Search Documents

By L. A. Matthews

The embrittlement of the Ti-6Al-4V alloy by hydrogen dosing at elevated temperature has been investigated. Metal coupons were subjected to isothermal treatments in a partial hydrogen atmosphere in the temperature range from 650ºC?950ºC and the degree of embrittlement after treatment was determined by drop-weight impact testing. Embrittlement was found to be inversely related to reaction temperature as a result of the combined effects of greater hydrogen absorption and titanium hydride precipitation at the lower isothermal temperatures. Keywords Titanium, hydrogen embrittlement, impact energy

Jan 1, 2011

By C. C. Pretorius

3D seismic surveys have become a well-accepted method of imaging sub-surface geology in the Bushveld Basin. Because both the seismic sources and receivers are placed at surface, most surveys are limited to imaging faults with >=7 m throw. However, some areas such as shaft infrastructure can benefit from the ability to detect smaller structures prior to shaft sinking. Borehole radar and logs such as the acoustic televiewer can identify these structures in the shaft barrel, but not within -200 m from the shaft. Therefore, Anglo Platinum has developed the method of high-resolution 3D vertical seismic profiles (VSPs) for structural sterilization of shaft sites prior to shaft sinking. VSPs utilize a surface source and sub-surface receivers deployed down a borehole. This means that seismic energy goes through the near surface only once, and therefore suffers less frequency loss than when both source and receivers are on surface. Furthermore, VSPs can record 3 component (3C) data which comprises P wave, Sh (horizontal) and Sv (vertical) wavemodes. Because S waves travel at about 0.6 of the velocity of a P wave, shorter wavelengths and therefore higher resolutions are possible using S wave data. VSPs were first used in their one dimensional, zero offset, mode to correlate borehole geology with seismic stratigraphy within seismic surveys. This showed that VSPs record -30% higher frequencies than surface seismic surveys. The program then applied two dimensional VSPs to the problem of imaging geological structure close to a borehole. 2D VSPs confirmed that high frequency results could be achieved, and that this enabled a more accurate image of the Merensky Reef than could be achieved by surface seismic data. Finally, the program has achieved the successful recording and processing of 3D VSP data, with both compressional and shear wavemodes giving a high resolution structural result. Although P wave data gives -30% frequency improvement over surface seismic surveys, S wave data can give as much as 80% improvement, further improving the structural image achieved. The high fixed costs of VSP surveys mean that they should ideally be run in multi-client batches, which can reduce the per borehole cost by up to 50%. Their application is anticipated to be on high value capital infrastructure with a geological risk from faults which are not imaged by surface seismics-generally in the sub-7 m range. Keywords VSP, 3D seismic, bushveld, platinum, resolution, imaging, shear waves, shaft sinking.

Jan 1, 2011

By J. D. Steenkamp

"In July of 2016 and June 2017, major tap-hole repairs were carried out during the annual shutdown of two 63 MVA submerged arc furnaces producing ferrochromium in South Africa. The hot excavations of the tap-holes allowed for a study of the wear profiles. Subsequent thermodynamic calculations allowed for the quantification of the potential for chemical reaction between refractory and slag or alloy contributing to wear in the tap-hole area. Results from both wear profiling and thermodynamic calculations are reported here. IntroductionFerrochrome (FeCr) is an essential alloy in the production of stainless steel (Gasik, 2013). In 2017, 11.7 Mt of FeCr was produced (Pariser et al., 2018.). The main producers were China (39%), South Africa (29%), Kazakstan (12%), and India (8%).A number of grades of FeCr are available to producers of stainless steel, including high-carbon ferrochrome (HCFeCr) (Gasik, 2013). HCFeCr typically contains 60–70% Cr and 4–6% C (Basson and Daavittila, 2013). When the chromium content is lower (50–55% Cr) and the carbon content higher (6–8% C), it is referred to as charge chrome. Charge chrome is typically produced in South Africa due to the grades of ore available.FeCr is produced in submerged arc furnaces (SAFs) and in open arc furnaces (Mc Dougall, 2013) by carbothermic reduction of oxide raw materials. In a SAF the electrode tips are submerged in a porous charge mix. In FeCr production, the electrical energy is mainly liberated by resistive heating of the wet coke bed (Barker et al., 2007; Steenkamp et al., 2017). Thermally conductive lining design philosophies are typically applied in SAFs producing FeCr (Steenkamp, Denton, and Hayman, 2017; Coetzee, Duncanson, and Sylven, 2010; Coetzee et al., 2010).In July of 2016 and June of 2017, major tap-hole repairs were carried out during the annual shutdown of two 63 MVA SAFs producing FeCr in South Africa. The hot excavations of the tap-holes allowed for a study of the wear profiles. Both SAFs were operated on the same site, by two different operating crews, using the same raw materials. The major difference between them was that the SAF excavated in 2016 was an open SAF (referred to as Furnace A) and the one excavated in 2017 was a closed SAF (referred to as Furnace B). In an open SAF, a gap exists between the top of the steel shell and the roof, with the top of the burden being open to the atmosphere and the CO-rich off-gas being combusted as it escapes. In a closed SAF, the roof is attached to the steel shell and the CO-rich off-gas is collected from an offtake on the roof for further treatment and used in downstream processes.The paper presented here reports the macro-scale observations on refractory wear for the two SAFs, and studies the potential for chemical wear of refractory by slag and alloy as a potential contributing factor to the tap-hole wear observed."

