1.1. Defining The Rockburst Problem: Chapter 1 - Introduction
1.1. Defining The Rockburst Problem: Chapter 1 - Introduction
1.1. Defining The Rockburst Problem: Chapter 1 - Introduction
1.
INTRODUCTION
1.1.
Chapter 1 Introduction
1.2.
Chapter 1 Introduction
pillars or active development faces and the use of remotely operated equipment have
been successfully applied at several mines. For all approaches, a good understanding of
the seismicity which has or is occurring in a mine is vital. As such, seismic monitoring
is one of the key aspects to successful management of seismic risk in underground
mining.
1.3.
Whilst the earliest recorded evidence of mine seismicity dates back prior to 1900 in the
Polish Upper-Silesian Coal Basin (Ortlepp, 2006), the first attempts at detailed
monitoring occurred in the United States, where mining induced seismicity was
discovered almost by accident in the late 1930s during experiments with the
measurement of shockwave propagation in an underground mine. It was later realised
that by capturing the shockwaves resulting from mine seismicity, there was an
opportunity to locate and measure the magnitude of mining induced seismic events.
Mine-wide microseismic monitoring systems were later installed in gold mines in South
Africa and Idaho in the 1960s while the first commercial monitoring system, the
Electrolab MP-250, became available in the late 1970s (Potvin et al., 2000). This system
was subsequently installed in several mines in Canada and the USA (MacDonald and
Muppalaneni, 1983; Davidge, 1984; Neumann, 1985; Oliver and MacDonald, 1985).
Following these early findings, research in the 1980s primarily involved the
development of sophisticated commercial seismic monitoring systems for use in a
mining setting. Three systems were developed:
The South African ISS system (de Kock and Mountfort, 1998).
The ISS is now the most commonly used seismic monitoring system in Australian
mines. With these high resolution and highly accurate systems in place at mines,
Chapter 1 Introduction
advanced seismological analysis of mine seismicity data was now possible, although not
widely adopted.
Research in the 1990s concentrated mainly on data analysis and defining the nature of
the rockburst problem, while refining the seismic systems hardware and software
(Potvin et al., 2000). Major rockburst and seismicity research thrusts were undertaken in
South Africa and Canada. The Canadian Rockburst Research Program (CRRP) was
undertaken from 19901995 with the aims of improving the understanding of the
rockburst problem and making use of seismic monitoring information as inputs into the
mine design process (CAMIRO, 1995). The research program led to the publication of
six volumes of information on defining rockburst damage mechanisms, ground support
for rockburst conditions, microseismic monitoring, numerical modelling and mine
design. The findings of this project were relevant to Australian mines, given the similar
mine layouts and mining methods to Canadian operations, and are frequently referenced
in this thesis.
Most South African Research in the 1990s and early 2000s was conducted under the
auspices of the Safety in Mines Research Advisory Committee (SIMRAC). This
research focussed on controlling (and to some extent predicting) rockbursts, ground
support and theoretical mine seismology (SIMRAC, 2001). Whilst the research is
largely focussed on South African, shallow dipping, tabular mining layouts, there are
relevant findings to rockburst management in Australia, particularly with regards to
ground support performance under rockburst conditions.
In recent years the focus of mine seismicity and rockburst research in Australia has been
on risk management. The Australian Centre for Geomechanics (ACG) launched the
Mine Seismicity and Rockburst Risk Management (MSRRM) project in 2000, of which
this thesis is part, with the aims of better defining the nature and extent of the problem
and by developing risk analysis tools to quantify the hazards associated with rockbursts
and mine seismicity (Potvin et al., 2000). This is due in part to the realisation that, given
the right set of conditions, seismic events and rockbursts will inevitably occur in mines.
Such events cannot be reliably predicted, therefore the best approach is to understand
the problem as well as possible through monitoring, then circumvent the risk associated
with mining induced seismicity and rockbursting by making changes at the planning
4
Chapter 1 Introduction
stage or by managing the risk tactically as mining progresses. In the last decade, the
proliferation of microseismic monitoring equipment in Australian mines has provided
an extraordinary opportunity to study the rockburst phenomenon in local mining
conditions.
The current state of understanding or recognition in Western Australia is reflected in
mining industry codes of practice. The Western Australian Mines Safety Inspection Act
(DOCEP, 2006a) contains several references to mining induced seismicity and
rockbursting.
Point
Standard
7.3
The potential for mining induced seismicity or rockbursts to occur in the mine has been assessed.
7.10
A seismic monitoring system is installed in a mine where seismic and rockburst activity causes
damage to the openings and/or the rock support and reinforcement systems in the mine.
