Search Documents

The value derived by good mine planning is often underestimated as many mines steer their attention to the more immediate and more easily recognised successes made by the day-to-day operational æbush fireÆ fighting. Similar to machine maintenance, the efforts made to quickly rectify short-term breakdowns are well recognised; whereas the value derived from the work in planning, scheduling and undertaking preventative maintenance is not so obvious or readily quantified. Too often the focus of the business shifts to building and supporting an efficient operations team at the expense of developing a highly experienced team of mine planning practitioners that are well trained with procedures and purpose-built tools to develop optimum mining strategies. A resource has an inherent value based on its quality parameters and its likely mining conditions. How well that value can be unlocked is governed, to a large degree, by the proficiency of the mine planning practitioners, as they, essentially, set the foundation of the mine business and the path along which the mine will operate. The importance of having a well documented and procedural mine planning methodology that integrates both a thorough technical and economic understanding of the resource cannot be over emphasised. The aim of the mine planning process is to develop a robust mine design and production capability which can best cater for future uncertainty whilst providing the best opportunity to maximise cash flow into the business. The planning process must be transparent and auditable, clearly summarising the process steps and assumptions made to arrive at a justifiable and defendable outcome. To provide maximum value to the business, the planning process must be repeatable and scalable to enable easy replication throughout the different mines and projects within an organisation. The diminishing numbers of highly experienced personnel within our industry, coupled with the growing pressure to clearly demonstrate that appropriate due diligence has been exercised, means that well documented and monitored mine planning processes using purposed built and innovative tools need to be developed to ensure optimum outcomes.

Jan 1, 2009

Exploration for economic nickel sulphide mineralisation has often been directed by rationalisations based on poorly understood genetic models. Among those that have enjoyed popularity over the past decade are the "layered gabbro" model and the "basal ultramafic" or Kambalda model. The new discoveries in the Mt Keith-Agnew area could well become a model for future nickel search. Exploration based on even the crudest state- ment of the Kambalda model - prospect the basal contact of Archaean ultramafic complexes for gossanous expressions of sulphide mineralisa- tion - has met with remarkable success. Refine- ments to this model, besides satisfying the geology of any particular deposit, have the major virtue that many of the genetic options - magmatic segregation, hydrothermal transport, sulphurisation, replacement sedimentary, structural control, etc. - are still open. Recent descriptions of highly magnesian silicate melts associated with the ultramafic rocks, do however, place some constraints on the formation of these deposits. Further evidence, from experimental studies on sulphide-silicate and sulphide-oxide phase relationships, and nickel-iron exchange in pyrrhotite, is taken with descriptions of sulphide nickel occurrences in Western Australia to examine genetic models for existing ore bodies.

Jan 1, 1972

By P N. Holtham

Magnetic separation is routinely used in the processing of titanium minerals, and the efficiency of these operations can be determined by measuring the magnetic properties of the process streams. The valuable components in the feed to a titanium minerals plant have specific magnetic susceptibilities ranging from -10-9 m3 kg-1 (zircon) to 10-6 m3 kg-1 (ilmenite). However, in an industrial environment, in which the mineral temperature may range from 20 to 120¦C, it is difficult to measure susceptibilities lower than about 10-7 m3 kg-1. Hence laboratory testing of spot samples from magnetic separators is currently required to obtain accurate performance data. This is an inherently slow process and where feed grades are variable, optimum performance of magnetic separators is generally not achieved. This paper describes the development of an inductance-based instrument for measuring very low levels of magnetic susceptibility on-line. The signal from the instrument can be used for process control purposes to maintain optimum magnetic separator performance. A prototype instrument has been installed in the zircon scavenger circuit in an Australian titanium minerals plant. The circuit consists of a roll magnetic separator treating a feed consisting mainly of monazite and stained zircon. The monazite-rich magnetic fraction is returned to the mine site for disposal, while the non-magnetic fraction is recycled to the zircon wet circuit. Splitter positions in the separator are adjusted when the final zircon product fails to meet specification, or the magnetic fraction contains more than 30 per cent zircon. Weekly composite grain counts show that sub-optimal operation of the magnetic separator results in significant zircon losses into the magnetics fraction. Using the instrument, the zircon content of this stream can now be accurately determined from the measured magnetic susceptibility, provided a correction for the effect of mineral temperature is applied using the Curie - Weiss relationship.

