π0 = ,π1 = ,π2 = ,
Q α0 r β0 c γ0 Q α1 r β1 c γ1 Q α2 r β2 c γ2
It can be found from Fig. 15 that PPVs decreased with the elevation ⎪ H D a
of monitoring points increasing and the PPVs were greater in z-direc- π3 = ,π4 = ,π5 = ,
⎨ Q α3 r β3 c γ3 Q α 4 r β4 c γ4 Q α5 r β5 c γ5
tion than in other directions. The maximum z-component PPV was only ⎪ f t
π6 = ,π7 =
0.392 cm/s, occurring at No.104492 monitoring point of the lowest ⎩ Q α6 r β6 c γ6 Q α7 r β7 c γ7 (3)
elevation, and the minimum z-component PPV was only 0.082 cm/s,
where πn donates the dimensionless form of the dependent variables,
occurring at No.104465 monitoring point of the highest elevation.
n = 0, 1, 2…, 7 and the exponents αn , βn and γn are dimensionless. πn
Overall, PPVs of A-slope in all directions were attenuated, indicating
can be calculated as follows:
that the altitude amplification effect was not obvious. Compared with
the aforementioned monitoring data, PPVs of A-slope were perceptibly v μ ρ
⎧ π0 = c ,π1 = r ,π2 = Qr −3 ,
small. The underlying reason may be that A-slope was controlled by ⎪ H D a
Fault F9 and part energy was absorbed when blasting seismic waves π3 = r ,π4 = r ,π5 = −1 2 ,
⎨ r c
crossed the fault zone. Consequently, the PPVs on A-slope were ob- ⎪ π = f ,π = t
6 r −1c 7
viously small. Moreover, the FFT spectrum analysis on particle vibra- ⎩ rc−1 (4)
tion velocity curves showed that the main frequency decreased from Substituting Eq. (4) into Eq. (2) finds
40 Hz to 22 Hz with the elevation of monitoring points increasing,
suggesting that PPVs on A-slope were small and the underground v μ ρ H D a f t
= Φ ⎛ , −3 , , , −1 2 , −1 , −1 ⎞
⎜ ⎟

mining blast had little influence on the stability of A-slope. c ⎝ r Qr r r r c r c rc ⎠ (5)

A new dimensionless parameter π8 is given by the combination of

6. Establishment of a PPV predicting model for open pit slopes in π2, π3 and π4, as follows:
Daye Iron Mine β
1 β1 3 ρr 1 β β
π8 = ⎛π23 ⎞ π3β2 π4 3 = ⎛⎜
β ⎞ ⎛H⎞ 2 ⎛D⎞ 3
⎜ ⎟ ⎟
⎝ Q ⎠ ⎝r ⎠ ⎝r⎠
3
6.1. Establishment of a PPV predicting model ⎝ ⎠ (6)

Previous studies have shown that the attenuation of blasting seismic where β1, β2 and β3 are unknown exponents of π2 , π3 and π4 , respec-
wave is affected by explosion source conditions, structural and geolo- tively.
gical characteristics of slope rock mass (e.g. lithology, joint, etc.) as well Assuming that ρ and c are constants, according to Eq. (5), the
β1
H β2 D β3
as vertical and horizontal distances from the explosion source (Marrara functional relation between ν and ( 1
Q1/3r −1 )()()r r
can be written

Fig. 10. Time history of vibration velocity of Node No.102078.

416
N. Jiang et al. Tunnelling and Underground Space Technology 70 (2017) 409–421

Fig. 11. Time history of vibration velocity at 1# field monitoring point.

