Optimizing Drilling in Open Pit Mine
Optimizing Drilling in Open Pit Mine
Optimizing Drilling in Open Pit Mine
J. Kemeny
University of Arizona, Tucson, Arizona, USA
J. Segui
Julius Kruttschnitt Mineral Research Centre , Australia
ABSTRACT: A necessary requirement for optimization over the complete mine/mill/leach operation is continuous and accurate data at all steps of the process allowing fragmentation, crushabilily/ grindability/ leachability, slope stability, and safety to be evaluated simultaneously. In an ideal sy.stem, these data would be
analyzed centrally and used in a feedback loop to modify mining operations and process-control variables as
necessary to improve performance. The objective of the project described in this paper is to demonstrate technologies that can increase the amount of information obtained during drilling, and understand how this information can best be used to improve blasting results, route blasted rock, and increase the efficiency of
downstream mineral processing. The technological goals of the project presented in this paper include development of various sensors, data-acquisition systems, and online analysis tools that will allow real-time characterization of the rock mass and bore-hole measurements of mineral content during drilling.
1 INTRODUCTION
Mining and mineral processing have traditionally
been approached as if they were separate entities.
However, the mining industry is beginning to look at
mining and milling as two interrelated components
that must be optimized as a whole. There is increasing realization that greater expenditures on
blasting can lead to tremendous crushing and grinding energy savings or to an increase in leach recovery (e.g. Bulow et al., 1998). One of the requirements for being able to optimize the complete
mine/mil 1/leach feedback loop is accurate, online,
and continuous data on key inlonnation on the state
of different parts of the "system." This information
includes the characteristics of the rock about to be
blasted, the characteristics of the blasted rock about
to be sent to the primary crusher, the characteristics
of the rock about to enter the flotation circuit, and so
on. In recent years, online systems have been developed to provide some of this information on a continuous basis.
This paper focuses on optimization of blasting, an
often overlooked part of the mine/mill/leach system.
The information vital to optimizing blast design includes characterization of the rock mass prior to
blasting; it is widely accepted that characterizing
fractures and other discontinuities in the rock mass
is one of the most important inputs to blast design to
achieve optimal rock fragmentation. The work de-
191
192
els. As described below, substantial data are available for each blast site. For the pre-blast stage: GPR
3D rock mass data, mine model geological information, blast-hole drill data from the drill monitoring
system, and geotechnical properties of the intact
rock together with ore content. Blast data: physical
characteristics of each drill hole (diameter, location,
depth, etc.), amount and type of explosive in each
hole, timing patterns, and video tape of the blast itself. For the post-blast stage: rock-mass characteristics (size distribution, particle shapes, etc.) across
the blast area, shape of blast pile, and other properties deemed useful. The first step was to use multivariate statistical techniques to help identify important relationships between pre-blast, post-blast and
actual blast design parameters. Then, using these
initially identified relationships, and knowledge of
existing blasting theory, empirical blast-design models are being developed. We propose to investigate
several modeling approaches including neural networks (ability to develop mappings between input
conditions and output parameters in complex environments) and fuzzy logic. Fuzzy-logic-based systems are well suited for making design decisions
with imprecise, incomplete and uncertain information.
3.1
Fi aine
Location
ot
the
relevant data would be integrated and analyzed ottline to produce a blast design In this case the time
frame tor commercialization is 1-2 years
4 FIELD TESTS IN OPEN-PIT COPPER MINES
The test sites for the work described here are located
in south eastern and south-western Arizona (indi
cated by the arrows in Figure 1) The Molenei min
ing district hosts the largest producing porphyrycoppei deposits in North America The mining com
plex consists ot several open-pit mining areas, a
concentrator with a capacity ot 75,000 tons ot ore
pei day, and the woild's largest solvent extraction
/electiowinning facility Over 780,000 tons ot rock
per day are outed to either n-pt crushing systems
or leaching stockpiles In 1999, the Morenci mining
distnct produced over one-billion pounds ot copper
Mineralization is associated with a co-magmatic
calc-alkaline series ot porphyrtc intrusions anging
in composition from dionte and granodionte to
quartz monzonite and granite (Trler et al 2002)
field
tests
in
southen
Azona
well as Paleozoic limestone and Precamban gante Where
the test mine is located, the geology shows tertay
intrusive rocks including andsite, dorte, ganodorite, quartz monzonite porphyiy and Jurassic quaitz
monzonite (Coopei 1971) The mine has an annual
ore production ot 40 million tons (ot which 22 mil
lion tons goes to a solvent extraction/electrowinmng
facility) The average giade ot copper is 0 3% and ot
Molybdenum is 0 03% The mill cutoff value foi
copper is 0 33% Cu
193
194
Cross-hole radar surveys are conducted using a zerooffset profile method to obtain arrival time versus
depth in adjacent boreholes. For the field tests in
Arizona, the borehole radar system transmitting at
either 50 MHz (for hole spacings between 20 and 30
feet) or 100 MHz (for holes spaced less than 20 feet
apart) was used between adjacent boreholes
(Hopkins el al. 2002). The bench where experiments
were conducted at one of the mines included a fault.