Jan 1, 2001

By T Goswami, M Edmondson

Henry Walker Eltin (HWE) are the mining contractors for Gloucester Coal at Stratford mine, which is in an environmentally sensitive area in the Avon Valley, NSW, Australia. Normal blasting at Stratford is done in a æstand upÆ or æbox cutÆ mode to suit the truck and shovel operations. Recently an opportunity of doing a throw blast became available when part of a ramp had to be blasted to accommodate changes to the pit design while not damaging the remainder of the ramp. The objective was to maximise throw into the void in order to minimise excavation costs by reducing the amount of overburden to be moved by the excavation fleet. It was necessary that the best result be achieved while staying within the vibration and airblast limits at a neighbouring property and nearby township. Both objectives were achieved through the application of OricaÆs knowledge of blast design, modelling and i-konTM electronic detonators. The i-kon system was used because of its known benefits in cast blasting and environmental control. The i-kon system provides the blaster with much greater accuracy, control and the flexibility to tailor blast designs. It has been found previously to enhance blast throw and provide cleaner high walls (Brent, 2002). The blast was a success, producing a muckpile with a high degree of looseness and consistent fragmentation, with an average throw to final spoil position of 37.5 per cent. This was achieved without any environmental exceedance. In addition, high speed filming showed that the coal edge had remained largely in place, indicating minimal coal loss. The results exceeded the expectations of HWE Stratford and the overall cost savings to HWE Stratford from this blast were recognised to be substantial. Further throw blasting opportunities at Stratford have been followed using the i-kon system to maximise benefits.

Jan 1, 2003

By S D. Golding, K Baublys, V Nanjappa, C J. Moran

Water reuse is becoming an integral component of the water management strategy on mine sites. This practise is being driven by corporate sustainability goals, community and societal pressures to demonstrate improved water stewardship, as well as climate and regulatory pressures. However, water reuse often results in water quality compromise which can then result in decreased recovery through problems in processing circuits, product quality, and an increased likelihood of discharge of water that cannot meet environmental regulatory requirements. On most mine sites, there is usually a disjuncture between water quantity management and water quality management with the latter being managed solely as an environmental problem. It is becoming increasingly clear that water quality and quantity must be managed as an integrated system. In order to integrate water quality and quantity management to achieve multiple objectives the dynamics of water and constituents must be understood. Using examples from a study being conducted at a coal mine in the Bowen Basin, this paper will outline the dynamics of water and salts on the site.

Jan 1, 2009

By M H. Randell, J B. Haines

The Honeymoon Uranium Project comprises the Honeymoon Deposit,Goulds Dam Deposit and Yarramba Prospect in the Southern Lake Fromeregion, South Australia. Both the Yarramba Prospect and HoneymoonDeposit (which includes Honeymoon and East Kalkaroo orebodies) arelocated in the Yarramba Palaeovalley. The Goulds Dam Deposit is about75 km north west of Honeymoon, in the Billeroo Palaeovalley. Exploration for sediment-hosted uranium began in the area in the late1960s, culminating in the discovery of Honeymoon and Goulds Dam inthe early 1970s. In 1982, a 25 L/s demonstration plant was built atHoneymoon to confirm suitability for in situ leaching. The project was put on hold in 1983 due to changes in government policy. Southern Cross Resources Australia Pty Ltd acquired the project in mid 1997. Roll-front deposits form from a migrating geochemical cell, anadvancing reduction-oxidation interface between oxygenateduranium-bearing groundwater and its reduced aquifer host. The source of metal is uranium-anomalous granites, which were eroded fromsurrounding ranges. The weathered granites and resultant sediments are stripped of uranium by oxidised groundwaters, to form solutions carrying uranyl carbonate complexes. The solutions percolate down-slope through permeable sand zones until contacting a reduced environment where uranium precipitates. Uranium mineralisation occurs interstitially between and as thin coatings on sand grains, usually in the form of uraninite or coffinite. The palaeovalleys (previously termed palaeochannels) are incised intounderlying Cambrian/Precambrian basement rocks and filled withsemi-consolidated, largely un-cemented, Tertiary sediments of the Eyre Formation. The late Palaeocene to middle Eocene Eyre Formation is the basal unit of the Tertiary succession in the Callabonna Sub-basin of the Lake Eyre Basin. Further to the north, the Lake Eyre Basin overlies the Jurassic-Cretaceous Eromanga Basin, which comprises much of the Great Australian Basin. Eyre Formation sediments are in turn overlain by the late Tertiary Namba Formation and terrestrial Quaternary units. At Honeymoon, the Eyre Formation includes immature, pyritic,carbonaceous sands and gravels interbedded with lignite andkaolinite-illite-montmorillonite clays. It was deposited by braidedstreams during epeirogenic uplift of the Olary Ranges, accompanied bysubsidence of the Lake Eyre Basin. The Eyre Formation comprises aconsistent stratigraphy of interbedded sand and clay units that arecontinuous over the Honeymoon Deposit but with local variationindicating some hydrological connection between aquifers.In the Honeymoon orebody, uranium occurs within coarse-grained,pyritic sands where the basal sand pinches out between overlyingcarbonaceous clay and the southern palaeovalley slope. Its characteristics are typical of the well-developed, planar limb of a roll-front uranium deposit. Zonal changes related to the reduction-oxidation interfaces across the deposit are readily observable. The orebody extends for nearly 1000 m along the valley margin, is 400 m wide at its maximum and averages about 6 m in thickness. The East Kalkaroo orebody also lies near the southern margin of theYarramba Palaeovalley. It extends discontinuously for approximately 3.5 km and averages 200 m width, with significant uranium present in the Middle as well as the Basal Member. The Billeroo Palaeovalley extends northward for 40 km sub-parallel tothe western side of the Benagerie Ridge, is up to 5 km wide and hasgently dipping banks. Eyre Formation within the palaeovalley isconcealed beneath 90 m of Quaternary and Namba cover, isapproximately 40 m thick, and is sub-divided into three upward-finingunits, which are remarkably similar to the Yarramba Palaeovalleysequence.

