Search Documents

By K. M. Godiwalla, S. Ranganathan

The thermal state in submerged-arc furnaces (S.A.F.) producing ferro alloys is important for the conditions of both the grade of alloy produced as well as specific energy consumption. The temperature profile in the S.A.F. is affected by the endothermic chemical reactions and the rate of gas evolution. The gas evolution commences at about 900 C due to the reduction of iron in the ore. A first-principle-based 2-D model has been developed to predict the temperature profile in the furnace. In this study, four ores of different grades have been considered in order to simulate the thermal profile in the S.A.F. The influence of the composition of the ore on the thermal profile and its significance in rationalisation of charge distribution to optimise the process are discussed.

Jan 1, 2007

By G. A. Harrison

The introduction of extra low profile (XLP) mechanized equipment within Anglo Platinum is aligned with the overall New Mining Technologies strategy to focus on continuing modernization of our mining operations` incorporating injury-free production and cost-effective mining technologies. The method allows for the introduction of a suite of ultra-low profile equipment in narrow reef (1.2 m) stoping operations. For the first time it is now possible to achieve total trackless mechanization at these widths and indications are that XLP mining has the potential to add more value when benchmarked with conventional mining and low profile (1.8 m) room and pillar methods. Various mine layouts were tested at different sites within Anglo Platinum and the XLP breast mining layout proved to be the most effective. This paper describes the results of the trials at Waterval Shaft and concludes with recommendations on our future strategy on implementation of XLP technology within Anglo Platinum.

Jan 1, 2006

By T. A. Ngcobo

Paper written on project work carried out in partial fulfilment of BSc Eng (Mining)degree The purpose of this report is to assess the risks faced by mines in dolomitic compartments in particular Kloof 8 shaft. The dolomites of the West Rand comprise one of South Africa?s most natural water resources, and overlie the reefs of the richest gold mines in the world. The pumping of water from the dolomitic aquifers to prevent flooding has caused the natural water table to subside, and the resulting cavities within the dolomite rock formations that overlie many gold deposits sometimes collapse, causing sinkholes. These impacts of the mining industry could worsen over time if preventative measures are not taken. This report discusses these impacts in detail, more specifically the risks to which operations in dolomitic compartments are exposed. A risk analysis was carried out to access the risks that the organization could face and it was this analysis that allowed one to examine the best methods to combat the problems. Other factors such as geological features and seismic events were also considered for the evaluation. To aid with the completion of this work various resources were utilized. A literature study was undertaken, information was gathered from various experts who specialize in dolomitic studies, evaluations were conducted and conclusions reached. The results from the project indicated that preventative measures should be implemented and correct procedures put in place to prevent any disasters.

Jan 1, 2006

By R. Jacobs, R. E. L. Pobe

The Moab Khotsong Metallurgical Complex comprises a gold recovery circuit (Noligwa Gold Plant) and a uranium recovery circuit (South Uranium Plant). The metallurgical complex employs the reverse leach technology for the recovery of uranium; specifically, uranium extraction prior to gold using acid leaching processing. The inherent uranium species in the ore are uraninite, coffinite and brannerite. The slurry that has been subjected to the uranium leach is further subjected to solid-liquid separation to separate the uranium in a solution through counter-current decantation and the solids (solids carrying all the gold), which are repulped and returned to the gold plant for gold processing. The uranium leaching process is critical to the uranium playing a pivotal role as a value adding by-product to the gold operation. The benefit of adopting this reverse leach process, for improved gold recovery, is that it exposes gold particles occluded within the uranium matrices and avails them for cyanidation. However, the drive for operational efficiency in uranium extraction must be weighed against the incremental operating cost against the backdrop of depressed spot uranium market prices. Over the years, efforts were directed at maintaining the benefit of extracting uranium to enhance gold recovery and simultaneously, optimising the operating cost model for the uranium operation to remain relevant to the gold operation. The aim of this paper and the corresponding presentation is to give an account of the uranium processing optimisation journey embarked on over the years; namely, the operational tradeoffs made in the process of ensuring that uranium continues to add value to the gold operation. These challenges were faced while striking a fine balance between processing cost and operational efficiencies against the prevailing depressed uranium spot market prices. Mineralogical- and laboratory tests conducted form the support basis for the results and observations articulated in the paper. Keywords: By-products, gold extraction, reverse leach benefit, uranium extraction