Jun 1, 2019

By B. Genc

"The risk of spontaneous combustion is well known in the South African coal mining industry. In the coming years it is very possible that the incidence of spontaneous combustion will increase from current levels, due to factors such as an increased rate of mining, re-working of previously mined seams, more stooping and total extraction for underground mines, and higher stripping ratios for surface mines, leading to more spoils. It is also fairly certain that coal mining will face tougher environmental emissions legislation in the near future. To determine the areas where the risks of spontaneous combustion are high, it is necessary to improve on our current laboratory procedures for testing and evaluating coal samples, combining the results with site and field data, and if necessary revising the laboratory rating system to refine our understanding of South African conditions.Currently, laboratory tests are conducted in order to determine both the Wits-EHAC index and the crossing-point temperature which, when combined, give an indication of the spontaneous combustion propensities of the coal samples. This procedure has enabled the establishment of a database of results to review and evaluate South African coal seams. Using this database, the high-risk areas in terms of spontaneous combustion are identified. Tests have been undertaken for five consecutive years, between 2008 and 2012. In total, 119 coal samples from different coal seams and production coalfields have been analysed and classified through a series of laboratory tests. A comprehensive database of these results is available, and is continually being updated as new test results are added. This database will continue to expand, and to provide the basis for an improved risk evaluation methodology for spontaneous combustion"

Jan 1, 2015

The roasting of sulphide ores and concentrates is often the first step in the production of metals or chemicals. In many processes, the production of sulphuric acid is viewed as a by-product, whereas in some plants production is an important economic factor. Regardless of the purpose, a pyrite roasting plant consists of mainly three plant sections: roasting, gas cleaning and sulphuric acid. With the addition of air, the pyrite concentrates are transformed into solid oxides and gaseous sulphur dioxide at temperatures of 600?1000°C. After cleaning and cooling, the sulphur dioxide in the roasting gas is further processed to sulphuric acid. Two types of reactors are used depending on the application: stationary or circulating fluid bed. For over 60 years, Outotec has progressively been developing the principle of fluidized bed technology in several different reactor types for a multitude of process applications. The versatility of the fluidized bed reactor system has manifested itself in the treatment of minerals, including solid fuels, and for metallurgical processes both in the ferrous and non-ferrous fields. Process applications have included roasting, calcining, combustion and charring of coals, as well as off-gas treatment. This paper provides a summary of the pyrite roasting technology currently used along with a simple cost comparison of pyrite roasting and sulphur burning processes.

Jan 1, 2009

By P. Turcotte, E. Karampinos, F. Mercier-Langevin

In some underground hard-rock mines, squeezing compressive ground conditions are influenced by the presence of rock foliation and high stress. In these cases, the orientation of the foliation with respect to the drift direction has a considerable impact on the magnitude of the resulting deformation. Irrespective of the reinforcement and support strategy, keeping drives developed sub-parallel to the rock foliation operational is difficult, and often requires excessive rehabilitation during the lifetime of the excavation. This study uses field observations and convergence measurements at the LaRonde and Lapa mines of Agnico Eagle Mines Ltd to provide guidelines of the anticipated squeezing levels at these operations.

