Search Documents

Paper written on project work carried out in partial fulfilment of B.Sc. Eng. (Extractive Metallurgy)degree The concepts behind support vector machines are very interesting both in theory and in practice, as they are based on a universal constructive learning procedure derived from the statistical learning theory developed by Vapnik. In this paper, their application to time series modelling is considered by means of simulated data from an autocatalytic reactor. In particular, reconstruction methods from non-linear dynamics are used to define a state space model for the process. Multivariate embedding techniques are compared to scalar embedding with respect to modelling. Keywords: System identification, Support vector machines, Multilayer perceptrons, Non-linear dynamics

Jan 1, 2003

By M. Kumral

Raw material fed into a processing or refining plant is required to be uniform in composition for several reasons. When the mined ore is highly variable in quality the only way to ensure consistency is to homogenize the ore prior to the process. The problem is more complicated in the case of multiple ore sources. The idea behind the research is that theoretical blending ratios can serve not only to meet predefined specific criteria but also to reduce grade fluctuations of variables under consideration. In this paper, the problem is formulated as a quadratic programming problem, whose objective function is in quadratic form and constraints are linear. A case study was conducted on a data set from an iron orebody to demonstrate the technique. The objective was to minimize the blend variability in terms of each variable (in this study, iron, silica, alumina and lime) grade of ores extracted in three different production faces. The stockpile capacity, lower and upper limits of each variable satisfying operational requirements and non-negativity were constrained to the model. A modified simplex method developed by Wolfe was used for solving the blending and homogenization problem. The promising results can be used as apart of the stacking and reclaiming design. Keywords: blending and homogenization problem, quadratic programming, Wolfe method.

Jan 1, 2005

By R. Poisel, C. Scavia, Preh A., W. Roth, M. Pirulli

ABSTRACT: Two methodologies for the three-dimensional simulation of the run out of rock avalanches are presented: one based on a granular approach and the other based on a continuum equivalent approach. An evaluation of the possible run out area of the Oselitzenbach landslide (Carnic Alps, Austria) has been carried out applying the two methodologies. The combined use of these two approaches makes it possible a comprehensive assessment of the run out area. RESUME: Deux méthodologies pour la simulation tridimensionnelle de la propagation des avalanches rocheuses sont présentées: la première est basée sur un modèle de type granulaire et la deuxième sur un modèle du continu équivalent. Une évaluation de la propagation éventuelle de l’avalanche de Oselitzenbach (Alpes Carniche, Autriche) a été réalisée en utilisant les deux méthodologies. Les résultats obtenus montrent que la combinaison des deux approches rend possible une analyse plus complète de la phase de propagation. KURZFASSUNG: Für die dreidimensionale Modellierung eines Felsmassensturzes (sturzstrom) werden zwei Methoden, eine granulare und eine kontinuumsmechanische Methode, vorgestellt. Beide Methoden wurden zur Abschätzung der Ausbreitung (runout) der Massenbewegung Oselitzenbach (Karnische Alpen, Österreich) herangezogen. Die kombinierte Anwendung beider Methoden ermöglichte eine umfassende Beurteilung des möglichen Ausbreitungsbereichs.

Jan 1, 2003

By P. W. Lotz

Traditionally, bottle roll tests have been used to optimize the extraction process for best gold extraction and recovery. Such tests typically involve exposing the ore to defined reagent concentrations (lime, cyanide and sufficient oxygen) and balance the gold input with recoveries and losses. With mounting pressures to accommodate sustainability issues during development work, it became apparent that much more information would be required on the interactions and kinetic behaviour of solid and solution species during the entire leach process. This information is vital to enable proactive management of critical issues such as cyanide, arsenic, water and tailings management. Inadequate management of these aspects has led to temporary closures and, in cases of non-compliance, to the dynamically changing regulations. MINTEK?s advanced leach facility (ALF) is being developed with the aim of bridging the knowledge gap between the gold leach process and the predictable environmental impacts upon discharge. The facility aims to monitor all parameters of relevance such as pH, Eh, dissolved oxygen, metals, cyanide species, and viscosity during the entire process to enable more informed decisions on practical interventions. The presented set of data reflects first experiences gained from processing a refractory gold ore concentrate. Attempts at optimizing the process were limited due to the developmental nature of the work. Data relating to leaches with differing preoxidation conditions and cyanide staging is presented.