Jan 1, 2020

By P. L. Crouse

"SynopsisFluorine – in the form of hydrofluoric acid, anhydrous hydrogen fluoride, elemental gaseous fluorine, fluoropolymers, volatile inorganic fluorides, and more – has played, and still plays, a major role in the nuclear industry. In order to enrich uranium, the metal has to be in the gaseous state. While more exotic methods are known, the standard and most cost-competitive way of achieving this is by means of uranium hexafluoride (UF6). This compound sublimates at low temperatures, and the vapour is enriched using centrifugal processes. The industrial preparation of uranium hexafluoride requires both elemental fluorine gas and anhydrous hydrogen fluoride (HF). HF is prepared by the reaction of sulphuric acid with fluorspar (CaF2). Fluorine gas in turn is prepared by the electrolysis of HF. Yellowcake is first converted to uranium tetrafluoride (UF4), using HF, after which the compound is treated with fluorine to yield UF6. After enrichment, the UF6 is reduced to UO2 for use in fuel elements in pellet form.South Africa has the largest reserves of fluorspar internationally, and is the third largest producer after Mexico and China. Fluorine technology has many associated difficulties, because of the reactivity of fluorine and the toxicity of HF. The main barriers to entry into the fluorochemical industry are thus the abilities to produce both HF and F2. Both these substances are produced locally, at the industrial scale, at Pelchem SOC Ltd. Should South Africa contemplate developing its own nuclear fuel cycle as part of the awaited new-build nuclear project, it will be imperative to leverage the existing skills with respect to fluorine technology, resident at both Pelchem and Necsa, for this purpose. This paper summarizes the fluorochemical skills developed locally over the past several decades, and suggests strategies for maintaining the technology base and developing it for the next generation of scientists and engineers.IntroductionFluorine chemistry has always had a very close relationship with the nuclear industry. Although the first mention of fluorspar was recorded in the sixteenth century, and the fundamentals had been well-developed by the Second World War, it was only during the Manhattan Project (Banks et al., 1994), when almost unlimited funds were made available for the large-scale industrialization of fluorine and related compounds, that the technology took off. Expertise in both the chemistry and the engineering aspects of fluorine is essential for the design and running of several of the unit processes in the nuclear fuel cycle."

Jan 1, 2015

By J. E. Nesset

The bubble size, together with the gas rate, determines the bubble surface area flux in flotation machines and, therefore, impacts directly on the kinetics of flotation. Understanding the variables that influence bubble size is one of the keys to understanding flotation recovery. Surprisingly, these relationships are poorly understood, largely due to the difficulties associated with gas dispersion measurements in industrial environments. The McGill gas dispersion sensors (bubble size, gas rate, gas holdup) are capable of reliably making such measurements. This paper reports on a study, using the McGill sensors in a 0.8 m3 Metso RCSTM pilot cell, which investigated the effect on bubble size of frother type and concentration, gas rate, temperature and rotor speed. A Denver laboratory cell was used to investigate the effect of elevation above sea level. Frother concentration proved to have the predominant influence on bubble size, and the effect of frother type can be normalised as a function of the critical coalescence concentration (CCC). Measurements made at many plant sites worldwide are compared to the model developed using the Metso RCSTM unit.

Jan 1, 2007

By P. den Hoed, J. Nell

Tests were conducted on samples of rutile, titaniferous slag and ilmenite; each sample was mixed with coke and chlorinated in a bubbling fluidized bed at 1000°C. Profiles of the rates of chlorination of species were constructed from chemical compositions of the bed residues at different durations. From the start FeO and MnO are chlorinated quickly and all but entirely in a fraction of the time it takes to chlorinate all the TiO2. When the level of FeO is significant, as in slag and ilmenite, FeO is chlorinated faster than TiO2is. In these oxides, there is an inverse relationship between the rates of chlorination of FeO and TiO2during this first stage; the relationship, moreover, correlates linearly with the initial percentage of FeO in the oxide. We identify three stages in the chlorination process; a first, in slag and ilmenite, when FeO is being chlorinated; a second, when TiO2is most readily chlorinated; and a third, towards the end, when the levels of TiO2are reduced to those of other species. The rates at which individual species are chlorinated are analysed in terms of their appreciation or depreciation with respect toTiO2

