Search Documents

By A. K. K. Rompel, G. K. Chunnett

Introduction Structure appears to be of secondary importance to the Bushveld Complex when viewed from a distance. However, a closer look already reveals that large fault zones bound the Bushveld Complex. These fault zones are deep-seated crustal lineaments of continental magnitude such as the Thabazimbi-Murchison Lineament, the Johannesburg-Barberton Lineament, the Palala Fault zone and the Rustenburg fault. Furthermore, the Bushveld Complex is transected by large faults such as the Vlakfontein Fault, visible in the Pilanesberg, the Steelport Fault and the Laersdfrift Fault in the Eastern Bushveld, the Brits Graben, and a multitude of dyke generations (Figure 1). When pursuing structures on a project scale, the lack of outcrop, particularly in the western part of the Bushveld, is of no help for the exploration geologist. Consequently, the first dataset that needs to be considered is the Landsatimagery. Older Landsat TM 5 or newer Landsat ETM 7 will immediately reveal areas of outcrop for field visits, and, after further processing, indicate the geological units, in particular the platiniferous Critical and the Main Zone of the Rustenburg Layered Suite (Figure 2). Airborne magnetic surveys are the subsequent dataset necessary for any decent interpretation of the geology under the ?Black Turf? cover. Currently Anglo Platinum acquires aeromagnetic surveys with a flight height of 20 m and a line spacing of 50 m. These result in high-resolution imagery absolutely crucial for dyke delineation and dip modelling, stratigraphic layer detection, IRUP detection and possibly pothole discovery and fault delineation (Figures 3 and 4). However, the aeromagnetic surveys and the Landsat imagery reveal structures very much confined to the surface and may be hundreds of metres above the platiniferous reef horizons. This may lead to inaccuracies. When determining the underground structures for shaft sinking and mine planning, a 3D seismic survey is advisable. Too many shafts have been sunk in the Witwatersrand Basin for gold and in the Bushveld for platinum without sufficient information of the structure beneath. Recent 3D seismic surveys in the Bushveld have proven the technology and visualized the reef horizons and led to an interpretation of the underground structure (Figures 5 and 6). [ ]

Jan 1, 2004

By M J. Daniel

The diamond industry is no newcomer to High Pressure Grinding Rolls (HPGR) as it has used the technology for more than 20 years, predominantly in secondary crushing and recrushing roles. In fact it could be argued that the diamond industry has led the way for the wider minerals industry to consider its application. In existing conventional secondary cone crushing applications, large valuable diamonds are" won" through meticulously managing the recovery process within defined particle size ranges. Conventional crushers operate with relatively large closed side settings, but have the potential to damage diamonds by making direct contact with the sides of the crusher. Cone crushers also result in steep product size distributions that run a high risk of losing many of the smaller, unliberated but more abundant diamonds from the rock. This current comminution paradigm of particle size reduction management results in recrushing processing applications or plants where HPGRs are often used.

Jan 1, 2010

By I. O. Flores, A. Anani, H. Li

Current technological advancement has increased the exploitation of deeper ore bodies using underground methods. In certain cases, the mineral deposit starts near the surface and continues deep within the earth's crust. Such deposits are best mined with both underground and surface methods. Two main techniques are used to determine the optimal transition depth from surface to underground - a disintegrated approach and an integrated approach. Both methods have limitations in their use to determine the optimal transition depth. The objective of this research is to investigate the conditions in which an integrated approach to the transition depth problem is superior to a disintegrated approach. Three binary integer linear programming models are developed for this purpose: a surface and underground model for the disintegrated instance and an integrated model. A case study is performed to validate the models, with a comparative sensitivity analysis to obtain operational insights.