Table 5 Let k = k1 k2 k3 , thus, the predicting model reflecting PPV on slopes of

PPVs of nodes and the corresponding field monitoring points. open pit subjected to underground mining is expressed as:
β 1
Type ID PPV (cm/s) Main frequency (Hz) 3 Q β β
v = k ⎛⎜ ⎞ ⎛H⎞ 2 ⎛D⎞ 3
⎟
x y z x y z ⎝ r ⎠ ⎝r ⎠ ⎝r⎠ (12)

Numerical simulation 102078 0.936 0.827 0.843 64.74 67.17 70.05 It can be obtained by comparing Eqs. (9) and (12) that when pro-
13486 0.912 0.793 0.810 65.52 65.03 63.69 pagating on an open pit slope, blast vibration is affected by both the
13188 0.649 0.486 0.588 71.30 72.53 66.94 buried depth of explosion source and the elevation difference between
7077 0.684 0.516 0.526 68.15 70.28 69.80
the explosion source and the monitoring point.
Field monitoring 1# 0.731 0.612 0.645 54.05 60.61 66.67
2# 0.712 0.689 0.596 56.34 73.91 50.63 It can be obtained by comparing Eqs. (9) and (12) that when pro-
3# 0.581 0.433 0.564 58.82 58.82 66.67 pagating on an open pit slope, blast vibration is affected by both the
4# 0.664 0.592 0.614 63.33 56.85 65.00 buried depth of explosion source and the elevation difference between
the explosion source and the monitoring point.

as
6.2. Application of the PPV predicting model in Daye Iron Mine
3 Q
⎡
lnv = ⎢α1 + β1ln ⎛⎜ ⎞ ⎤ + ⎡α + β ln ⎛ H ⎞ ⎤ + ⎡α + β ln ⎛ D ⎞ ⎤
⎟
⎥ 2 2 3 3 To further study the dynamic response of pit slope in different
⎣ ⎝ r ⎠⎦ ⎣ ⎝ r ⎠⎦ ⎣ ⎝ r ⎠⎦ (7) mining conditions, the materials and blast loading applying method
described in Section 4.1 are adopted to simulate 5 cases of explosion
( ), then
3Q
Let lnv0 = α1 + β1ln r source at 5 different locations with varying elevations (−84 m, −96 m,
lnv0 = α1 + (β1lnQ)/3−β1lnr (8) −108 m, −120 m, −132 m). PPVs at different positions of the north
slope, south slope and A-slope are extracted. A total of 90 groups of data
where α1, α2 and α3 are unknown coefficients and β1 is attenuation
were obtained and parts of them are listed in Table 7.
index and reflects the influence of the medium. −β1ln r suggests that
Fitting Eq. (12) to the simulated data listed in Table 7 obtained the
PPV decreases with r increasing and α1 + (β1lnQ)/3 reflects the influ-
PPV predicting models for the south slope, north slope and A-slope. To
ence of both medium and mass of explosive.
justify the accuracy of PPV predicting models’ rationality, Eq. (9) was
Let lnk1 = lnα1, then
fit to the simulated data listed in Table 7. The resultant correlation
β1
3 Q coefficients from Eqs. (9) and (12), as listed in Table 8, were compared
v0 = k1 ⎛⎜ ⎞
⎟
to evaluate the accuracy of PPV prediction.
⎝ r ⎠ (9)
The following conclusions can be obtained from Table 8.
Eq. (9) is the classical Sadovski’s formula, in which the influence of
elevation variation is not taken into account (AQSIQ, 2014). Sub- (a) The PPV predicting models are different for different locations due
stituting Eq. (9) into Eq. (7) finds to differences in lithology and geological structures. Therefore, the
blast vibration of different pit slopes should be analyzed respec-
H D
lnv = lnv0 + ⎡α2 + β2ln ⎛ ⎞ ⎤ + ⎡α3 + β3ln ⎛ ⎞ ⎤ tively.
⎣ r
⎝ ⎠⎦ ⎣ ⎝ r ⎠⎦ (10)
(b) The correlation coefficients of the PPV predicting formulas based on
Let lnk2 = lnα2 and lnk3 = lnα3 , then Eq. (10) can be expressed as Eq. (12) are greater than those of the classical Sadovski’s formulas,
β1 indicating that the blast vibration wave propagation law of open pit
3 Q β β
v = k1 k2 k3 ⎜⎛ ⎞ H⎞ 2 ⎛D⎞ 3 slopes subjected to underground mining is too complex to be de-
⎟ ⎛
⎝ r ⎠ ⎝r ⎠ ⎝r⎠ (11) scribed using the classical Sadovski’s formula. The PPV predicting