Jan 1, 2003

By A. M. Bryukhanov, A. G. Mnukhin

The report states accident rates for gassy/dusty coal mines represented as an ergatic system "man – machine - environment" with further analysis by the method of conditional probabilities. A coal mine has been analyzed with allowance for the attending personnel, equipment and environment as a sequence of events and states. Probabilities of operational states have been determined and prediction of mine behavior has been made on the basis of investigation findings for equipment and the personnel states, and analysis of the environment (i.e. mine and mining conditions). Utilization of expert estimation methods has been considered for determining major operational parameters of gassy/dusty mines. The importance of standardized values of permissible occupational risks (permissible probability of accident, PA) is demonstrated. It is also stated that a procedure is essential to determine these values, involving regular check-ups of the integral safety criterion at all mines and further development of customized recommendations for individual mines on efficient improvement of accident situations. Methodology to determine the assessment criteria has been developed and scientifically substantiated for permissible accident rates in hazardous industries.

Jan 1, 2003

An effective electrowinning process in hydrometallurgical industry requires fluoride levels in the base metal solution to be less than 10 mg/l. Selective removal of the fluoride ions from base solution is thus desired, if fluoride was added to control calcium and magnesium in the circuit. Consequently, adsorption of fluoride onto activated alumina was studied in a batch and a column set-up. The effects of base metal solution pH, temperature, initial concentration and flow rate on activated alumina performance were investigated in either a batch or column configuration. A two-level factorial experimental design was implemented in studying column dynamics. Results demonstrate that activated alumina is an effective adsorbent for selective removal of fluoride from base solution. In the batch operation, fluoride was removed to values below the maximum allowable concentration (10 mg/[l]) when pH was = 8. In the column adsorption step at 55°C and 600mg/[l] initial concentration, up to 16 bed volumes were processed before breakthrough level was reached. Desorption step using 1% sodium hydroxide solution achieved an elution of 8 bed volumes. The activated alumina (AA) had a capacity of 8.65 gF/[l]AA at the 10mgF/[l] fluoride breakthrough level during the column adsorption test. Keywords: Fluoride; activated alumina; adsorption; recovery; regeneration.

Jan 1, 2009

By W. C. Joughin

Problems with the stability and performance of rockpasses in deep-level mines are common. This paper presents the results of an investigation into the recorded performance of rockpasses in deep-level gold mines. Records relevant to the geological environment in which the passes are excavated, their excavation details, and their historical performance are not at all well documented on mines. In general, the quality of data available from mines is poor. The investigation showed that very large spans frequently developed in passes during use, and that rock mass quality, the orientation of the pass with respect to the dip of the strata, and the magnitude of the stress acting normal to the axis of the pass were the main factors of influence. From the historical information obtained, the risks of instability in rockpasses are identified.

Jan 1, 2005

By A. T. Haile

Detailed monitoring of the response of the rock mass between rockbolt reinforcement units subjected to a simulated seismic source has been successfully conducted. This work has shown the increase in amplification of the Peak Particle Velocity (PPV) with distance from a rock bolt unit, and, a minimum PPV for rockburst damage for the given site characteristics. This understanding, at this site, will allow the design of a suitable rock bolt spacing to prevent unravelling of the rock mass between the rock bolt units for a given level of seismicity. Or, alternatively, an estimation of the requirement for a suitable fabric support system, for the anticipated level of seismicity, can be made.

Jan 1, 2001

By A. A. B. Douglas

In this paper the origins of the mortar intruded concrete concept and its development and application in the construction of watertight bulkheads in South Africa are described. The events leading to the commissioning and execution of the West Driefontein Experimental Plug and the Virginia/Merriespruit Boundary Plugs, and their roles in the proving of the technique, follow. The development of the first Transvaal and Orange Free State Chamber of Mines Code of Practice for Mortar Intrusion Concrete for Underground Plugs and Bulkheads, Johannesburg, November 1962,is explained. The saga of the October 1968 flood and subsequent salvage of the West Driefontein Gold Mine is recounted. The paper concludes with mention of the validation programme carried out for the Placer Dome Western Areas Joint Venture and their South Deep water barrier programme.