7.11
A seismic monitoring system where installed is capable of detecting, processing and displaying a
representative sample of the range of seismic events occurring in real-time.
7.12
Table 1.1
A more obvious indicator of how seriously the issue is taken in Australia is the number
of mine-wide seismic monitoring systems installed at mines (see Figure 1.1). There has
been a steady increase in the number of systems, due to several factors: an increasing
numbers of mines as metal prices have risen, deepening of existing mines as near
surface resources are exhausted, and the increasing awareness brought about by
widespread publication of research material and major incidences such as that at
Beaconsfield.
Chapter 1 Introduction
NUMBER OF SYSTEMS
50
40
30
20
10
0
1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
YEAR
Figure 1.1
Some of the more recent research work has focussed on the relationship between the
generation of seismic events and output from numerical modelling. Much of this work
has focussed on the use of seismic data in the calibration of mine-wide numerical
models (Beck et al., 1997; Wiles et al., 2000; Diederichs et al., 2002). These techniques
have been shown to be useful in complimenting calibration techniques involving
conventional instrumentation, such as closure measurements, stress cells and
extensometers, or underground observations. More recently, regional scale non-linear
numerical models have been used to achieve some success in forecasting the amount of
energy dissipated as a result of rock mass yield, of which a certain percentage is
measurable seismicity (Beck et al., 2006). Forecasting the amount of damage a seismic
event causes to an excavation remains unresolved using these techniques.
Ultimately, the interaction between seismically induced ground motion and a
structurally intricate and compositionally heterogeneous rock mass is extremely
complicated. As such, the problem may lend itself to an empirical or probabilistic
approach, which accounts for excavation specific factors.
Chapter 1 Introduction
1.4.
Research hypothesis
This thesis is concerned with rockburst damage. That is, given a seismic event occurs at
or near an excavation, what level of damage to that excavation can be expected? More
specifically, in a seismically active mine where an excavation is subjected to an episode
of dynamic loading, the amount of damage done is dependant upon local site conditions.
It is hypothesised that the degree of expected rockburst damage can be shown to be
related to a number of excavation specific, simple to obtain, objective and quantifiable
factors. These factors are:
The magnitude of the seismic event and distance from the event to the
excavation (defined by an estimate of the peak particle velocity (PPV) at the
excavation boundary);
An estimate of the ratio of rock mass strength (UCS) to the in situ stress (1)
as determined by a well calibrated numerical model;
The presence of stability reducing rock mass structure which may contribute
to an increased failure thickness.
The aim of this thesis is to demonstrate that this relationship exists and an outcome will
be a practical tool for use in seismically active underground mines. This tool will allow
mine sites practitioners to determine the likelihood of rockburst damage occurring at
any location in a mine, based on the factors listed above.
A parameter termed the Rockburst Damage Potential based on data from several
hundred incidences of rockburst damage from seismically active mines in Australia and
Canada, will combine these factors to give a rating of the expected rockburst damage
for a particular excavation.
Chapter 1 Introduction
1.5.
Methodology
The methods used to achieve the thesis aims are shown in Table 1.2 below.
Investigate the current state of knowledge regarding mine seismicity and
rockbursting. In particular, identify the gaps in existing knowledge with
respect to:
Literature Review
Australian and Canadian mines. The mines should cover a range of rock mass
Build the rockburst damage potential method into existing software used by
Method at Underground
Mines
Table 1.2
Research methodology
Chapter 1 Introduction
1.6.
Chapter 1 Introduction
a repeatable and reliable testing technique. There is no widely accepted technique for
the design and selection of ground support for dynamic loading conditions. The results
of the simulated rockburst testing carried out as part of this research have been
presented in a format in this thesis which can be used to aid in this process. This thesis
focuses on the role of ground support, since changing or upgrading ground support is
the most obvious and common tactical approach employed at underground mines to
mitigate seismic risk.
The quality and amount of data available to study the rockburst phenomenon is
illustrated in the chart in Figure 1.2. This thesis aims to increase the amount of data at
the top of the pyramid, which can be classed as very high quality but of which there is
little available. In the case of ground support, this thesis will make use of the top two
categories of the pyramid, both case histories and in situ testing.
Rockburst
Case Histories
Increasing
quality of
data
In Situ Testing
Laboratory Testing
Numerical Modelling
10
Chapter 1 Introduction
1.7.
Organisation of thesis
Chapter 1 Introduction Introduces the research field and describes the thesis
topic.
11
Chapter 1 Introduction
Chapter 9 References.
Appendix C Logitfit Matlab Routine Provides Matlab code for the routine
used to perform logistic regression on the rockburst case history data.
12