Jan 1, 2001

By Valenta R. K

Copper-gold deposits associated with iron oxide rich rocks (ironstones) form a distinctive and significant ore association which occurs widely, but not exclusively, in the Proterozoic. The ironstone association is geochemically distinct from the BIF-hosted gold association (Phillips and Groves, 1983). Not only is the former association higher in its copper and commonly uranium, FEE, ¦bismuth, molybdenum contents, but also the ironstones in many cases do not originate as chemical sediments. Although both associations form in metamorphic environs, they reflect differing tectonostratigraphic settings and with consequent differences in ore-forming fluids and ore-forming mechanisms. This paper discusses the nature and origin of the ironstone-related copper-gold association, with particular reference to the Tennant Creek district in northern Australia, and emphasises the two stage origin of mineralisation. The genetic model developed has significant implications for other ore-bearing environments.

Jan 1, 1990

By Li YQ, Yoon RH

The photoresponse of natural mineral pyrite and galena electrodes has been studied in aqueous solutions on freshly fractured surfaces (some fractured in situ) using moderate overvoltages. When fractured in situ, the electrode potential of both minerals instantly drops several tenths of a volt. Pyrite exhibits a spontaneous anodic photocurrent at fracture, which indicates upward band bending (depletion layer) on the n-type electrodes. The negative surface charge responsible for the depletion layer is believed to arise from a surface state that is intrinsic to pyrite surfaces, possibly modified by hydroxide ion adsorption. The flat band potential, Vty, of pyrite is at approximately -1.0 V (SCE) at pH 9.2 but to reach V(b, pyrite has to be cathodically decomposed. In contrast, galena assumes the flat band potential at fracture. Subsequent polarisation of galena at potentials positive of Vth produces an anodic photocurrent (upward band bending) and polarisation negative of Vro a cathodic photocurrent. Most of the galena electrodes studied were degenerate n-type and at potentials negative of Vtb, no cathodic photocurrents are observed, indicating the bands cannot be bent downward, consistent with the Fermi level at the surface merging with the conduction band edge near Vn,. Several were close to intrinsic, with the photocurrent changing sign near Vn,.

Jan 1, 1993

By Nemcik JA, Gale W. J

The magnitude and orientation of the in situ stressfield has been determined as a key factor in controlling the stability of openings for both coal mine development and extraction. Monitoring of roadway and pillar behaviour in conjunction with numerical analysis has identified specific stress control measures that have been successfully applied to improve productivity in underground coal mining operations. The influence of the major horizontal stressfield on the stability of openings has been identified at a wide range of sites including mines in Australia, Britain, the USA, New Zealand and Japan. Stress mapping, stress measurement and stress monitoring techniques have been applied to define the in situ stress regime and, together with the correlation of measurements of rock deformation and rock reinforcement behaviour, has enabled appropriate stress control measures to be developed. Stress control measures for development headings have required the selection of roadway orientation, roadway sequencing, reinforcement density and pillar size. The monitoring of behaviour at a range of trial sites has confirmed significant improvements in development stability. Stress control measures for longwall extraction have required rational selection of longwall orientation and extraction direction. Where a mine layout design incorporating the optimum longwall layout design has not been possible, the use of sacrificial roadways has been successfully applied to protect key roadways from stress concentration effects. Comprehensive in situ