Jan 1, 2006

By C. Atilla Öztürk

Determining the cutability and mechanical properties of rocks is vital for the feasibility studies of a mine or tunnel projects. Many scientists have investigated the correlation between these parameters in order to predict specific energy (SE) and other important rock cutability properties for many years. On the other hand, textural properties of rocks have been quantified in detail by Howarth and Rowlands (1987). Hence, it was possible to correlate mechanical and textural properties of rocks by using a parameter known as the texture coefficient (TC). The main objective of this study was to use the test results taken from a research project carried out by Bilgin and Shahrair1tocorrelate the mechanical and cutability properties of rocks with textural properties of rocks. For this purpose, uniaxial compressive strength (UCS), uniaxial tensile strength (UTS), cone indent or (CI),point load index (PLI), unit volume weight (?), porosity, SE, and normal and cutting force values, were correlated with textural properties of rocks. Hence it was possible to make illuminating interpretations about the relationships between the mechanical and cutability with the textural properties of rocks. Keywords: Cutting properties, mechanical properties, specific energy, texture coefficient.

Jan 1, 2004

INFACON conjures up many things for many people. For the fortunate few who have been actively involved with this event since its inception, it brings back memories of far places, stimulating debates and above all, the people who have over the years contributed to the outstanding success of the Congress and who have added to the wealth of knowledge on all aspects of ferroalloys. The International Ferroalloy Congress (INFACON) has been held every three years since 1974. INFACON X will be held in Cape Town in February 2004. Simple arithmetic shows that next year?s Congress should in fact be INFACON XI. Thus, one Congress was skipped. INFACON 74, as the name implies, was held in 1974, in Johannesburg, South Africa. The event was sponsored by the Ferro Alloy Producers? Association of South Africa (FAPA), the National Institute for Metallurgy (NIM - Mintek as it is known today) and the South African Institute of Mining and Metallurgy (SAIMM). You will see some of the same sponsors involved in INFACON X. Dr. Robbie Robinson, renowned academic and past president of the SAIMM, was the Chairman of the Organising Committee. The editor was Dr. Helen Glen. Anyone fortunate enough to possess a copy of the proceedings of this first Congress will agree that from a literary and technical perspective, it is still the benchmark for subsequent events. A unique feature of INFACON 74 was the use of rapporteurs (my first duty) to ensure that the discussions following the presentations were recorded (in direct speech) as accurately as possible. This involved time-consuming correspondence (can you imagine life before e-mail?) to all parts of the world between the author and the person posing the question, with the rapporteurs acting as go-betweens. It is understandable that the much-awaited proceedings only saw the light in 1975. The names recorded in the proceedings as the members of all six committees read like the who?s-who of the South African ferroalloy industry at the time: Dr. Peter Jochens, Mr. Harvey Reid, Mr. Piet Streicher, Prof David Howat, Mr. Sigurd Selmer-Olsen - to name but a few. The Congress was held in the new, very fashionable and upmarket Carlton Hotel in the CBD. In the Foreword of the proceedings, Dr. Robinson provides an insight into the event. Three hundred and thirty six delegates from twenty two countries attended the Congress, which was opened by the Honourable P.G.J. Koornhof, Minister of Mines. Again, the opening address makes fascinating reading. During the Congress an ad hoc committee was convened as the first International Committee on the Organisation of Ferroalloys Congresses, with representation from the main ferroalloy associations of the world. The following decisions were recorded: ? The Congress would be continued at three-yearly intervals under the name of INFACON ? Every third Congress would be held in South Africa ? NIM (Mintek) would provide the permanent secretariat. Six technical sessions were held, each preceded by a plenary address with a closing address by the session chairman. In all, thirty technical papers were presented. Prior to the closing address, a panel discussion on factors affecting the future of the ferrochromium industry provided stimulating debate in the form of a question-and-answer session. The hosting country for the next INFACON would be announced. Eventually, word got round that the next INFACON would be held in Japan. However, due to unforeseen circumstances, the event was cancelled ? thus no INFACON 77. INFACON 80 was subsequently held in Lausanne, Switzerland from 12th to 16th October 1980. The event was organised by IFPEO, Institut des Producteurs de Ferro-Alliages d?Europe Occidentale. The Congress was officially opened by the President of IPFEO, Mr. R. Björklund, who was also President of the Organising Committee. The technical sessions were held in the Palais de Beaulieu. The Congress was well attended by 400 participants from 28 countries as well as 100 accompanying ladies. A total of four plenary addresses and 18 technical papers were presented. As with the first event, an international panel discussion (questions and answers) preceded the closing address. In his closing address, Dr. Louw Alberts, Chairman of the International Committee on Ferroalloys and President of NIM, presented an insight into the factors affecting growth in the industry and where growth was likely to take place. Twelve exhibitors had stands at the Congress. Attendees could choose from four post-event tours to Belguim/France (SADACEM/Compagnie Universelle d?Acetylene et d?Electro-Metallurgie), France (CUAEM), Finland (Outokumpu Oy) and Norway (Elkem Fiskaa plant).