417
N. Jiang et al.
model with elevation difference and mining depth taken into ac-
count can better describe the influence of open pit subjected to
underground mining on the propagation of waves.
(c) The correlation coefficient of z-component velocity is greater than
that of x- and y-component velocity, showing that z-component
velocity is more prone to be affected by elevation difference and
mining depth.
(d) The correlation coefficient of the PPV predicting formula for A-
slope is slightly smaller than that for the north slope and the south
418
Tunnelling and Underground Space Technology 70 (2017) 409–421
Fig. 12. Numerical monitoring points on the south slope
and the north slope.
slope possibly because A-slope is controlled by Fault F9 and the
attenuation law of blast vibration in it is more complex.
(e) The resultant PPV predicting formulas have good accuracy. Their
correlation coefficients are all greater than 0.721. Furthermore, the
resultant PPV is a combination of velocities in all directions and can
better reflect the blast vibration dynamic response of pit slopes.
Therefore, when evaluating the influence of underground mining
blast on pit slopes, the resultant PPV predicting formula is re-
commended. For Daye Iron Mine, the resultant PPV predicting
Fig. 13. PPVs at monitoring points on the south slope.
N. Jiang et al. Tunnelling and Underground Space Technology 70 (2017) 409–421

Fig. 14. PPVs at monitoring points on the north slope.

Fig. 15. PPVs at monitoring points on A-slope.

Table 6 formulas are

Variables related to PPVs on pit slopes subjected to underground mining.
0.841
H −1.432 D 0.296
Variables Symbol Dimension ⎪ ( ) () ()
⎧12.089 3 Q
r r r
(North slope)
⎪ 0.281
H −1.152 D 0.980
( ) () ()
3Q
Vibration displacement μ L v = 3.437 (South slope)
PPV v LT−1
⎨ r r r
⎪ 0.729
LT−2 H −1.221 D 0.191
( ) () ()
3Q
Particle vibration acceleration a ⎪5.161 (A-slope)
Frequency f T−1 ⎩
r r r
(13)
Mass of explosives Q M
Horizontal distance r L
Elevation difference H L 7. Conclusions
Mining depth D L
Density of rock mass ρ ML−3
Phase velocity c LT−1 (1) According to fan-shaped distribution of medium-length boreholes
Time t T in underground mining, a numerical model is established to analyze
the characteristics of blast loadings by adopting the dynamic finite
element software LS-DYNA. The blast loadings on boundaries out-
L symbolizes the concept of length, M symbolizes the concept of mass, and T symbolizes
side the fracture zone are selected as the input loadings in the
the concept of time.
consequent numerical simulations, so that the blasting vibration

419
N. Jiang et al. Tunnelling and Underground Space Technology 70 (2017) 409–421

Table 7
Part of the simulated PPV data.

Slope Mining depth (m) Horizontal distance (m) Elevation difference (m) x-direction velocity (cm/s) y-direction velocity (cm/s) z-direction velocity (cm/s)

South slope 84.0 211.5 259.3 0.149 0.115 0.061

96.0 219.9 99.1 0.703 0.665 0.714
108.0 146.2 122.8 0.697 0.588 0.604
120.0 240.2 112.4 0.684 0.641 0.689
132.0 262.5 133.5 0.655 0.602 0.654
North slope 84.0 213.8 264.2 0.156 0.113 0.059
96.0 225.3 96.1 0.777 0.662 0.692
108.0 230.4 108.5 0.747 0.638 0.704
120.0 256.9 122.2 0.704 0.602 0.671
132.0 270.5 142.5 0.686 0.582 0.658
A-slope 84.0 233.0 121.0 0.386 0.380 0.379
96.0 215.4 101.6 0.402 0.398 0.423
108.0 256.9 145.3 0.372 0.381 0.392
120.0 201.0 262.9 0.162 0.198 0.115
132.0 141.8 125.5 0.368 0.378 0.372

Table 8
PPV predicting formulas for all pit slopes in Daye Iron Mine.