Jan 1, 1992

Integrated geological interpretation of aerial photographs and Landsat images revealed that the gold mineralisation at Kyauk Pahto took place in a narrow zone of tension fractures within part of a previously unrecognised dextral wrench fault system. The latter is the wider manifestation of the well known Sagaing (right-lateral) strike-slip fault which has a length of more than 1 000 km, and it demarcates the boundary between the Indo-Australian and Asian plates. A subtle, right-handed kink affects the full width of the Sagaing Wrench-fault System at about the latitude of Kyauk pahto, and the extra dilatancy imposed on the tension fracture zone facilitated episodic ingress of mineralising hydrothermal fluids by "seismic pumping".

Jan 1, 1987

Through the stress synthesis of many exanples of macrostructure from eastern to western parts of China, it can be concluded that in Chinese tectonics there are two push- twisting forces, one of which ("Rl", from SE to NW) came from the Pacific Ocean plate thrusting over the Chinese continent plate in the eastern part of China; the other ("R2", from SW to NE) resulted from the thrusting of the India plate. The meeting line could be situated along the belt from 104¦-106¦E longitude in the middle part of China. To the south of Qinling the meeting line was an abyssal and giant fault situated from Longmen to the Er Inland Sea where the Tibet-Gansi fold system was intersqueezed to the Sidluan platform; to the north of Qinling, situated on the Fengxiang subsidence of the Eerduosi basin, the meeting line was located about 105¦E in Mongolia and appeared in the form of an arcuate structure, i.e. the Mongolia arcuate structure, which could extend to Beijiaer Lake towards the north. A new special type of structure named rotation-apart structure occurred from Gansu, Qinghai province to Ningxia, Neimeng autonomous region, in the west of Shandong province and in the western part of Chaidamu Basin.

Jan 1, 1987

By M D. Raine, J K. Bartlett, S C. Dominy

The Bendigo Goldfield is located in the state of Victoria, Australia, and is one of a number of fields that make up the Central Victorian Goldfield. Since 1993, Bendigo Mining NL has been carrying out an extensive re-evaluation of the goldfield and is currently well advanced in developing an underground mine. The sediment-hosted auriferous quartz reefs are structurally complex and closely associated with anticlinal axes and reverse faulting. The gold within the quartz reefs is characterised by both its coarse nature (100 up to >5000 ¦m) and its erratic distribution; as a result, the Bendigo Goldfield is classed as an extreme nugget effect system. This characteristic ensures that the Bendigo mineralisation is very challenging to sample and evaluate. In this case, diamond drilling is a good measure of structure and geological continuity, but a poor measure of grade and grade distribution. Drilling, combined with on-reef development, bulk sampling and detailed geological studies are required for resource evaluation. For grade control a balance has to be sought between sample representativness, cost, turn around time and practicality. Extreme nugget effect deposits like Bendigo, require the use of bulk samples in order to be representative. However, the cost and turn around time is significant. As a result, a micro-bulk sampling system has been introduced that attempts to overcome these problems by providing a fast, practical method that can define ore from waste. Bulk sampling and on-vein development are still necessary for resource and reserve estimation.