Sep 1, 2021

By R. J. Davidson, W. D. Douglas, J. A. Tumilty

SYNOPSIS Two pilot plants operating on different types of carbon were used to generate steady-state adsorption data in the treatment of gold-plant solutions of high and low tenor. A pilot plant comprising three 4 m long columns and using 3 mm extruded activated carbon was used to treat pregnant solutions with a gold concentration of 5 to 8 g/t at pH 10 to 11,5. Column influent was pumped at a super¬ficial flow velocity of 0,5 m/min while the carbon was moved daily so as to effect gold loadings on the carbon of 5 to 10 k /t. With a carbon-bed depth of 12 m, an adsorption efficiency of 99,9 per cent was obtained in a bed depth of 9 m, which was subsequently reduced to 8 m when the depth of the carbon bed was reduced to 8 m. This improvement in efficiency was attributed to the reduced residence time of the carbon from 12 to 8 days. A corresponding decrease in adsorbed silica indicated that some silica fouling of the carbon had occurred. A second pilot plant, comprising a single column 6 m in length and using coconut-shell carbon of 1,0 to 2,4 mm, treated a solution with a gold concentration of 0,7 g/t at pH 8,0. Feed solution was pumped at a superficial flow velocity of 1,0 m/min while carbon was moved every 8 hours to give gold loadings of more than 5 kg/t. An adsorp¬tion efficiency of 99,5 per cent was indicated at steady state. Intermittent washing of the column with dilute hydrochloric acid was shown to have a beneficial effect on the gold adsorption and to reduce the pressure drop across the column. Intermittent washing with dilute sulphuric acid indicated that there had been some passivation of the car¬bon during the treatment of a solution with a gold concentration of only 0,2 g/t. Compared with other adsorption systems, the countercurrent moving bed appears to offer significant advantages in terms of metallurgical efficiency and capital investment. The inability normally associated with moving-bed systems to treat solutions containing suspended solids may be overcome to a degree by the use of relatively coarse extrud¬ed carbon.

Jan 1, 1985

By W. Ferreira

Steel is produced in a basic oxygen furnace from hot metal obtained from a blast furnace. A sintered iron ore with good hightemperature properties (strength and permeability) should be used as feed to the blast furnace. The quality of this sintered ore depends on the reactivity of the iron ore used as feed to the sinter plant during the lime assimilation step in the sintering process. The penetration test is the standard method for evaluating the reactivity of iron ore with lime. It is, however, difficult to determine the exact depth of penetration from the standard test. A new test method is proposed that allows automatic evaluation of iron ores in terms of lime assimilation with increasing temperature. A comparison of the coefficients of variation for the new and standard methods for each ore type demonstrates that the results of the new test are more reproducible and more precise than those of the standard method. The test is also less time-consuming and easier to implement.

Jan 1, 2014

By B. Aydin, A. S. Aloglu, S. Kahraman, E. Saygin

"The coal strength parameter is used in operations such as the selection and design of mechanical excavators and coal crushing/grinding equipment. However, coring or cutting prismatic specimens from coal for the standard strength tests is difficult, and sometimes not possible. For this reason a method for the indirect determination of coal strength from easy tests would be useful. In this study, the predictability of coal strength from the needle penetration test was investigated. This is a very easy and practical test method. Point load and needle penetration index (NPI) tests were carried out on coal specimens from Çayirhan coalfield in Turkey. The point load indexes were converted to the uniaxial compressive strength (UCS) values by using the general conversion factor. The UCS and NPI values were evaluated using regression analysis. A strong linear relationship was found between the two parameters. The conversion factor for the UCS-NPI ratio is 0.35 for the tested coal samples. It is concluded that the UCS of coal can be predicted from the NPI.IntroductionKnowledge of coal strength is important for the selection and design of mechanical excavators and coal crushing and grinding equipment. Uniaxial compressive strength of coal is used in designing pillar dimensions in room and pillar mining. Coals always contain bedding planes and cleats; they are also brittle and fragile. Therefore, coring or cutting prismatic specimens from coal for the standard strength tests is difficult, and sometimes not possible. For this reason, a method for the estimation of coal strength from easy tests would be useful.The needle penetration test is a nondestructive index test that is applicable both in the field and the laboratory and does not require any special sample preparation. The test has been used for the estimation of physical and mechanical properties of weak or soft rocks (Okada et al., 1985; Yamaguchi et al., 1997; Takahashi, Noto, and Yokokawa, 1988; Uchida et al., 2004; Aydan et al., 2006; Aydan, Watanabe, and Tokashiki, 2008; Aydan, 2012; Erguler and Ulusay, 2007; Park, Obara, and Kan, 2011; Ulusay and Erguler, 2012; Aydan and Ulusay, 2013; Aydan. Sato, and Yagi, 2014). This study investigates the predictability of coal strength from the needle penetration test."