Jan 1, 2004

By J. P. Theron

Paper written on project work carried out in partial fulfilment of Phd.(Process Engineering) degree Process modelling is an essential element in the development of advanced (model-based) process control systems, accounting for up to 80% of the cost of development. Often, models based on historic process data are the only viable option when dealing with processes that cannot be modeled cost-effectively from first principles. Despite major advances that have recently been made in the field of nonlinear process modeling, some basic idea of the dynamic behaviour of a process system is important prior to system identification. While many process systems seem unlikely to be linear, the possible nonlinear aspects of their dynamics are not necessarily supported by the data. This may have significant implications for subsequent system identification or interpretation of the data. For example, it is quite possible that an underlying nonlinear deterministic signal may be obscured by a high degree of measurement or process noise, to the extent that it may be critical in the choice of model to fit to the data. The method of surrogate data analysis has recently been proposed as a means to classify the underlying dynamics of complex systems, and in this paper the use of Fourier transform surrogates to classify process systems is explored. Keywords: modelling, process control, simulation, surrogate data analysis, nonlinear systems.

Jan 1, 2004

By M. Moys

The use of temperature measurements for unusual applications in the analysis and control of column flotation and milling is discussed. This is followed by a review of some of the authors current research in the application of the discrete element method to the modelling of particulate processes, specifically milling, fluidized beds and chute flow.

Jan 1, 2005

By N M. Vegter, C B. Kohler, E H. O Meyer

Zincor operates four Lurgi fluidised-bed roasters at its roast-leach-electrowin refinery in Springs, South Africa. In 2006, Zincor?s four roasters had reached ages ranging between 26 and 38 years, while the expected life of this type of roaster is generally 20 to 25 years. Based on its condition, Roaster 3 was the first to be rebuilt in 2006. Roaster 4 is being rebuilt in 2007 and the last two roasters will be rebuilt over the next few years as necessary. The main damage to Roaster 3 was caused by buckling of the roaster steel shell. This caused the freeboard and dome areas to shift, resulting in cracks in the dome brick lining and crumbling of the bricks lining the freeboard. The steel shell buckled due to the combined effects of too-high operating temperatures and corrosion caused by the following two main factors: ? Thicker external insulation was installed during a previous maintenance shut which resulted in and increased steel shell temperature and exacerbated the normal metal fatigue; ? SO2/SO3 containing gas that had permeated through the brick lining, passed through its dew point during frequent planned and unplanned shuts over the life of the roaster. This caused acid formation between the refractory brick lining and steel shell, resulting in corrosion of the shell. The initial plan was to replace only the damaged areas of Roaster 3, but upon dismantling more damage was found and the roaster was rebuilt from the tuyeres upwards. During the rebuild, the opportunity was taken to modify the design of some of the brickwork to extend its life and make future repairs easier. This paper discusses the mode of failure of Roaster 3 and the lessons learnt during its dismantling and rebuilding.