Jan 1, 2015

By F. X. Prins, L. R. Nelson

Tight control of electrode paste additions, hard and soft paste levels, baked zone position and electrode temperatures are integral to superior Söderberg electrode management practices. Paste addition is typically effected through briquettes, cylinders or large blocks. Each electrode paste form has its advantages and disadvantages, but probably a more crucial element from an electrode control perspective is the frequency and amount of electrode paste addition into a Söderberg electrode column. The detrimental effects of excessive soft paste levels leading to Stokes? Law segregation of paste in the electrode column are well documented, while several operators have effectively implemented the use of electrode thermocouples to monitor paste temperatures and automated slipping algorithms to control baking conditions within electrodes. Somewhat less well appreciated are the negative influences of inadequate soft paste levels, especially when coupled with excessive paste cylinder or block additions and resulting high hard paste levels. The alternative of insufficient hard paste levels and associated overheated electrode paste must also be considered, and can contribute to paste segregation directly. This condition also has the potential to generate gas mixtures in the electrode column that are concurrently a potential health hazard, while being prone to spontaneous combustion. Results of industrial trials on electric furnaces using Söderberg electrodes in a variety of applications from ferrochromium, to ferromanganese and even platinum-bearing base metal sulphides smelting are presented to illustrate the influence of paste loading practices on the dynamics of hard and soft paste levels. Possible electrode failure mechanisms attributable to inadequate paste level control will be discussed through some industrial examples. Suitable paste loading practices to mitigate the risks of inappropriate paste levels and associated electrode incidents are presented.

Jan 1, 2004

By J. P. Josien, Didier C.

ABSTRACT: France now faces the closure of most of its mining sites. Surface instability risk may sometimes remain, especially where large underground voids exist. This paper illustrates the importance of failure mechanisms identification in order to define properly how to manage the residual risk. RESUME: La France fait aujourd’hui face à la fermeture de ses exploitations minières. Des risques d’instabilités de surface peuvent persister, notamment lorsque des vides résiduels existent au sein des travaux souterrains. L’article illustre l’importance que présente la nature des mécanismes de rupture potentiels dans le choix des mesures compensatoires à adopter pour gérer le risque résiduel à long terme. ZUSAMMENFASSUNG: In Frankreich stellt sich derzeit das Problem der definitiven Schliessung des meisten Bergwerke. In manchen fällen ist die Stabilität der Oberfläche nicht garantiert, speziell wenn unter Tage nicht verfüllte Hohlräume verblieben sind. Dieser Bericht zeigt die Wichtigkeit, bei der Auswahl geeigneter Langzeit-Risikomanagementmethoden, der verschiedenen Mechanismen mit denen ein eventueller Bruch an die Tagesoberfläche geraten kann.

Jan 1, 2003

By E. L. Thyse

Keywords: Pyrometallurgy, converter, nickel, matte, distribution Metallic elements occupy a wide range in granulated nickel converter matte phases. The distribution of these metallic elements, particularly during crystallization path events, appears to be poorly understood. The present study serves to investigate the distribution of Fe, Co, Ni, Cu, and Pb between nickel-sulphide, copper-sulphide, and alloy phases. A novel combination of thermo-chemical modelling, electron microscopy, and electron probe microanalysis was applied to understand crystallization path events and the associated metallic element distribution. This study has demonstrated that major Ni incorporated during copper-sulphide crystallization would serve to mobilize minor Co towards distribution in copper-sulphide phases. In addition, solid solution substitution of Cu by Ni is manifested by compositional zoning of copper-sulphide phases. The role played by Ni during alloy crystallization suggests that minor Co was also mobilized, along with an increase in principal Ni. Fe, by contrast, seems to be incompatible with principal Ni. In addition, solid solution substitution of Ni with PGEs and Cu is manifested by compositional zoning of NiCu-alloy phases.

Jan 1, 2011

By A. Cichowicz

In this study, techniques used to design earthquake-resistant structures are applied to the deep-level mining environment. Earthquake engineers seek to identify the cause of potentially damaging resonance, and design or reinforce structures to reduce these effects. Similar techniques have been used to analyse the dynamic behaviour of the rock mass surrounding a stope in a deep South African gold mine. A three-dimensional seismic array installed in a stope at TauTona Mine (previously Western Deep Levels, East Mine) provided an opportunity to examine the evolution of the wave field close to an underground opening. A set of three-component geophones installed in a borehole above the panel revealed the development of surface waves as wave fronts reached the excavation, and an array of vertical component geophones fixed to the hangingwall provided a two-dimensional map of site effects. While the collected data do not include damaging seismic events, quantitative analysis enabled phenomena which could lead to damage, to be recognized. In the vicinity of the excavation, the energy of the seismic signal is generally transformed from high frequency to low frequency. It was found that seismic events with relatively high corner frequency (150 Hz?200 Hz) usually excite several modes of vibration (30 Hz to 110 Hz), while events with low corner frequency (30 Hz?50 Hz) do not always excite higher modes of vibration in the structure. This implies that the rock mass around the excavation is a complex medium, and should be studied using the multi-degree-of-freedom model. A strong structural effect (due to the excavation shape) was revealed as a low frequency coda wave (at 40 Hz and 60?70 Hz), well developed on the skin of the hangingwall. A strong site effect was revealed by changes in signal properties between hangingwall geophones, exhibiting strong resonances around 160 Hz and in the range 200 Hz to 300 Hz.