Jan 1, 2017

In China, thin coal seams are rich in resources but are technically challenging. The mineable reserves in these seams account for 20.4% of the total coal resources, while the current production accounts for only 10.4% of the total annual production. Characteristics such as narrow mining space, low level of mechanization, poor working environment, and high cost of mining, restrict the development of mining safety and efficiency. Recently, fully mechanized mining technology has developed rapidly for thin coal seams, the level of yield and efficiency has reached or exceeded the international standard, and some state-owned key coal mines are considering automation of their mining process. In thin coal seams with hard stone bands that contain concentrations of pyrite, a specialized software, LS-DYNA, is used to calculate the rational blasting parameters that are used in the deep-hole pre-splitting blasting. Using this method the hard stone bands are fractured effectively, and hence increasing the coal productivity. In addition, mining advance rate were increased by enhancing the level of fully mechanized equipment and safety improved by increasing gas drainage from the gas outburst prone seam located some 7 m below the coal horizon. At present, thin coal seam mining technology faces many challenges, including the low level of equipment automation, the low advance rate in mixed coal-rock ground, and the large number of the mine personnel underground. By lowering the labour intensity and improving efficiency through automation and other measures, more efficient working faces can be implemented in thin coal seams.

Jan 1, 2012

By Thermo Electron, Tim Sennett

Summary ?SamStat Development ?SamStat Description ?Accuracy ?Availability ?Cost of ownership ?Capital ?Installation ?Operating/Maintenance ?Case Studies ?Modikwa ?Combined concentrate sampling ?Cerro Verde ?State of the art sampling & analysis ?Ok Tedi Final Tails ?brownfields installation ?Conclusion

Jan 1, 2005

By I. M. Morsi

The present study investigates the feasibility of using some materials as start-up slags for the extraction of magnesium metal from Egyptian dolomite ores by the Magnethermic technique. The reduction of roasted dolomite (dolime) takes place in a closed evacuated electric arc furnace (small Magnetherm reactor) using ferrosilicon as a reducing agent and liquid start-up slag as a reduction bath under an inert atmosphere. The operational parameters affecting the stabilization of the start-up slag composition and the extraction of magnesium such as reducing agent stoichiometry, reduction temperature, and time were investigated. The dolomitic charge was in the form of sintered briquettes. The charge was reduced under an argon atmosphere in the Magnetherm reactor for constant period and constant temperature. The results showed that the clinker can be used as a start-up slag for the magnethermic process. Also, the optimum conditions for the extraction process were found at 2XSi stoichiometry to give maximum reduction of 80% after 15 minutes reduction time at a temperature of 1550°C. XRD and XRF measurements were used to determine the phases present and the extent of reduction of magnesium, respectively. Keywords magnesium, dolomite ore, magnetherm, magnetherm reactor, silicothermic reduction.

