See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/234794783

Blast simulation in underground mines using the combined finite-discrete

element method

Article · September 2007

CITATION READS

1 775

5 authors, including:

Riyu Wei Warna M Karunasena

The University of Queensland University of Southern Queensland 
4 PUBLICATIONS   31 CITATIONS    191 PUBLICATIONS   4,180 CITATIONS   

SEE PROFILE SEE PROFILE

Nagaratnam Sivakugan
James Cook University
226 PUBLICATIONS   5,308 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

GFRP-reinforced concrete columns View project

Structural behaviour of low grade timber beams and slabs View project

All content following this page was uploaded by Warna M Karunasena on 03 March 2014.

The user has requested enhancement of the downloaded file.

Proceedings of the 7th WSEAS International Conference on Simulation, Modelling and Optimization, Beijing, China, September 15-17, 2007 256

Blast Simulation in Underground Mines Using the Combined

Finite-Discrete Element Method

RIYU WEIa, SHENG ZHANGa,b, WARNA KARUNASENAc, NAGARATNAM

SIVAKUGANd, HONGWU ZHANGb
a
Advanced Computational Modelling Centre, University of Queensland, St Lucia, Qld
4072, AUSTRALIA
b
Department of Engineering Mechanics, Dalian University of Technology, Dalian
116024, CHINA
c
Faculty of Engineering and Surveying, University of Southern Queensland,
Toowoomba, Qld 4350, AUSTRALIA
d
School of Engineering, James Cook University, Townsville, Qld 4811, AUSTRALI
rywei@acm http://www.acmc.uq.edu.au

Abstract : The purpose of this paper is to investigate the application of a combined finite-discrete element
program, such as ELFEN, for the simulation and prediction of backfill stope failure risks due to blast loading
effects in underground mines. A real size of a backfilled stope blasted in the nearby rock of the paste fill is
modelled. The dynamic fracture propagation and fragmentation in rock and the behaviour of paste fill by using
the proposed model are provided and discussed. The results show the proposed model can be applied to provide
an effective means to simulate substantial insight into the dynamic fracture processes during the blasting and
estimate failure risks of the paste fill by using the peak particle velocity as a measure of blast damage.

Key-Words: finite-discrete element method; paste backfill; blast simulation; rock damage;underground mine

1 Introduction fracture propagation and fragmentation and

Paste fill, which is a kind of backfill material in an predicting the stability of paste fill in backfilled
underground mine, consists of tailing of minimum stopes due to mine blast.
15% of fill size less than 20 µm, water and small The combined finite-discrete element method is a
percentage of binder. Recently paste fill has gained recently developed numerical method aiming at
rapid acceptance as an alternative backfill material to modeling failing, fracturing and fragmenting solids.
the conventional cemented hydraulic fills (Belem & The discrete element method (DEM) developed
Benzaazoua [1], Rankine & Sivakugan [2]). Paste fill originally by Cundall and Strack [3] is that the
has substantial benefits to mining operations domain of interest is treated as an assemblage of rigid
including an effective means of tailings disposal, or deformable blocks/particles/bodies and the
improvement of local and regional rock stability, contacts among them can be identified and
greater ore recovery and greatly reduced continuously updated during the entire
environmental impacts. Once the ore from a stope is deformation/motion process. The effect of the blast
removed, the stope is filled with paste fill. As the pressure waveform on the fracture process was
mining sequence progresses, stopes adjacent to the studied using the DEM (Donze et al [4]). To simulate
fill are mined, and the fill is subjected to blast loads the complex behavior of blasting which causes
and subsequently exposed. The fill must remain extensive fracture and fragmentation processes,
stable at this time to provide support to the area and Munjiza et al. [5, 6] developed an interesting
to avoid reprocessing the material. The purpose of detonation gas model based on the combined
this study is to utilize the combined finite-discrete FEM/DEM.
element program ELFEN for simulating the dynamic
Proceedings of the 7th WSEAS International Conference on Simulation, Modelling and Optimization, Beijing, China, September 15-17, 2007 257