Jan 1, 2000

By N. Demirel, O. Gölbasi

Draglines are massive machines extensively utilized in opencast mines for overburden stripping. The demanding working environment induces fractures, wear and tear, and fatigue failures in dragline components and eventuates in extended maintenance, lengthy downtimes, and loss of production. The bucket is the main source of external loads on the machinery, since interactions with ground materials take place in this region. This study aims to develop a generic finite element model of the stress on an operating bucket. This entails (i) three-dimensional modelling of a dragline bucket, (ii) analytical estimation of resistive forces in the bucket movement, (iii) three-dimensional simulation of the moving bucket using finite element analysis (FEA), and (iv) sensitivity analysis to examine the effect of formation characteristics on stress variation. Simulation results imply that the drag hitch and digging teeth are the elements of the bucket that are most prone to failure. In addition, sensitivity analysis indicates that internal friction angle of the formation is the dominant parameter leading fluctuations in stress values. Changes in stress level are least influenced by formation density.

Jan 1, 2015

By M. Yavuz

One of the challenging problems for optimization in mining operations is to choose the best equipment among the alternatives. Equipment selection is an important task for mine management due to its operational cost, and is also an integral part of mine planning and design. Equipment selection is not a well-defined process because it involves the interaction of several subjective factors or criteria. Besides, decisions are often complicated and may even embody contradictions. Therefore, equipment selection is considered as a multi-criteria decision-making process, and suitable decision-making methods should be employed in this process. In this study, the loader selection for Aegean Lignite Colliery was made by using both the analytic hierarchy process and Yager’s method. Owing to the misusage of Yager’s method in the past, a new procedure is proposed in this paper for making proper decisions. It is highly recommended in this purposed method that criteria or alternatives should be grouped so as not to exceed the limitations of human performance (nine criteria/alternatives). The most appropriate solution for the loader selection was investigated by obeying this limitation for each method, and the results were compared. At the end of the decision-making process, a sensitivity analysis was applied for each method in order to see how a criterion affects the final decision. The advantages and disadvantages encountered during the application of each decision-making method are presented.

Jan 1, 2015

By C. -T. Mutale, A. S. Tuling

Titanium can exist in three different oxidation states. Its oxides are numerous and in presence of iron and oxygen, iron-titanium oxide compounds can also form. Some of the oxides are non-stoichiometric, making the determination of the thermodynamic properties of titanium and its oxides in molten steel and in slag difficult. Therefore, for the assessment of titanium activity in molten steel using the electrochemical measurements of oxygen partial pressure, the identification of oxides in equilibrium with titanium is important. This paper reports on an investigation of solid titanium oxide inclusions, which have been identified by combining different methods of analysis on quenched samples. These samples were taken from a Fe-Ti-melt at 1600°C, quenched in cold water then mounted and polished for optical microscopy. The oxide inclusions were analysed using transmission electron microscopy (TEM) with energy dispersive spectroscopy (EDS). Both single crystal and polycrystalline diffraction patterns were used to identify the oxides by comparing the d spacings with those in the ICDD X-Ray Diffraction (XRD) database. XRD was used to analyse the oxide particles on the top of slow cooled samples. It was found in the TEM study that the similarity of the diffraction patterns of the various phases meant that unique identification was difficult to achieve in some of the diffraction patterns. However, TiO, Ti2O3, Ti3O5, Ti4O7were identified in quenched samples, while TiO2, FeO.TiO2and iron oxides were seen in slow cooled samples. It was also confirmed that the type of oxides is related to titanium content of the samples. Keywords: Titanium, Titanium Oxides, Iron, TEM, XRD, thermodynamic.