Jan 1, 2009

By D. Malovichko

Seismic events and pillar failures observed in two platinum mines in the Rustenburg area, South Africa, were investigated. We studied the sources of approximately 300 largest seismic events recorded in 2009. Moment tensors of these events were estimated from amplitudes and polarities of P- and S-waves. The inverted mechanisms contain isotropic implosive components in many instances. Analyses of uncertainty of the moment tensor solutions confirmed the significance of the implosive component in general. Several documented cases of pillar failure were considered in detail. For each of these a seismic moment tensor was calculated from an elastic stress model assuming an instant removal of a failed pillar. The geometries and magnitudes of the theoretical seismic moment tensors are in agreement with the observed (i.e. inverted from seismic data). The results indicate that either pillar failure or fracturing in the proximity of the reef (within a fraction of the dominant wavelength of seismic waves) are responsible for a significant portion of the large events recorded in these mines.

Jan 1, 2012

By D. Norval

Keywords: Pyrometallurgy, slag, metal, entrained, granulation, ferromanganese Water granulation of certain slags with high metal entrainment has customarily not been practised, not only due to the belief that such processes are too dangerous and unproven in the pyrometallurgical industry, but also that it might not be economically viable or environmentally acceptable. General slag granulation, on the other hand, is extensively applied, although in some cases crudely, with disastrous effects, as a result of not recognising the high heat capacities prevalent in slags. Thus, purpose-designed granulation and solidification processes are discussed, taking into account specific applications and the individual signatures of alloys and slags, emphasising that no slag, matte, or alloy responds in similar fashion to a granulation practice. This paper addresses three issues, namely: ? Water-granulation of BOF steel slag, without processing it to render it environmentally safe, physically and chemically acceptable for use by the construction industry, and the process economical ? Processing air-cooled BOF steel slag to change its physical and chemical properties to be acceptable for various uses and extracting metallic and bound iron ? Processing air-cooled or molten manganese slag to lower its manganese content to a level that is acceptable from an environmental and health point of view, and extracting FeMn-alloy of acceptable composition. The latter process culminated in a patented process to extract FeMn from air-cooled and molten manganese slag, and, at the same time, produce a hydraulic (cementitious) slag, thus transforming a possible hazardous by-product into virtually 100% saleable products.

Jan 1, 2011

By M. Hopf

The use of water-cooled copper components as sidewall elements of submerged arc furnaces and other pyrometallurgical applications has become increasingly common. The high cooling efficiency of copper has made it possible to increase the service life of furnace refractories when compared to conventional refractory linings. However, even the self-healing effect of ?freezing? worn areas of the lining does not necessarily protect the copper coolers against wear or corrosion completely. In order to avoid furnace explosions, various methods have been used to monitor the wear of coolers. Unfortunately, these have not been able consistently to identify localized wear. Thos Begbie (South African) and SAVEWAY (German) jointly developed technology to embed a line sensor into the copper cooler during the casting process. This system allows an exhaustive monitoring of the copper cooler. A SAVELINE sensor is placed inside the copper-cooling panel between the water-cooling passage and the hot face of the panel. The sensor follows the meandering shape of the cooling water channels. In the first instance, the sensor provides analogue information as the thickness of the copper reduces. Secondly, this wear signal will be confirmed by a signal of interruption if the sensor is washed away. Because of the structure of the sensor, a self-diagnosis is possible and indications can be confirmed (double proofed).Subsequent to the manufacture of some test pieces, coolers for a large submerged arc furnace were manufactured. The sensorized coolers have been installed at the intermittent surface (slag line) where the largest amount of corrosion has been encountered. The system has been operating since December 2005 in what is considered one of the largest submerged arc furnaces in South Africa. Keywords: Pyrometallurgy, furnace, copper coolers, freeze lining, monitoring, Saveway

Jan 1, 2006

By A. C. du Preez

Mintek has been involved in extensive test work since the early 1990s on the recovery of nickel and cobalt from leach liquors saturated in calcium, using synergistic solvent extraction systems. During this period the Nicksyn? reagent was developed, optimized, commercially manufactured, and tested by Tati Nickel on a demonstration plant for more than 2800 operating hours. Efficient recovery of nickel without the co-extraction of calcium, thus avoiding gypsum formation in the extraction and stripping circuits, was illustrated. This synergistic system was recently evaluated on a laboratory scale for the recovery of nickel and cobalt from synthetic lateritic sulphate leach liquor containing about 3 g/L nickel, 0.5 g/L cobalt, 0.7 g/L manganese, 20 g/L magnesium, and with calcium at saturation. Extraction and stripping parameters were determined for this feed liquor and are discussed in this paper.