Jun 1, 2013

By A. W. S. Schumann

G. H. HENDERSON INTRODUCTION It is a pleasure to have read Mr Schuman's wide ranging and highly informative paper and also to congratulate him. Before proceeding to give views on the use and preservation of coal in South Africa, it is perhaps apposite to look at the South African scene on the world canvas. Mr Schuman's figures in his Table I are highly revealing. The Table shows expected consumption, export and import demand in the year 1980 by the continents of the world. The Republic of South Africa is shown as one of only four exporting countries at that date, although in the interim, Australia has become an exporter and will continue to grow as an exporter of coal. Huge fields have been proved in recent years. There is reason to think that for some years, coal and its economical uses have been under-estimated. The importation of oil carries a constant question mark and the exciting possibilities expressed in the uses of Dranium a decade ago for power generation, appear to have met with some setback. Present indications are that breederreactors will be delayed until about 1990. The capital cost of such reactors, too, is enormous. By 1980 it has been estimated that 20 per cent of the energy required by the leading ten nations of the world will come from nuclear power, but 80 per cent would then remain for fossil fuels and hydro generation. COAL DEMAND AND USAGE It is of interest to record that coal production in all the countries of Europe has declined markedly in the last 25 years. This is due to competition from oil, gas and nuclear power and also for sociological reasons; but there are signs that efforts will be made to resuscitate, at least in some measure, the mining of coking coals for which there is a strong and growing demand. Possibly all hard coal now mined in Germany has coking properties and that country is self-sufficient in this commodity. Germany might perhaps consider selling coking coal to her neighbours and importing additional coal for power-raising purposes. It has been frequently stated in American publications that the coal reserves of the D.S.A. are sufficient for 1 000 years. What that means, I do not know, but it is at least an indication that America has huge reserves of coal. Domestic and export demand in the U.S.A. in 1968 exceeded production for the first time. In 1969, the United States' output amounted to some 556 million tons, an increase of less than 2 per cent on the previous year. The output achieved fell short of demand, domestic and export, by more than 7 000 000 tons. This has caused and is continuing to cause anxiety in the U.S.A. In many countries young men for a number of reasons, largely environmental and social, are not coming forward in sufficient numbers for mining requirements and the same pattern has revealed itself in the Republic. This is, indeed, alarming because mining men in large numbers will be required to meet the estimated tonnage requirements set out in the following Table: It is difficult to visualise the replacement of these coal requirements either by oil or by nuclear fuels. The coal must be mined and men are required for the purpose. Unless a greater section of the population is allowed to participate in the more responsible aspects of mining, there is a serious likelihood that demand will not be matched by output. In addition to local demand, a lively export trade is possible. Exports are dependent on the future capacity of the South African Railways and Harbours and on Government policy. An export figure of 10000000 tons per year during the decade can be visualised and this figure would raise total estimated sales production to a figure of over 100 000 000 tons in 1980. At this point, it is necessary to examine the desirability or otherwise of exporting coal. How much coal

Jan 11, 1970

By C. Birch, J. Perold

"Due to the challenging geological environment of the Postmasburg Manganese Field (PMF), a study was conducted to determine if any benefits would derive from the inclusion of resistivity data during threedimensional (3D) modelling of the manganese resource. This was achieved by estimating manganese resources from 2011/2012 drilling data and comparing them with manganese resources estimated from the same drilling data and resistivity data collected during 2013 and 2017.Both models were adjusted to limit their extent to the same 3D modelling space. Significant volume and tonnage differences were observed for all lithological units. The greatest differences were noted in the manganiferous zones of alteration – 7.200 Mt for the geological model versus 3.700 Mt for the geoelectric model.This study showed that the inclusion of resistivity data can reduce exploration costs significantly, as a direct consequence of the resistivity data allowing more accurate siting of boreholes. This decreases the number of boreholes, samples, and analyses required due to the 3D electrical delineation of mineralized areas prior to drilling.An additional benefit is the ability to more correctly forecast the net present value of an operation due to more accurate estimation of manganese resources and stripping ratios. This is clearly demonstrated by the estimated gross profit estimation of R409 million for the geological resource model versus R264 million for the geoelectric resource model.The addition of resistivity data can, therefore, reduce exploration costs and can increase confidence in geological and financial modelling. It would be reasonable to conclude that this approach could also be used for karsthosted massive sulphide deposits. IntroductionAccording to Corathers (2014) and Gajigo, Mutambatsere. and Adjei (2011), South Africa holds between 75–80% of global manganese reserves, followed by Ukraine (10%), Australia (3%), India (3%), and Gabon (2%). The known, land-based, manganese resources located elsewhere are estimated at 2%.The majority of economically important sedimentary manganese ore deposits in South Africa are situated in the Northern Cape Province (Gutzmer, 1996). Manganese ores of the Postmasburg Manganese Field (PMF) were discovered in 1922, and the Kalahari Manganese Field (KMF) in 1940. According to Astrup and Tsikos (1998) and Gutzmer (1996), mining of manganese ore from the PMF ceased during 1989 in favour of the superior quality of manganese ore from the KMF."