Up to now, some commercial DEM software, empty stopes. Paste fills or sand fills are used to fill
such as ELFEN (Rockfield [7]) and UDEC (Itasca [8]) the voids.
etc., have appeared. These software already show
significant potential in the investigation of
discontinuum modelling. ELFEN, a combined
finite/discrete element code, enables to model intact
behaviour, interactions along existing discontinuities,
and it has been successfully applied to the simulation
of intense fracturing associated with surface mine
blasting, mineral grinding, rock slope (Eberhardt et al.
[9]) and underground rock caving (Munjiza et al. [5]).
Taking the advantages of ELFEN, a numerical model
is developed for the simulation of a backfill stope
failure due to blast loading effects in underground
mines. The remainder of the paper is structured as
follows: Section 2 describes the mining method and
the development of the numerical model. Section 3
analyses the dynamic fracture processes and
discusses the numerical results. The conclusions are
drawn in Sections 4.
Figure 1 Schematic diagram of under ground mine
(cut and fill method)
2 Mining and Model Descriptions
There are several mining methods. These can be The elevation view of stope shows the access drives
classified as open surface mining and underground and the blasting pattern in Fig. 2. The blasting
mining. Most of the mines are using underground advances from bottom to top. The access drives
mining methods over open pit surface mining. comes from neighboring stope. There is more than
Underground mining can be divided into room and one access drive for one stope. The blasting is done in
pillar method and cut and fill method. different batches. The bottom level of stope is blasted
first and the crushed material is removed. Then the
blasting is done for above layer. The filling is done
2.1 Cut and fill mining after blasting whole stope and removing all the
In this study, the cut-and-fill mining method is used material.
for underground mining. Cut-and-fill mining is
applied for mining of steeply dipping ore-bodies in
strata with good to moderate stability and a
comparatively high grade mineralization. It allows
selective mining to recover high grade sections
separately and leave low grade rock behind in stopes.
The procedure of cut and fill mining can be
explained in Fig. 1. Spiral ramps are made in footwall.
It provides the access drive to undercut. Cut-and-fill
mining excavates the ore in horizontal slices, starting
from a bottom undercut, advancing upward. After the Figure 2 The procedure of drilling and blasting
ore was drilled and blasted, the muck is loaded and
removed from the stope. When the full stope area has The plan view of stope arrangement is given in Fig.3.
been mined out, voids are backfilled with sand In this case the blasting is done in a secondary stope,
tailings or waste rock. The fill serves to support both which is critical for design.
stope walls and working platform for equipment, It has access through the rock (neighboring stope)
when mining the next slice. The fill often consisting as shown in Fig. 2. The ore is blasted and removed
of de-slimed sand tailings from the dressing plant of to create the voids. The rock and neighboring stopes
mine, at times complemented by waste rock support these voids.
produced by development excavations, dumped in
Proceedings of the 7th WSEAS International Conference on Simulation, Modelling and Optimization, Beijing, China, September 15-17, 2007 258

The model is developed using the finite-discrete

element program ELFEN. As it is the simulation of
blast in underground mines, the non-reflecting
boundary is assumed in the surrounding boundary of
the model. Furthermore, the bottom edge is fixed,
the top edge is fixed in the vertical direction, the left
and right edges are fixed in the horizontal direction,
and other inside boundaries are free. The model is
meshed with triangular elements by Delaunay
triangulation method. The material properties used
Figure 3 Plan view of stope arrangement in this simulation are listed in Table 1.

2.2 Model description Table 1 Material properties

Model is based on the scenario of a single column of Paste Fill Mine Rock
emulsion explosive in a rock mass, as shown in Fig.4.
The rock to be blasted is 35 m in height and 5 m in Young’s Modulus (GPa) 5.5 95
thickness. The blast hole is 89 mm in diameter. The Poisson’s Ratio 0.2 0.2
blast hole drilled from bottom to top, is 25 m deep in 3
this model, and is assumed to consist of 20 m length Density (kg/m ) 2100 2700
of explosive material and 5 m of stemming Cohesion (MPa) 0.179 0.2
material. The stemming material is used to ensure Damping 0.0638 0.03
that the gases provide a load to the walls of the blast
hole rather than escaping from the blast hole, and Tensile Strength (MPa) 6 5
does not provide any strength to the system. Fracture Energy (J/mm) 0.06 0.06
Therefore, the stemming material is not modelled.
For the purpose of this modelling, it is assumed that A time-varying pressure pulse function is used to
there is no paste fill in the vicinity of the explosive model the blast loading for the numerical model,
column. which is applied on the walls of an “equivalent
cavity” as defined by (Sharpe [10]).

P(t ) = t n p 0 e − xt (1)

where t is the duration of blast loading, and P (t ) is

the pressure at time t . p0 denotes the initial peak
pressure value. p0 , n and x are constants, which
determine the shape of the pressure curve. p0 is
chosen so that a maximum pressure of 2.255 × 108 Pa
is achieved. In the paper, the values n = 1 and
x = 2000 are used.

3 Results and Discussion

The dynamic fracture process starts from a
continuum representation of the solid material by
finite elements. Fracturing in the model is controlled
according to a fracture criterion specified through a
constitutive model, i.e. Rankine crack model. The
processes of the fractures and fragmentations from
t = 10µs up to t = 350 µs are shown in Fig. 5, in
Figure 4 Geometry of the analysis model for blast
simulation
Proceedings of the 7th WSEAS International Conference on Simulation, Modelling and Optimization, Beijing, China, September 15-17, 2007 259