5.3
Figure 5 Vibrations recorded on the drill stem by accelerometers. The horizontal axis is time (seconds)
and the vertical axis is acceleration (g).
Data are being analyzed to determine if vibration of
the drill stem can be used to identify fractures.
Commercialization potential depends on value
added by additional information gained from geophysical measurements under investigation. A system based on vibration measurements made on the
drill rig has the shortest path to commercialization
because it can be incorporated into existing commercial systems that collect and display other drill
data. The project's drilling partner is interested in
commercializing the technology if it proves viable,
so that commercialization within a timeframe of 1-2
years is possible.
195
196
Fgue 7 Specific enegy contours of the Iwo shot locations Lighlei colos coespond to higher specific energy values
compared to darke colos which indicate lelatively low specific energy values Low specific enegy s assoc
ated with softei rock The stiaight line indicates the trace ot the fault line on the bench
6 CONCLUSIONS
Woik to date has demonstrated the feasibility ot ntegatng dulling, rock-mass, blasting and post-blast
liagmentation data to improve blast design Data
horn field tests has been used successfully to nnpiove blast-fragmentation models Thus, an adaptive
blast-design tool that would allow blasting engineers
to bettei optimize blast paiameters including the lo
cation ot boeholes, the charge per hole, and the
timing ot detonation, has strong commercialization
potential With this system, blasting could be opti
mized tor specific downstieam piocesses on a holeby-hole basis, and would be applicable to most any
piocess including crushing and grinding, leaching,
and disposal on a waste pile
Modeling work to date is based on thiee parameteis that aie available tor each blast hole drilling
specific enegy explosive energy per volume ot
197
ACKNOWLEDGEMENTS
This work is supported by the Director, Office of Indus
trial Technologies, of the U.S. Department of Energy un
der Contract No. DE-AC03-76SF00098. The authors
gratefully acknowledge the contributions of their col
leagues at the Lawrence Berkeley National Laboratory
and the University of Arizona. The authors are also grate
ful to industry partners Phelps Dodge Mining Co., Aquila
Mining Systems, LTD., and SPLIT Engineering, LLC,
who provided technical assistance, and access to mines
and drill rigs.
REFERENCES
BoBo. T.W.. 1997. Optimizing blast fragmentation at Cyprus
Sierrita Mine using the Split Image Analysis System. MS
Thesis. University of Arizona.
Bulow. B.. Smallbone. P.. and Walker. B.. 1998. Blasting for
reduced process plant costs at Argyle Diamond Mine.
Australasian Institute of Mining and Metallurgy. Pub. Se
iles n4. pp. 199-208.
Cooper. J.R. 1971. Mesozoic Stratigraphy of the Sierrita
Mountains. Pima County, Arizona. Geological Survey
Professional Paper 65H-D.
Girdner. K., Kemeny. J.. Annavarapu. S. and McGill, R..
1996. The Split System for analyzing the size distribution
of fragmented rock. Proceedings of the Fragblast - 5
Workshop on Mats, of Blast Fragmentation, J. Franklin
andT. Katsabanis. eds.. Balkema. pp. 101 - 108
Hopkins. D.. R. Haught. M. Karaca. D. Trler. L. Myer. .1.
Kemeny. P. Lever and M. Lowery. 2000. Blasting optimi
zation using seismic analysis and x-ray fluorescence
198