Jan 1, 2014

By E. R. Davis, R. J. Knott

The paper discusses some of the tools of management in the field of project control, especially in engineering. Project control includes planning and cost engineering, and involves various phases: the conceptual Design and Planning Phase, the Definitive Design and Capital Budgeting Phase, and the Implementation Phase. Each of these phases is discussed in in, together with the use that can be made of data banks, progress reports, and computer systems.

Jan 1, 1982

By J. N. Van der Merwe, M. Mathey

"The South African squat coal pillar strength formula, developed by researchers from the Chamber of Mines Research Organization in the 1980s, predicts an exponential increase in coal pillar strength once a critical width-to-height ratio of 5 is exceeded. The arguments that have been proposed in favour of this formula are discussed critically in this paper.Field experience with squat pillars in the USA and evidence from published physical, analytical, and numerical model pillar studies corroborate the fact that a squat effect is unlikely to occur in coal pillars at width-to-height ratios less than 10. Furthermore, an exponential increase in peak strength of coal pillars does not exist. An alternative design criterion for squat pillars in South Africa is therefore suggested. IntroductionFive decades of coal pillar research in South Africa have produced a wealth of information on the mechanisms contributing to the strength and failure of in-situ coal pillars and accompanying phenomena. The relevant knowledge was sourced from, and for, a South African coal mining environment that remained largely unchanged in the second half of the 20th century. The depth of mining seldom exceeded 250 m, with mined seam thicknesses averaging 3 m and typical pillar width-to-height ratios (w/h) of 3–4.It is therefore not surprising that comparatively little research has been directed at the prediction of coal pillar performance in deeper deposits, where squat pillars of larger width-to-height ratios will be required to support the overlying strata. However, a thorough understanding of squat pillar performance will be necessary should the remaining deep coal deposits of the country be extracted in future, for instance in the Waterberg or Ermelo coalfields.A squat coal pillar strength formula has already been proposed in South Africa. The origin of this formula will be discussed in some detail in the following sections and its appropriateness will be critically reviewed against a great amount of evidence from relevant squat coal pillar research conducted internationally."

Jan 1, 2016

By S. E. Olsen, M. Tangstad

The main source of Mn in raw materials for SiMn production is Mn-ore and Mn-rich slag from the high carbon FeMn production. The amount of slag per tonne of SiMn metal is mainly determined by the ore/slag ratio. Increasing share of FeMn slag at expense of Mn-ore will lead to larger slag/metal ratio in the SiMn process. High volume of slag leads to an increased consumption of energy and probably to higher losses of metal inclusions in the final slag. Excavation of a SiMn furnace has shown that only modest pre-reduction of Mn-ore with CO gas seems to be obtained. Nearly all reduction of MnO was finished at the top of the cokebed. Dissolution and reduction of quartz obviously takes place in the cokebed zone after the main reduction of manganese oxide is finished. Probably the ?pick up? of Si in the metal is quite fast and takes place as the metal trickles down through the cokebed towards the metal bath. The distribution of Si between SiMn alloys and multicomponent MnO-SiO2-CaO-Al2O3-MgO slags is mainly determined by the process temperature, the silica content of the slag and its R-ratio = (CaO+MgO)/Al2O3. As an example, the equilibrium content of Si in the alloy will increase by about 6 % if the R-ratio is reduced from 2 to 1, provided constant temperature and silica content. The effect of temperature is also considerable. The equilibrium content of silicon will increase by approximately 6 % per 50°C in the temperature range 1550°C to 1700°C. The equilibrium content of MnO in SiMn slags depends first of all on the temperature and secondly on the silica content of the slag. At 1600°C the MnO content decreases from about 9 % at silica saturation to a minimum of about 3-4% when the silica content is reduced to about 40-45%.

Jan 1, 2004

By M. L. Bakker

Keywords: Pyrometallurgy, top submerged lance (TSL), nickel, PGM, calcium ferrite, lime ferrite The ISASMELT? top submerged lance (TSL) bath smelting process was developed in Mount Isa, Australia by Mount Isa Mines Limited (now a subsidiary of Xstrata plc) during the 1980s. By the end of 2011, the total installed capacity of the ISASMELT? technology will exceed 9 000 000 metric tons per year of feed materials in copper and lead smelters around the world. Commercial plants, operating in Belgium and Germany, are also batch converting copper materials in ISASMELT? furnaces. This TSL technology is equally effective for continuous converting processes, whereupon it is called ISACONVERT?. This paper presents the recently patented ISACONVERT? process for the continuous converting of nickel and Platinum Group Metal (PGM) containing mattes using the calcium ferrite slag system. The paper focuses on the potential application of the ISACONVERT? technology to existing nickel and PGM smelting complexes.