Slope Direction Eq. (9) Correlation coefficient Eq. (12) Correlation coefficient

North slope x 3 Q 0.803 0.478 3 Q 0.964 H −1.595 D 0.410 0.694

v = 7.354 ⎛
⎝
⎞
r ⎠
v = 10.794 ⎛
⎝
⎞
r ⎠ ( )
r ( )
r
y 3 Q 0.467 0.395 3 Q 0.616 H −1.450 D 0.442 0.727
v = 2.02 ⎛
⎝
⎞
r ⎠
v = 3.341 ⎛
⎝
⎞
r ⎠ ( )
r ( )
r
z 3 Q 0.518 0.546 3 Q 0.764 H −1.323 D 0.247 0.777
v = 2.881 ⎛
⎝
⎞
r ⎠
v = 5.668 ⎛
⎝
⎞
r ⎠ ( )
r ( )
r
Resultant 3 Q 0.609 0.502 3 Q 0.841 H −1.432 D 0.296 0.755
v = 6.472 ⎛
⎝
⎞
r ⎠
v = 12.089 ⎛
⎝
⎞
r ⎠ ( )
r ( )
r
North slope x 3 Q 1.044 0.682 3 Q 0.371 H −1.289 D 1.074 0.711
v = 21.011 ⎛
⎝
⎞
r ⎠
v = 2.749 ⎛
⎝
⎞
r ⎠ ( )
r ( )
r
y 3 Q 0.762 0.602 3 Q 0.192 H −1.170 D 0.976 0.754
v = 7.100 ⎛
⎝
⎞
r ⎠
v = 1.366 ⎛
⎝
⎞
r ⎠ ( )
r ( )
r
z 3 Q 0.867 0.744 3 Q 0.135 H −1.084 D 1.105 0.862
v = 11.680 ⎛
⎝
⎞
r ⎠
v = 1.420 ⎛
⎝
⎞
r ⎠ ( )
r ( )
r
Resultant 3 Q 0.891 0.715 3 Q 0.281 H −1.152 D 0.980 0.808
v = 21.112 ⎛
⎝
⎞
r ⎠
v = 3.437 ⎛
⎝
⎞
r ⎠ ( )
r ( )
r
A-slope x 3 Q 0.572 0.494 3 Q 1.046 H −1.354 D 0.022 0.667
v = 2.013 ⎛
⎝
⎞
r ⎠
v = 7.193 ⎛
⎝
⎞
r ⎠ ( )
r ( )
r
y 3 Q 0.441 0.458 3 Q 0.724 H −1.026 D 0.081 0.692
v = 1.383 ⎛
⎝
⎞
r ⎠
v = 2.937 ⎛
⎝
⎞
r ⎠ ( )
r ( )
r
z 3 Q 0.488 0.515 3 Q 0.345 H −1.410 D 0.602 0.704
v = 1.479 ⎛
⎝
⎞
r ⎠
v = 1.035 ⎛
⎝
⎞
r ⎠ ( )
r ( )
r
Resultant 3 Q 0.490 0.503 3 Q 0.729 H −1.221 D 0.191 0.721
v = 2.706 ⎛
⎝
⎞
r ⎠
v = 5.161 ⎛
⎝
⎞
r ⎠ ( )
r ( )
r

velocity distribution of open pit subjected to underground mining Acknowledgments