Jan 1, 2003

By G Solly, S Parsons, P Leevers, S Ni

Hematite ore has been mined in the Southern Middleback Ranges as blast furnace feed for the steelworks at Whyalla, by what was then BHP, since 1991. With the floating of BHP Integrated Steel as OneSteel in 2001 and the subsequent loss of key mining personnel, breakdowns in grade control methods and procedures occurred. These were compounded by changes in orebody characteristics and grade distribution leading to poor reserve recovery and poor grade control, both which resulted in loss of control of ore specification and severe deviation from the mining schedule. As a result, a total review of grade control, mining method and scheduling procedures was undertaken as well as a review of contractor quality assurance. It was found that sampling methods were poor and unrepresentative, grade control blocking processes were outdated and did not reflect the geology of the deposit, the ore mining method was outdated and did not respect the geometry of the orebody, reconciliation methods were being completed globally and were not reflective of individual areas of the deposit, and that deviations from the mine plan were occurring due to access to high-grade areas outside the schedule in order to bring stockpiles back on grade. Based on these findings, updated sample splitter systems were fitted to drill hole rigs, double bench sampling was implemented (2 + 4 m samples as opposed to 1 + 8 m sample) and a three-dimensional cellular grade control system introduced to better reflect the geology and orientation of the deposit. At the same time ore mining was changed from front end loaders mining on 8 m benches to excavator mining on three lifts on 8 m benches; this aided selectivity and respected the geometry of the orebodies aiding cleaner recovery. Reconciliation is now completed on a blast-by-blast basis, which allows for quick and easy identification of poorly performing areas of the resource model that can then be incorporated back into resource model updates. Regular daily and weekly meetings are also held with emphasis on the mining schedule to ensure no deviation from the schedule in the long term. These improvements have resulted in a lift in reserve recovery both globally and locally associated at the same time with improvement in grade.

Jan 1, 2005

By M I. Pownceby, J Manuel, S Suthers, A Poliakov, E Donskoi, N Ware

"The mining industry is constantly seeking methods for optimising the exploitation of mineral resources, such as iron ore, in a sustainable and eco-efficient manner for economic and social benefit. When faced with resources that have mineralogical or textural complexity, poorer grades or geological variability, it is particularly important to apply a comprehensive geometallurgical approach to the development of the mining operation, from resource definition through to processed product. Integral to this approach is the quantitative characterisation of representative samples across the whole orebody.A key factor that associates the geological characteristics of iron ores to their processing response is ore texture. While the classification of samples on the basis of chemical or physical properties alone is often not sufficient to predict their processing behaviour, classification and quantification of texture types can be effectively correlated to metallurgical performance in processes such as blending, beneficiation, pellet making, sintering and ironmaking. This paper describes some of the research conducted by CSIRO pertaining to the design and rationale of a texture-based classification system for Australian iron ores; some of the tools developed and used to apply such classification; and the application of the generated data to quantifying, modelling and predicting sintering behaviour and performance in the blast furnace.CITATION:Suthers, S, Pownceby, M I, Manuel, J, Ware, N, Donskoi, E and Poliakov, A, 2016. Geometallurgical Characterisation of Australian Iron Ores – from Ore to Processed Product, in Proceedings The Third AusIMM International Geometallurgy Conference (GeoMet) 2016, pp 351–360 (The Australasian Institute of Mining and Metallurgy: Melbourne)."

Jun 15, 2016

By Muchi I

To clarify the characteristics of non-uniform flow of gas through layered burdens of ore and coke, flow equations of continuity and motion were solved numerically taking account of the geometrical effect of a bed. The results are compared with that predicted by an approximate solution. A mathematical-kinetic model for determining the distributions of process variables over the region from the top to the melting level of blast furnace with the horizontal or the inclined layers of burdens is developed in consideration of the uneven gas flow. By the use of this model, the longitudinal and the radial distributions of process variables in an existing blast furnace in Japan were calculated with the aid of a digital computer. Computed results show that the process variables and the melting level change remarkably over the cross-section of the furnace and that the patterns of these distributions are similar to the results observed by other investigators.

Jan 1, 1975

By D Chalmers

Drill and blast mining techniques are used to mine coal outburst prone and high gas content coals in Australia. As the risk of an outburst is increased in high gas content coal, it is considered that drill and blast is the safest way to mine coals that refuse to yield gas even after considerable gas drainage efforts. The availability of suitable explosives for drill and blast in Australia is dependent on the æBuxtonÆ tests. With the diminished coal operations in England, the type and supply of such explosives is tenuous, causing significant disruption to the production of coal in some mines. In an attempt to free up the availability of explosives a review of the methods used for the testing of coal mine explosives was undertaken to ascertain the scientific basis for these tests. It was hoped that this comparison would allow regulators to have the confidence in explosives that had been tested elsewhere. Although this was not achieved, it has spawned a major research project that aims to better understand the mechanisms involved so that either confirmation of the current testing regime can be achieved or more suitable and scientifically based tests can be adopted. This paper looks at the ongoing research program investigating the parameters that are involved with methane ignitions from explosives. By utilising high-speed data capture and processing the research project is trying to view an ignition as it happens to identify the mechanisms involved so that better explosives can be designed for the coal mining industry.