Mar 1, 2019

By R. A. Wildenboer, R. F. Sandenbergh

Rare earth elements (REE) are present at concentrations of approximately 0.36 % in phosphogypsum stacks located in Phalaborwa, South Africa. The REE are present in the phosphogypsum in solid solution in the gypsum and as fluorine and aluminium rich precipitates. Recovery of the REE offers opportunities for valorization and reducing the environmental impact of the phosphogypsum waste. This paper presents the results of investigations into the leaching behaviour of rare earth elements and other impurities from phosphogypsum using aqueous sulfuric acid. The effects of acid concentration, temperature, residence time, and the influence of impurities were evaluated for their impact on the solubility of rare earth elements in a sulfate medium. It was found that a significant fraction of the REE can be leached from the phosphogypsum in a range of conditions much less severe than traditional hard rock rare earth processing conditions. The REE were leached at a sulfuric acid concentration of 110 g/l, temperature of 30°C – 40°C and a residence time of approximately 8 hours.

Nov 6, 2024

Grains are subject to contamination by mycotoxins which are metabolic by-products of various fungi. The fungal infestations produce local concentrations of the mycotoxins thousands of times higher than the average. These facts make the sampling of the grain for assessment of the mean mycotoxin levels potentially very difficult. In the context of sampling theory, this morphology of the infestations creates a very high distributional and intrinsic heterogeneity of the grain shipment. The distributions of mycotoxin concentrations in seaborne shipments that exist at loading and the number and intensity of infestations that develop under shipping conditions are unknown. All grain shipments carry spores that may give rise to mycotoxin-producing colonies; moisture and temperature determine the final extent of the problem. This paper develops the value of the sampling constant with respect to intrinsic heterogeneity based on data on observed Ochratoxin A (OTA) concentrations in individual wheat kernels. A useful relationship between the distribution of mycotoxin concentrations on a kernel to kernel basis and the distribution of concentration over subsamples is developed.

Jan 1, 2009

By B. P. Watson, A. Carpede, W. O. Kekana, N. Fernandes, W. Theron, M. P. Mahlangu, G. Betz

The research described in this paper was done to confirm the Upper Group 2 (UG2) PlatMine peak pillar strength formula (Watson et al., 2007), which was determined from a back-analysis of failed and unfailed pillars. Underground measurements were made on a stable pillar that was loaded by firstly reducing it’s length and then by mining the surrounding pillars until pillar failure took place. The pillar was instrumented with suitably positioned strain cells and closure meters, which allowed both the average pillar stress and strain to be determined. The paper describes the methodology applied to identify a suitable position for the instrumentation, as well as the results. A stress/strain curve is presented for a UG2 pillar with a w/h ratio of 2.0, at Booysendal Platinum Mine. The measured pillar strength was similar to the predicted strength using the PlatMine pillar strength formula for UG2 pillars. The PlatMine formula has been successfully implemented on Booysendal Platinum Mine, and about 3 670 pillars have been cut without a single failure. An additional revenue of US$1.3 billion was calculated for the 25-year life of the mine as a direct result of the improved pillar design, given the January 2020 platinum group metals basket price. An extended life of mine and better mining efficiencies will also be realized.