Jan 1, 2011

By T. Ertekin, X. Dong

This paper identifies the need for the development of specialized inverse solution techniques for the analysis of pressure transient data for the characterization of the transport and sorption properties of coal seams. Central to the proposed inverse solution technique is a specially structured artificial neural network (ANN). The most important transport and sorption characteristics of coal seams are butt and face cleat permeabilities, macropore porosity, Langmuir volume constant, Langmuir pressure constant, and sorption time constant. Conventional methods, such as type curve matching techniques and laboratory-derived measurements are handicapped in terms of being representative over the entire domain of interest. In this paper, we utilize analytical forward solution techniques for various combinations of characteristics of coal seams to generate training and testing patterns. These training patterns are shown to an ANN structured to learn the ubiquitous non-linear relationships that exist between coal seam properties, input conditions and the pressure transients measured. The trained networks are then extensively tested to ensure that they have effectively learned the training patterns. In the final stage of the development, the networks are run in the prediction mode. The responses of the networks are found to be in agreement with the input data. The developed networks are able to generate results generally within 5% error margin for permeability and porosity and 20% error margin for the sorption characteristics of coal seams. A number of examples highlighting the capabilities and degree of effectiveness of the network in characterizing the coal seams areincluded. Keywords: artificial neural networks (ANNs); coal seam characteristics; pressure transient analysis

Jan 1, 2003

By T. Y. Yunsel

Estimation techniques such as polygonal, triangular prism, trapezoid, isopach maps, and inverse distance methods are often used in ore or industrial minerals deposit evaluation. These techniques do not express the variability of the deposit and do not allow a determination of the reliability of the estimates. However, geostatistical methods can express a measure of the error associated with the estimates, by finding weighting coefficients for a given mining block, and can also help with data configuration that minimizes the error. This work addresses an application study on the quality and reserve characteristics of the cement raw materials of the Adana Cement Factory in Adana, Turkey, based on the spatial distribution and variability of the chemical components (Si02, CaO, A1203, Fe203). The study has been carried out using a geostatistical procedure that is useful for site assessment, characterization, and monitoring situations where data are collected spatially. Directional and omnidirectional experimental variograms of the cement raw material variables showed that neither strong geometric nor severe zonal anisotropy exists in the data. The most evident spatial dependence structure expressing the continuity for omnidirectional experimental variograms were characterized by exponential and spherical variogram models. These models have been used in cross¬-validation analysis, which proved that these models, their parameters, and kriging parameters are applicable for the study area. Quality contour maps of the deposits at given levels underground were estimated using a kriging interpolation technique. Anomalies such as bullseyes and drift were not observed in the maps that were generated. Kriged maps showed the spatial distribution of quality continuity and variability of the deposits. Grade-tonnage curves and total tonnage estimates in the particular grade were determined using ordinary kriging in order to improve the mining operation and planning. Consequently, local uncertainty and the probability of extreme values occurring are tools of prime importance for the mine planning, the optimum mix of raw materials coming from different quarry stopes.

Jan 1, 2012

By J. H. Zietsman

It has previously been observed that industrial high-titania slags display a compositional invariance, remaining close to stoichio-metric M3O5. The present study focuses on this invariance and the phenomena which may cause it. The invariance was studied using thermochemical calculations and a mechanism is proposed for the observed behaviour. The mechanism suggests that a combination of thermochemical equilibrium and transient slag solidification and melting results in the observed compositional invariance. The proposed mechanism provides new insights and prompts questions about other mechanisms in the smelting process.

Jan 1, 2004

By J. H. Zietsman

It has previously been observed that industrial high-titania slags display a compositional invariance, remaining close to stoichio-metric M3O5. The present study focuses on this invariance and the phenomena which may cause it. The invariance was studied using thermochemical calculations and a mechanism is proposed for the observed behaviour. The mechanism suggests that a combination of thermochemical equilibrium and transient slag solidification and melting results in the observed compositional invariance. The proposed mechanism provides new insights and prompts questions about other mechanisms in the smelting process.