Jan 1, 2007

By A Havenga, F, J Hough, R Grobbelaar

After the closure of mines they naturally start to fill up with water. As a result, hydraulic gradients develop between adjacent mines and different hydraulic water pressures are exerted onto peripheral areas or compartments within mines. This results in water flow between mines, or onto the surface. This flow is referred to as intermine flow. The geometry of the collieries in the central Witbank Coalfield make intermine flow possible in several areas. Since the Department of Water Affairs and Forestry declared intermine flow as one of its greatest concerns for granting closure to South African mines, much research into this phenomenon has been undertaken. The importance of the quantification lies in the DepartmentÆs goal of sustainable water management, and the possible long-term financial implications these flows could have for individual mines. The challenges of determining intermine flow are numerous, and attention is focused on the identification of areas where these flows can take place. Detailed mining plans on a regional scale are required, and detailed coal-seam floor data are combined to provide targets for possible interaction between these mines. The quantification of these flows is problematic due to the uncertainties regarding geometrical configurations and the variation in site-specific hydraulic characteristics of the coal and overlying lithilogical layers. To this end, use has been made of several analytic solutions and numerical flow modelling. Stage curves (volume versus elevation curves) for the opencast and underground workings have been calculated and first-order estimates for intermine flow or decant from these areas have been determined. The numerical modelling methodology entailed a down-scaling approach, starting with a broad regional model covering the entire area, followed by modelling the interactions between interconnected mines and finally looking in detail at the major areas of interaction. By considering different scenarios within the context of these flows, possible management options for intermine flow can be evaluated in terms of volumes and qualities that might emanate in the post-closure phases. Challenges for the future include obtaining site-specific information on the groundwater hydraulics and acceptable ranges for model parameters, and verifying model results over time.

Jan 1, 2003

By Teague JWS, Jung RF, Kadletz O, Jones DR, Jones CJ

A study of the passive treatment system for Hilton minewater revealed that it consists of four main process steps. The water, which contains soluble and particulate iron and zinc and is supersaturated with CO , is pumped to the surface and then passes throuqh a degassinq unit into settling ponds where larger particles of iron hydroxide, containing some adsorbed zinc, can settle out. The still turbid supernatant is pumped to the top of a hill, from where it is allowed to trickle in a broad sheet, releasing CO , towards an earthen channel in the valley below. Consequently the pArises from around 6 to 7.5. This in turn promotes further decrease in the concentration of soluble zinc by increased adsorption of zinc on to iron hydroxide particles. The channel then transports the water to the third process in the system - a 53 ML pond acting primarily as a flocculation-sedimentation basin. The clarified water overflows into a second channel, downstream of the pond, which constitutes the fourth process in the system. Here, copious quantities of filamentous algae further reduce zinc concentrations to acceptable levels before the water is collected for recycle into mine operations. The work reported here represents a detailed study of this system and describes the optimisation of the operation of each of the individual steps involved.

Jan 1, 1988

By Torres F, Cerda I, Gimenez P, Gonzalez G, Castelli L

Slurry density measurement, specially in cyclone feed streams, is fundamental to grinding circuit control. With the purpose of increasing the availability of such a measurement, the design of a soft-sensor in an actual grinding plant is addressed in this paper. The system is designed to substitute the momentary unavailability of the real density sensor, either because it has failed or because it has been removed for any other reason. The ARMAX model structure is determined using the stepwise regression method. Starting with a list of the candidate variables, the method selects, one by one, and in a systematic manner the variables which best model the measurement to be replaced, choosing them from the set of presumably correlated candidate variables. In this paper not only the single measurements are included in the list of candidates, but also combinations of such measurements having physical significance. A phenomenological model has been built with the purpose of determining these combinations. It turned out that in most of the cases the components of combined measurements having physical significance were selected by the stepwise regression method, instead of the single measurements. Actual grinding plant data is used to determine the model structure, to estimate the model parameters and to test the predictive capability of the developed soft-sensors. Good results are shown to be obtained during periods of up to 30 hours of the density soft-sensor operation.

Jan 1, 1995

The Olympic Dam copper-uranium-gold-silver deposit, located approximately 520 km north-northwest of Adelaide, South Australia (Figure 1) was discovered in 1975 by Western Mining Corporation. A joint venture to facilitate development of the deposit was formed in mid-1979 between WMC (51 per cent) and the British Petroleum Group (BP, 49 per cent). Production commenced in 1988, and current annual output from the mining and treatment of 2.2 Mt of ore is approximately 65 000 t of refined copper and 1400 t of uranium oxide. Significant amounts of refined gold and silver are also produced.

Jan 1, 1993