Aug 1, 2021

By S. Kahraman

The prediction of rock properties from indirect testing methods is important, particularly for preliminary investigations since indirect tests are easier and cheaper than the direct tests. In this study, we investigate the predictability of the uniaxial compressive strength (UCS) and Brazilian tensile strength (BTS) of rocks from the indentation hardness index (IHI) obtained using point load apparatus. Forty-six different rock types, 14 of which were igneous, 15 were metamorphic, and 17 were sedimentary were tested in the laboratory. The UCS and BTS values were correlated with the corresponding IHI values and the results were statistically analysed. The influence of rock classes on the relationships was also investigated. A strong correlation between UCS and IHI was found for all data. The correlation between BTS and IHI is not as strong as the correlation between UCS and IHI. However, it is in the acceptable limits. When the regression analyses were repeated for igneous, metamorphic, and sedimentary rocks, the correlation coefficients were generally increased. The results show the UCS and BTS can be estimated from IHI. In addition, the effect of rock classes on the relationships between IHI and both UCS and BTS is important.

Jan 1, 2012

By Q. P. Campbell

Activated carbons used in the precious metals extraction industry are characterized by large internal surface areas and a great affinity for metal ions. The purpose of this research is to prepare activated carbon from a South African bituminous coal by physical activation that is suitable and cost-effective for use in the extraction of metals. The quality of the coal-based activated carbon may not prove to be as good as activated carbon produced from other traditional sources, but the production costs involved may make South African coal a feasible alternative feedstock. The activated carbons produced were characterized by Brunauer, Emmett, and Teller (BET) surface area, activated carbon pH, and phenol adsorption studies and the results compared to the results from a commercially available activated carbon, Norit RO 0.8 (control sample). Bituminous coals from various sources including Witbank Seam 4 and Free State coal were used in this study. The preparation method chosen was physical activation using superheated steam. The effects of process variables such as activation time (1-3 h) and temperature (600-800°C) were studied in order to optimize those parameters. The activated carbon surface area was characterized by means of nitrogen adsorption isotherms at 77K. BET surface area analysis showed that Witbank Seam 4 coal activated at a temperature of 800°C and activation time of 3 hours resulted in a surface area of 340 m2/g. Quality control of each sample was performed by measuring the pH of a known amount of the prepared activated carbon in distilled water over time. Results showed that the pH of some of the prepared activated carbons reached a value of 11. Phenol adsorption results for the different activated carbons prepared corresponded well to the results obtained for the Norit RO 0.8 activated carbon sample.

Jan 1, 2012

By H. Han, Z. Gao, Y. Gao, W. Sun, J. Wang

Strategic Ca-bearing minerals like scheelite and fluorite are usually separated by flotation, yet the separation is quite difficult due to their similar surface Ca active sites. Compared with the conventional strong collector sodium oleate (NaOL), lead nitrate-benzohydroxamic acid (Pb-BHA), complex collector shows a better selectivity for scheelite, yet its high cost hinders its wide application. Therefore, a novel collector (Pb-BHA-NaOL complex) was initially developed in this work to overcome the abovementioned dilemma. The flotation results of a single mineral showed that Pb-BHA-NaOL had a better selectivity for scheelite than Pb-BHA. Under the conditions of sequential assembly of Pb, BHA and NaOL at the molar ratio of 240:120:1 and pH 9.0, selective separation of scheelite from fluorite could be achieved without adding any depressants. In addition, for the same preferred scheelite recovery, Pb- BHA-NaOL could also save approximately 20% Pb and BHA compared with Pb-BHA. The analyses of the zeta potential measurement and X-ray photoelectron spectroscopy indicated that Pb-BHA-NaOL showed stronger adsorption on scheelite relative to fluorite. Meanwhile, Pb-BHA-NaOL showed more adsorption modes and higher utilisation of Pb and BHA on scheelite than Pb-BHA, which was due to the accelerator, NaOL. Keywords: Scheelite, fluorite, lead nitrate, benzohydroxamic acid, sodium oleate, flotation