can be simulated.
(2) The numerical simulations were validated using field monitoring
data. Numerical simulations show that PPV and its main frequency
on the south slope, north slope and A-slope decrease with elevation
increasing and the altitude amplification effect is not obvious. The
main frequency of blasting vibration is greater than the natural
frequency of pit slopes, and the effect of resonance is weak. It is
quite difficult to set up field monitoring points on pit slopes, and the
impacts of mining blasting on the stability of pit slopes are limited
because the simulated PPVs on them are all less than 0.747 cm/s.
(3) By theoretical analysis, with mining depth and elevation difference
between explosion source and monitoring point taken into account,
a PPV predicting model suitable for open pit slopes subjected to
underground mining is established. Based on numerical simula-
tions, PPV predicting formulas for slopes in Daye Iron Mine are
proposed and can be used to conveniently assess and control the
potential risk of blasting seismic effect.
The study was sponsored by the National Natural Science
Foundation of China (Grant No. 41372312) and the Fundamental
Research Funds for the Central Universities, China University of
Geosciences (Wuhan) (No. 2016140).
References
Afeni, T.B., Osasan, S.K., 2009. Assessment of noise and ground vibration induced during
blasting operations in an open pit mine-A case study on Ewekoro limestone quarry,
Nigeria. Mining Sci. Technol. (China) 19, 420–424.
Ak, H., Iphar, M., Yavuz, M., Konuk, A., 2009. Evaluation of ground vibration effect of
blasting operations in a magnesite mine. Soil Dyn. Earthquake Eng. 29, 669–676.
AQSIQ, 2014. Safety Regulations for Blasting Beijing. China WaterPower Press, China (in
Chinese).
Azizabadi, H.R.M., Mansouri, H., Fouché, O., 2014. Coupling of two methods, waveform
superposition and numerical, to model blast vibration effect on slope stability in
jointed rock masses. Comput. Geotech. 61, 42–49.
Chen, M., Lu, W., Li, P., Liu, M., Zhou, C., Zhao, G., 2011. Elevation amplification effect of
blasting vibration velocity in rock slope. Chin. J. Rock Mechan. Eng. 30, 2189–2195
(in Chinese).