Jan 1, 2004

By M. Berger

The ‘Metallurgical Operations’ Department of Grinaker-LTA Process Engineering (GLTA) currently undertakes then primary mining as well as the re-deposition of tailings forTicor South Africa (Ticor) at their Hillendale mine near Empangeni, KwaZulu-Natal, South Africa. The mining plant, which was designed by GLTA on a Build, Own, Operate and Transfer (BOOT) basis, is capable of mining 780 000 tons per month of ROM material which is then pumped to the Primary Wet Plant for treatment by Ticor. Hydraulic mining via high pressure water cannons and subsequent transfer of the heavy mineral slurry at an RD of between 1.35 and 1.48, has never before been undertaken successfully at this level of production on an in situ reserve. The presentation will take the form of a case study in which the unique site operating conditions will be detailed and discussed, as will the innovative process and engineering solutions employed by GLTA to overcome the engineering and production challenges encountered. In order to protect the intellectual property involved in this operation, this paper is fairly general in nature, and attempts merely to provide the reader with a broad perspective understanding of the operation, and some of the difficulties encountered during the early stages of production.

Jan 1, 2003

By S. J. Lötter, J. T. Nel, W. Purcell

"SynopsisGlow discharge optical emission spectroscopy (GD-OES) is an analytical technique used in the analysis of solid, conducting materials. Though primarily of interest as a depth profiling technique on samples with varying layers of both conducting and non-conducting materials, it is also capable of rapid bulk analysis of homogenous, solid samples. GD-OES combines the advantages of ICP-OES (wide detection range, speed, and lack of interferences) with the solid sampling of XRF techniques. This allows the analyst to not only quantify the elemental composition of a sample, but to evaluate it in terms of homogeneity with depth, a field in which auger electron spectroscopy (AES) and secondary ion mass spectrometry (SIMS) have traditionally been the primary techniques. Although GD-OES does not replace these useful techniques, it does offer various advantages over them, making it an excellent complementary analytical tool.In GD-OES analysis, a low-pressure glow discharge plasma is generated with the sample material acting as a cathode, accelerating the cations in the plasma towards the sample surface. This bombardment causes the sample material to ‘sputter’, essentially knocking free atoms or molecules of analyte material, which are then drawn into the plasma where they are excited. The light emitted by this excitation is then diffracted to separate the wavelengths emitted by the specific elements and detected by a spectrophotometer. The intensity of the signal is directly proportional to the quantity of analyte element present in the sample, allowing simple calibration and quantitative determination of the elements. Technological improvements made in the past twenty years or so have significantly increased the usefulness of GD-OES for surface analysis. Faster plasma stabilization and start-up allow for the quantification of surface layers as thin as 1 nm. Sputter rates have been accurately measured for many common materials, allowing them to be built into software libraries in the instrument’s control software. This dramatically expands the usefulness of this software and the ease of performing analyses.GD-OES analysis of nuclear materials allows for the rapid determination of the elemental composition without the requirement of initial dissolution. The thickness of any corrosion layers on nuclear materials can also be determined."

Jan 1, 2015

By G. van Hout, D. Rahal

Caving is a mass mining method that is normally applied to large, low grade orebodies because of its low production cost and high capacity. However, equipment breakdown and poor draw management can cause mining costs to exceed that of all other mining methods. One way to minimize the effect of disruptions in the mining cycle is to use Mixed Integer Linear Programming (MILP) for long-term scheduling. This paper describes a MILP goal program which has the dual objectives of minimizing deviation from the ideal draw profile while achieving a production target. Schedule optimization is performed using a life of mine approach in which all production periods are optimized simultaneously. The underlying assumption being that material mixing in the short-term (between monthly optimization cycles) has a minimal effect on the long-term state of the panel. Schedule optimization is carried out using six general constraint types to model and control production. The constraint types define: deviation from ideal practice, panel state, material flow conservation, production quality, material flow capacity and production control. These constraints define production targets and drive production toward the ideal profile while maintaining correct mining practice. Application of the model using production scheduling data for a De Beers kimberlite mine. It is shown how different production control constraints regulate production from individual drawpoints as well as recovery of the ideal panel profile by implementing an optimised draw schedule.