Jan 1, 2020

By M. Gao

The Zinifex Century Mine employs a large number of Detritor mills for fine and ultra-fine grinding to liberate silica and sphalerite at 80 per cent passing 7.5 microns before final zinc flotation. A joint project between CSIRO Minerals and the Zinifex Century Mine was carried out to evaluate the Detritor mill performance. Plant surveys were conducted around the Detritors in conjunction with laboratory tests. The survey data showed that the Detritors in the ultra-fine milling circuit performed well. They reduced a feed F80 sizing of 46 microns to a product P80 sizing of 7.5 microns at a gross energy consumption of 37.2 kWh/t. Scale-up of ultra-fine milling Detritors was studied using a 0.55 kW laboratory batch unit and a 355 kW full-scale Detritor by processing Century zinc concentrate with 1 mm sand media. The results indicated that the 0.55 kW laboratory mill underestimated the energy consumption of the full-scale Detritor by about 20 per cent for a product P80 of seven microns. The performances of a horizontal IsaMill and a vertical Detritor were compared as ultra-fine milling devices for processing Century zinc concentrate. The experimental data demonstrated that, when using the same grinding media, the energy efficiency of the two mills is similar despite the difference in mill designs. The Detritors used for regrind duty at the Century Mine performed poorly when processing feed at 80 per cent passing 88 micron. The energy consumption was high and the product size distribution was broad with particles above 100 microns in the feed unbroken and particles below ten microns overground. A coarser ceramic media of 3 mm was therefore tested in a laboratory Detritor to process the regrind Detritor feed as an alternative to the 1 mm Colorado sand used at the Century Mine. The ceramic media reduced the energy consumption significantly and produced a narrower product size distribution, which is more desirable for the downstream flotation process.

Jan 1, 2007

By A. Yañez, J. Yianatos, P. Vallejos, R. Grau

This paper presents an evaluation of the effect that the mineral feed characteristics have on the metallurgical performance in an industrial flotation circuit. Variables such as mineralogy, particle size, solid content and mineral grade were evaluated using a flotation simulator, which was previously developed and validated using industrial data from different concentrators. The metallurgical performance was evaluated not only from the point of view of recovery and concentrate grade, but also by analysing less common variables such as froth carrying rate, froth and collection recovery, water recovery and others. Then, new data from the industrial flotation circuit was used to complement this analysis and evaluate the simulator predictability by comparing simulated and measured data. The results allowed for the quantification of the significant effect that mineralogy and particle size have on recovery and/or concentrate grade. Therefore, the grinding stage and flotation cells design become important factors to avoid a significant loss in metallurgical performance when feed mineral characteristics change in an industrial operation. In this context, an evaluation of cells design and arrangement should be performed to compensate the loss in recovery and concentrate grade in order to find an optimal operating condition. Keywords: Flotation circuits, mineral feed, modelling, particle size, liberation, mineralogy

Jan 1, 2020

By S. Khoshjavan

In this investigation, the effects of different coal chemical properties were studied to estimate the coal Hardgrove Grindability Index (HGI) values index. An artificial neural network (ANN) method for 300 data-sets was used for evaluating the HGI values. Ten input parameters were used, and the outputs of the models were compared in order to select the best model for this study. A threelayer ANN was found to be optimum with architecture of five neurons in each of the first and second hidden layers, and one neuron in the output layer. The correlation coefficients (R2) for the training and test data were 0.962 and 0.82 respectively. Sensitivity analysis showed that volatile material, carbon, hydrogen, Btu, nitrogen, and fixed carbon (all on a dry basis) have the greatest effect on HGI, and moisture, oxygen (dry), ash (dry), and total sulphur (dry) the least effect. Keywords coal chemical properties, Hardgrove Grindability Index, artificial neural network, back-propagation neural network.

Jun 1, 2013