which the black lines indicate cracks, and the red and Due to the short detonation period that ends at about
blue zones are paste fill and mine rock, respectively. t = 5µs , the compressive and tensile stress fields
should radiate from the blast hole. Fig. 5(a) shows
initiation and propagation of a large number of radial
cracks starting around the blast hole, i.e. the vicinity
of the compressive failure zone. Then the tensile
stress field causes the radial cracks to extend, which
is then followed by stress releases around the
extending cracks. The differences in the crack
growth velocities increase as time increases. This is
because the cracking is affected by different spatial
microscopic strengths for an applied stress level. As
the fracture processes continue extending, a large
number of separate interacting fragments, i.e.,
discrete elements are created,. This will result in
further fracturing of both the remaining continuum
(a) t = 10µs (b) t = 50 µs and previously created discrete elements.
The peak velocities of individual material points
along different height levels in paste fill are shown
in Figs. 6-11. It can be observed that the peak
velocities decrease quickly as the distance from blast
hole increases. The peak velocities in Figs. 6 and 11
are lower than in Figs. 7-10. This is because the
horizontal surfaces at various heights are vertical
(what does this mean?) to the blast hole and the
individual material points in Figs. 7-10 are closer to
the blast hole. Figs. 12-14 show velocities against
times at material points I, II and III, as specified in
Figs. 6-8. It is noteworthy that the velocities of these
points are still subject to violent ocillation after the
peak phase as the time increases. An explanation for
this is that the processes of the fractures and
fragmentation result not only from detonation of
(c) t = 100µs (d) t = 150µs explosive but also from considerable fragments
interacting each other.

Figure 6 Peak velocity at height of 45 m in paste fill

(e) t = 200 µs (f) t = 350 µs

Figure 5 Fracture patterns at different times

Proceedings of the 7th WSEAS International Conference on Simulation, Modelling and Optimization, Beijing, China, September 15-17, 2007 260

Figure 7 Peak velocity at height of 40 m in paste fill Figure 11 Peak velocity at height of 20 m in paste fill

0.25

0.20

Velocity [m/s].
0.15

0.10

0.05

0.00
0.10 0.15 0.20 0.25 0.30 0.35 0.40
Time [s]

Figure 8 Peak velocity at height of 35 m in paste fill Figure 12 Velocity at Point I (See Fig. 6)

0.5

0.4
Velocity [m/s].

0.3

0.2

0.1

0
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35
Time [s]

Figure 9 Peak velocity at height of 30 m in paste fill Figure 13 Velocity at Point II (See Fig.7)

0.50

0.40
Velocity [m/s].

0.30

0.20

0.10

0.00
0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35
Time [s]

Figure 10 Peak velocity at height of 25 m in paste fill Figure 14 Velocity at Point III (See Fig. 8)
Proceedings of the 7th WSEAS International Conference on Simulation, Modelling and Optimization, Beijing, China, September 15-17, 2007 261

4 Conclusion Numerical methods in Engineering, Vol. 49,

A real-size model has been developed using a 2000, pp.1495-1520.
combined finite-discrete element program ELFEN [7] Rockfield, ELFEN 2D/3D Numerical Modelling
to predict and simulate backfill stope failure risks Package version 3.7, Rockfield Software Ltd.,
due to blast loading effects in underground mines. Swansea, 2004.
The dynamic fracture processes are shown starting [8] Itasca, UDEC Version 4.0. Itasca Consulting
from detonation of explosive to fracturing to Group Inc, Minneapolis, 2004.
separate interacting fragments. The extensions of [9] E. Eberhardt, D. Stead, and J. S. Coggan,
fracture processes related to blast-induced hole Numerical Analysis of Initiation and
breakdown start from compressive stress fields to Progressive Failure in Natural Rock
tensile stress fields, and mainly result in the motion Slopes―the 1991 Randa Rockslide,
and interaction of separate fragments. The velocities International Journal of Rock Mechanics and
for individual material points in paste fill have been Mining Sciences, Vol. 41, 2004, pp. 69-87.
determined to investigate the local and regional [10] J.A. Sharpe, The Production of Elastic Waves
stability. The combined finite-discrete element by Explosion Pressures. I: Theory and
method has demonstrated that it is possible to have Empirical Field Observations, Geophysics,
significant potential to predict failure risks of Vol.7, 1942, pp. 144-154.
backfilled stopes and provide assistance in mining
design.

Acknowledgement
The work reported has been supported by the
Australia-China Special Fund For S&T Cooperation
(CH040024), which is greatly acknowledged.

References:
[1] T. Belem, and M. Benzaazoua, An Overview of
the Use of Paste Fill Technology as a Ground
Support Method in Cut and Fill Mines.
Syposium on Ground support in Mining and
Underground Construction, London, 2004,
pp.637-650.
[2] R. M. Rankine and N. Sivakugan, Backfilling of
Underground Mines with Paste, 9th Australian
–New Zealand Conference on Geomechanics,
Auckland, Feb 8-11, 2004, pp.225-232.
[3] P. A. Cundall and O. D. L. Strack, A Discrete
Numerical Model for Granular Assemblies,
Geotechnique, Vol.29, 1979, pp.47-65.
[4] F. V. Donzé, S. A. Magnier, and J. Bouchez,
Modelling Fractures in Rock Blasting.
International Journal of Rock Mechanics and
Mining Sciences, Vol. 34, 1997, pp.1153-1163.
[5] A. Munjiza, D.R.J. Owen, and N.A. Bicanic, A
Combined Finite-Discrete Element Method in
Transient Dynamics of Fracturing Solids.
International Journal of Engineering
Computations, Vol.12, 1995, pp.145-174.
[6] A. Munjiza, J.P. Latham, and K.R.F. Andrews,
Detonation Gas Model for Combined
Finite-Discrete Element Simulation of Fracture
and Fragmentation, International Journal for

View publication stats