Jan 1, 2003

By F. M. L. Mulaudzi

Behaviour studies of metal dusting processes and associated filamentous carbon formation were undertaken on Alloys 600, 601, and 800 using a simulated metal dusting environment of 25C0-70H2-5H20 (vol.%) at 650°C. These samples were studied visually and by optical microscopy, scanning electron microscopy with energy-dispersive X-ray spectroscopy (SEM-EDX), transmission electron microscopy with energy-dispersive X-ray spectroscopy (TEM-EDX) and x-ray diffraction (XRD). Visual examination and SEM surface observation showed that Alloy 800 suffered metal dusting attack at an early exposure reaction period. The coke deposit amounts increased significantly with reaction time from 168 hours to 336 hours in Alloys 600, 601, and 800. Alloy 800 showed pitting after 168 hours' exposure, and the degree of pitting increased after 336 hours. XRD showed all these alloys had a common austenitic phase, with Alloy 800 also having ferritic metal particles. Alloy 800 after 168 hours' exposure had a complex phase mixture on the surface, consisting of ferritic Fe and austenitic FeNi phases. For coke deposit, there were also Fe304, Fe3C and graphitic carbon. The existence of Fe, Ni, and Cr metal particles, and also graphitic carbon in coke deposits, was confirmed by EDX analyses.

Jan 1, 2013

By D. R. Hack

The complexity of commercial mining software has gained rapidly over the past decade as hardware developments have surpassed software capabilities at a staggering rate. Software users, recognizing the limitations of off-the-shelf applications, have adapted embedded scripting languages. Custom programming has always had a vital role in mine computing innovations, yielding a dynamic research environment in industry and academia. Fields such as industrial minerals can benefit from customization of existing software, since the majority of mine planning software was designed for large metal mines. Different customization approaches utilizing embedded scripting languages are presented, with examples of sand and gravel applications in development. Following implementation by Surpac and development of the Surpac Command Language, Tcl has been widely applied. Mintec introduced Python, also extensible and portable amongst various operating systems, for use in their MineSight products. In early 2000, Gemcom introduced the use of Microsoft Visual Basic for Applications, widely used customization tools for a decade. Datamine implemented a flexible multi-language approach, with all program commands and processes available through the COM Server Interface. This extends customization from the existing macro language to scripting in HTML, JavaScript, or any language that supports COM.

Jan 1, 2003

By K. N. Karparov

The classical limit equilibrium methods are suspected to be ineffective in predicting the potential for highwall failure in many instances in coal mines around the world. Slope engineers have for many years recognized a block thrust failure mechanism for slope failures, but little work had been done before 2000 to explain the actual mechanisms which must be responsible for the failure. The main reason for this is that limit equilibrium methods implicitly assume rigid blocks, and the resulting force equations must be satisfied everywhere simultaneously for them to have any meaning. This paper will show that the material involved in the slope failure is not rigid; indeed it undergoes considerable permanent deformation during failure. This observation allows the authors to treat the block boundaries independently, because they need not maintain a constant spatial relationship with one another, as is assumed in other methods. To enable analysis of this type the authors assume that the weight of the blocks is evenly distributed. This is reasonable, because the slope material is not strong enough to be self-supporting without some sort of constraint, or strong enough to be able to apply point loads to the surrounding material. The purpose of this paper is to extend the practical application of the analytical method developed from mechanism studies to provide an objective assessment of the risk of slope failure, and therefore guidelines for more stable slope designs. This work is based on studies carried out on two failures in an open pit coal mine in South Africa, and provides a methodology to assess the potential for failure more objectively than is possible with currently accepted methods, while at the same time remaining sufficiently simple to allow a ?back of the cigarette box? assessment by geotechnical engineers on site.

Jan 1, 2009

By K. Coleman, C. Visser, N. Selby, X. W. Pan, T. Letwaba

A prototype weightometer soft sensor is under development at DebTech, De Beers in partnership with Venetia Mine. Its main function consists of an online data source; data validation; dynamic weightometer models; online detection of weightometer failure; alarming and replacing a weightometer when it fails. 31 dynamic models have been developed. All models with R2>= 85%are accepted as good models and they generally have satisfactory results with a daily error of less than 8% within 90% confidence level. By using those models, a set of soft sensors can be set up to cover all weightometers in the entire processes starting from primary crushing to the stock piles of dense medium separation (DMS). Some practical issues and concerns on model development are discussed, including data collection, data analysis, data pre-processing, variable selection, time delay, model training and model validation. When deploying the weightometer soft sensor system, the potential benefits could include: reduction in plant downtime; reduction in unplanned maintenance and even replacement of non-critical weightometers, particularly where nuclear weightometers are used. Keywords: soft sensor, weightometer, dynamic modelling, neural network modelling

Jan 1, 2003