Jan 1, 2004

By M. Jele

? Sulphuric acid plant in umbogintwini. ? Double absorption plant, capacity of 550MTPD. ? Build by simon curve during the year 1973. ? Side stream plants such as: ? Liquid SO2 Plant, ? Liquid SO3 and ? Oleum plant

Jan 1, 2009

By J. T. Nel

Non-transfer-arc linear plasma systems are highly suited to manufacturing nanoparticles. In this paper, the manufacturing of such particles with specially designed plasma systems is discussed. In general, it entails the volatilization of a metal halide that is injected in the high-temperature region of a plasma flame, where it is reacted with a suitable reagent, for example O2, to produce nanooxide powders. Metal powders can also be introduced directly in the plasma. Due to the very short residence time of only milliseconds in the reaction zone, as well as the fact that the reaction takes place in the gas phase, combined with very rapid quenching in the order of 10 000°C.s-1, nanoparticles are produced. At Necsa the manufacturing of such particles has been demonstrated on various scales. A pilot plant for the production of nano-Al2O3, and two pilot plants, one using SiCl4 and the other one using SiF4 as feed material, were built to produce fumed, pyrogenic, nano-SiO2. The alumina plant had a production rate of 8?10 kg.h-1 and the silica plants had production rates of 2 and 10 kg.h-1 respectively. Perfectly spherical alumina particles with a particle size distribution of between 10 and 200 nm were produced on a continuous system. Particle size analysis for the silica by SEM measurements showed a d50 of 60 nm and a particle size distribution of 30?100 nm. The specific surface area was in the range of 170m2.g-1. Other samples of plasma produced nanoparticles such ZnO and B4C will also be presented.

Jan 1, 2010

By B. Schoevaerts

The ever-increasing depth of mineral extraction presents a challenging environment for hard-rock underground mines. High in situ stresses and associated seismicity with potential rockburst hazards are the major decisive factors contributing to the choice of a ground support regime. Conventional ground support systems, designed primarily for static loads, are not always capable of providing safe working conditions for underground personnel in seismically active mines. Systems specifically developed to resist dynamic loading and allowing for larger deformations are therefore preferred alternatives. High-tensile chain-link mesh has a proven record of successful use in open cut operations in various rockfall barrier installations due to its high energy absorption capacity. It has also been used in underground operations in various parts of the world. Its application in South African underground mining, however, is limited due to the high labour intensity. This paper describes a method of mechanized installation of a chain-link mesh as evaluation at Gold Fields' South Deep Gold mine situated some 45 km south-west of Johannesburg. A purpose-built mechanized roll mesh handler, developed in Australia, was used in this trial. The handler is compatible with all standard multi-boom underground drill rigs and is operated utilizing the hydraulic circuit normally used for the feed arrangement. A number of key performance indicators have been specified as success criteria by Gold Fields: ? A 30% faster installation compared to the conventional methodology ? The mesh should sustain any blast damage when applied to the blasting face ? No unravelling of the mesh when cut ? No failure of mesh under normal conditions ? Reliability of the MESHA® installation handler (availability >90% ? Compatibility of the MESHA® installation handler with AC 282 boomers. Although the energy-absorbing mesh combined with yielding rockbolts is a ground support system of choice for rockburst-prone conditions, the operational upsides recognized during the trial imply that this product can also be competitively used in less demanding ground conditions where weld mesh or shotcrete are customarily used as a primary surface support.

Jan 1, 2012

By J. A. L. Napier

It has been established that significant time-dependent stope convergence may occur over time periods of hours and days in certain hard-rock gold and platinum mines. The source of this time-dependent behaviour appears to be associated with both pre-existing discontinuities and with mining-induced fractures that form near the stope face. These induced fractures may be associated with blasting processes and may also be formed in response to high stress concentrations in the unmined regions immediately ahead of the stope face. In shallower platinum mining operations, time-dependent behaviour is, however, observed to be much less marked unless some form of specific mining-induced fracturing occurs. One particular case of considerable interest is the time-dependent behaviour that is found to be associated with the formation and deployment of crush pillars. The purpose of the paper is to present a simple limit-equilibrium computational model of this behaviour that is sensitive to both the formation sequence and the size of planned crush pillars in a mine layout. This model provides a useful means to optimize the sizing of crush pillars, and at the same time may be used to identify potentially hazardous circumstances in which pillars may not crush in a stable manner.

Jan 1, 2012