420
N. Jiang et al.
Chen, M., Lu, W., Yan, P., Zhou, C., 2014. New method for dynamic stability analysis of
rock slope under blasting vibration based on equivalent acceleration and Sarma
method. Can. Geotech. J. 51, 441–448.
Choi, B.H., Ryu, C.H., Deb, D., Jung, Y.B., Jeong, J.H., 2013. Case study of establishing a
safe blasting criterion for the pit slopes of an open-pit coal mine. Int. J. Rock Mech.
Min. Sci. 57, 1–10.
Deb, D., Jha, A.K., 2010. Estimation of blast induced peak particle velocity at under-
ground mine structures originating from neighbouring surface mine. Mining Technol.
119, 14–21.
Deb, D., Kaushik, K.N.R., Choi, B.H., Ryu, C.H., Jung, Y.B., Sunwoo, C., 2011. Stability
assessment of a pit slope under blast loading: a case study of Pasir coal mine. Geotech.
Geol. Eng. 29, 419–429.
Deng, X.F., Chen, S.G., Zhu, J.B., Zhou, Y.X., Zhao, Z.Y., Zhao, J., 2015. UDEC-AUTODYN
hybrid modeling of a large-scale underground explosion test. Rock Mech. Rock Eng.
48 (2), 737.
Deng, X.F., Zhu, J.B., Chen, S.G., Zhao, J., 2012. Some fundamental issues and verifica-
tion of 3DEC in modeling wave propagation in jointed rock masses. Rock Mech. Rock
Eng. 45 (5), 943–951.
Deng, X.F., Zhu, J.B., Chen, S.G., Zhao, Z.Y., Zhou, Y.X., Zhao, J., 2014. Numerical study
on tunnel damage subject to blast-induced shock wave in jointed rock masses. Tunn.
Undergr. Space Technol. 43, 88–100.
Esen, S., Onederra, I., Bilgin, H.A., 2003. Modelling the size of the crushed zone around a
blasthole. Int. J. Rock Mech. Min. Sci. 40, 485–495.
Guo, L., Zhang, G., Zhang, D., 2004. Analysis and evaluation on stability of high and steep
slope under energy-concentrated pre-split blasting. In: 2004 International
Symposium on Progress in Safety Science and Technology. Science Press, Shanghai,
China.
Hu, Y., Lu, W., Chen, M., Yan, P., Yang, J., 2014. Comparison of blast-induced damage
between presplit and smooth blasting of high rock slope. Rock Mech. Rock Eng. 47,
1307–1320 (in Chinese).
Hustrulid, W.A., 1999. Blasting Principles for Open Pit Mining: General Design Concepts.
Balkema.
Kesimal, A., Ercikdi, B., Cihangir, F., 2008. Environmental impacts of blast-induced ac-
celeration on slope instability at a limestone quarry. Environ. Geol. 54, 381–389.
Langhaar, H.L., 1951. Dimensional Analysis and Theory of Models. Wiley, New York.
Li, H.B., Xia, X., Li, J.C., Zhao, J., Liu, B., Liu, Y.Q., 2011. Rock damage control in bedrock
blasting excavation for a nuclear power plant. Int. J. Rock Mech. Min. Sci. 48 (2),
210–218.
Li, N., Zhang, X., Yu, C., 2007. Research on dynamic response of prestressed anchorcables
subjected to blasting load on slope. Chin. J. Rock Mechan. Eng. 26, 2593–2600 (in
Chinese).
Li, W.G., Zhang, J.C., 2007. Study on rock mass bedding slope stability under blast seism.
Explosion Shock Waves 27, 426–430 (in Chinese).
Liu, Y.Q., Yang, J., Yu, D.Y., 2010. Study on simulation of discontinuous deformation
analysis for slope stability under blast impact. Acta Armamentarii 31, 74–78.
Lu, W., Yang, J., Chen, M., Zhou, C., 2011. An equivalent method for blasting vibration
simulation. Simul. Model. Pract. Theory 19, 2050–2062.
Lu, Y., 2005. Underground blast induced ground shock and its modelling using artificial
neural network. Comput. Geotech. 32, 164–178.
Marrara, F., Suhadolc, P., 1998. Site amplifications in the city of Benevento (Italy):
comparison of observed and estimated ground motion from explosive sources. J.
Seismolog. 2, 125–143.
Miao, G., 2013. Study on stress and deformation characteristics of caving method un-
derground stope under the influence of high steep slope. China University of
Geosciences. (in Chinese).
Ming, F., Zhu, W., Li, D., 2012. Effect of blasting vibration frequency on slope stability. J.
Central South Univ. (Sci. Technol.) 43, 4439–4445 (in Chinese).
421
Tunnelling and Underground Space Technology 70 (2017) 409–421