Jan 1, 2007

The Annual General Meeting of the Institute was held in Kelvin House, Johannesburg, on Wednesday, 30th August, 1972. Professor D. D. Howat (President) was in the chair. There were also present sixty-two Fellows, twenty-four Members, two Associates, three Students and eighteen Visitors, making a total of one hundred and nine. The President declared the meeting open at 4.05 p.m. OBITUARIES The President: It is my sad duty to announce the death of six members of this Institute. The first of these is Dr A. J. Orenstein, Honorary Life Fellow, who joined the Institute in 1916 and died on Seventh July. Dr Orenstein became almost a legendary figure in Medicine as a result of his work in the control and elimination of malaria during the digging of the Panama Canal and I believe he was actually engaged working until four days before his death. The other members who have died are: F. Meyer, died on 8th June; S. D. Loxton, Fellow, died on 13th May; J. Innes, Fellow, died on l0th January, 1970; O. B. Prentis, a Member, died on 4th June, and J. Brits, Associate, died on 27th June. As a mark of respect to the memory of the deceased and in sympathy with the bereaved may I ask you to rise and observe a few moments silence. MINUTES The President: 'The second item on the Agenda, the minutes of the meetings held on March 22nd and May 10th, 1972, as tabled. May I confirm these minutes'? Agreed. WELCOME TO VISITORS 'On behalf of the Council, may I say how delighted we are to have such a splendid gathering with us on this our Annual General Meeting. Dames en here vir die omvang van my Afrikaanse woordeskat verdien ek weinig lof, boonop is dit nog Afrikaans met 'n Skotse aksent, wat u asseblief moet verskoon. Tog wil ek se hoe dankbaar ons is dat sy Edele, die Minister van Mynwese, dr Koornhof en mev Koornhof vandag hier teenwoordig kan wees. In addition to our own newly appointed Minister of Mines, we are also honoured by the presence of the Minister of Mines of Rhodesia, Mr I. B. Dillon. I think not even our oldest members can recall an occasion on which two Ministers of Mines were present at a meeting of the Institute. It is a particular pleasure to welcome them this afternoon. I am happy to say that Minister Koornhof is an Honorary-Vice President of the Institute, and Minister Dillon is an Honorary Member. The presence of your goodselves, gentlemen, with us this afternoon, I am sure underlines the great significance which we believe both countries attach to the economic exploitation of the vast mineral resources of Southern Africa. For over 70 years, this Institute and its members have been making their contribution to this great task of mineral exploitation, and we feel that your presence with us today, is a token of the confidence of your governments in what this Institute and its members have been endeavouring to do. Also present with us today, is Mr R. C. J. Goode, President of the Chamber of Mines, and Honorary President of this Institute. I was going to say I welcome Mr Goode, but it is rather difficult to welcome a man to his own home, and this is the case with Mr Goode. He is a very distinguished and a very recent past president of this Institute and he remains now, as he has been for as long as I can remember, one of our most active and valuable members. This I think, ladies and gentlemen, is the second rather unique feature of this occasion this afternoon, because no one else can recollect an occasion on which the President of the Chamber of Mines was also a past president and an active member of our Institute. A special word of welcome, Mr Goode. Mr Tommy Gibbs, our Government Mining Engineer, is in a somewhat similar position as he is also a member and an old and valued friend of our Institute, - welcome to you, Mr Gibbs. Mention I think, must be made of the fact that we are very glad to have with us Dr A. J. A. Roux, President of the Atomic Energy Board, and Mr Stanley Craib, President of the Associated Scientific and Technical Societies. Then, within the great family of the A.S. and T.S., we have the presidents, or the vice-presidents of I believe 13 of the constituent societies. These gentlemen, like all the rest of us, have the rather tiring task of supporting and upholding their fellow presidents on occasions such as this. We particularly appreciate their presence this afternoon and the effort they have made to be with us. We are happy to have with us: Dr R. E. Robinson, Director-General, National Institute of Metallurgy, Mr E. Boden, Manager, Associated Scientific and Technical Societies, Dr H. G. Denkhaus, President, The S.A. Institute of Mechanical Engineers, Mr G. Goedhals, Vice-President, S.A. Institute of Electrical Engineers, Mr E. Dalton, President" S.A. Institute of Certificated

By R. G. Tarr

During an intense and accurate study of initiation systems used on an AngloGold mine, Mine A, poor drilling practices were revealed. Graphic results of all actual measurements taken are shown below. [ ] Accurate drilling of shot holes will determine whether a mining operation is payable or not and at what pay limit mining can take place. Losing 30 cm of a potential 120 cm-hole length will result in a number of consequences, i.e.: ?Higher stoping costs ?Drilling of one shot hole twice in every five shifts ?Increased explosive cost as a result ?Excess stress on hangingwall and footwall due to bad face shape ?Excessive blast damage before the area is worked out. No wonder we have such a high-risk operation. Drilling accuracy is crucial to the profitability on any mine as profits are directly related to the quality and utilization of the drilled shot holes.

Jan 1, 2000

By B. Samanta

This paper investigates the performance of the neural network technique in a placer deposit situated at the Nome district in Alaska. Initially a single neural network model was constructed to estimate the gold grade. However, at a later stage, an ensemble model consisting of multiple networks was also constructed via the Adaboost algorithm. The reason behind the use of the ensemble model of the Adaboost algorithm was to examine if the ensemble model provided superior performance to the single neural network. As indicated by other studies, Adaboost did not have a better performance than a single neural network in this specific application. This may be due to the high noise inherent in the gold data used in this project. Consequently R2 values of gold prediction are also poor for both the single and multiple networks.

Jan 1, 2005