Hallquist, J.O., 2007. LS-DYNA keyword user's manual. Livermore Software Technology
Corporation.
Baczynski, N.R.P., 2010. Deformation rebound in a blast-damaged cutback rock slope. In:
Zhao, J., L.V., D.J.-P., M.J.-F., (Eds.), European Rock Mechanics Symposium.
Taylor & Francis-Balkema, Lausanne, Switzerland, pp. 605–608.
Saadat, M., Khandelwal, M., Monjezi, M., 2014. An ANN-based approach to predict blast-
induced ground vibration of Gol-E-Gohar iron ore mine, Iran. J. Rock Mech. Geotech.
Eng. 6, 67–76.
Sangroya, R., Choudhury, D., 2013. Stability analysis of soil slope subjected to blast in-
duced vibrations using FLAC3D. In: 2013 Congress on Stability and Performance of
Slopes and Embankments. San Diego, California: American Society of Civil Engineers
(ASCE).
Song, X.L., Zhang, J.C., Guo, X.B., Xiao, Z.X., 2009. Influence of blasting on the properties
of weak intercalation of a layered rock slope. Int. J. Miner., Metall. Mater. 16, 7–11.
Tang, H., 2007. Field Experimental and Theoretical Study on Blasting Wave Propagation
Characteristics Under Different Terrains. Institute of Rock and Soil Mechanics,
Chinese Academy of Sciences, Wuhan (in Chinese).
Tang, H., Li, H.B., 2011. Study of blasting vibration formula of reflecting amplification
effect on elevation. Rock Soil Mech. 32, 820–824 (in Chinese).
Thote, N.R., Ramana, C.V., 2013. Investigation into effect of blasting on slope stability in
opencast coal mines. In: 10th International Symposium on Rock Fragmentation by
Blasting. CRC Press, Delhi, India, pp. 763–768.
Wu, C., Hao, H., 2005. Numerical study of characteristics of underground blast induced
surface ground motion and their effect on above-ground structures. Part I. Ground
motion characteristics. Soil Dyn. Earthquake Eng. 25, 27–38.
Xie, L.X., Lu, W.B., Zhang, Q.B., Jiang, Q.H., Chen, M., Zhao, J., 2017. Analysis of damage
mechanisms and optimization of cut blasting design under high in-situ stresses. Tunn.
Undergr. Space Technol. 66, 19–33.
Xie, L.X., Lu, W.B., Zhang, Q.B., Jiang, Q.H., Wang, G.H., Zhao, J., 2016. Damage evo-
lution mechanisms of rock in deep tunnels induced by cut blasting. Tunn. Undergr.
Space Technol. 58, 257–270.
Xu, J., Liu, H., Gao, S., 2009. Study on natural period of homogeneous rock slope and its
estimation method. World Earthquake Eng. 25, 61–69 (in Chinese).
Xu, M., Peng, D., 2006. Finite element analysis of dynamic response on blasting vibration
in slope excavation of a hydroelectric power station. Chin. J. Geotech. Eng. 28,
770–775 (in Chinese).
Yan, Y., Zhang, Y., Huang, C., 2014. Impact of blasting vibration on soil slope stability.
Electron. J. Geotech. Eng. 19, 6559–6568.
Yang, Y., Li, Y., Liu, J., Xu, S., 2011. Dynamic stability of open pit slope under blast
loading. J. Northeastern Univ. 32, 1472–1475 (in Chinese).
Yao, Y.K., 2008. Study on failure mechanism and forecasting models of instability of high
rock slope. China Univ. Geosci. (in Chinese).
Yi, C., Sjöberg, J., Johansson, D., 2017. Numerical modelling for blast-induced frag-
mentation in sublevel caving mines. Tunn. Undergr. Space Technol. 68, 167–173.
Zhang, Z., Luan, L., Yao, J., Xie, J., Li, X., 2011. Study on the influence of stability for
higher precipitous open pit slope under mining blasting. Adv. Mater. Res. 255,
3822–3826.
Zheng, Y., Chen, C., Zhu, X., Pan, Z., Ou, Z., Zhou, Y., Xia, T., 2014. Node quasi-static
stability analysis of rock slope under excavation blasting based on UDEC. Chin. J.
Rock Mech. Eng. 33, 3932–3940 (in Chinese).
Zhong, D.W., Wu, L., Chen, H., 2010. Model test and numerical simulation study of dy-
namic characteristics of rock slope under blast loading. Chin. J. Rock Mech. Eng. 29,
2964–2971 (in Chinese).
Zhu, J.B., Deng, X.F., Zhao, X.B., Zhao, J., 2013. A numerical study on wave transmission
across multiple intersecting joint sets in rock masses with UDEC. Rock Mech. Rock
Eng. 46 (6), 1429–1442.