Search Documents

By D. deV. van der Spuy

A procedure for optimizing the secondary cooling zone of continuous casters is proposed and illustrated by a billet caster example. The procedure consists of quantifying deviations from the required conditions of the secondary cooling zone?as far as temperatures, metallurgical length and reheating are concerned?by means of a ?cost function?. The optimal spray settings are then found by minimizing the total ?cost?. The billet caster examples serve to emphasize the importance of specifying the required conditions in a consistent manner. Some envisaged practical limitations of this approach are discussed.

Jan 1, 1999

By R. Minango

Solvent extraction of Cu from acidic leach, using an oxime extractant, e.g. LIX 984 or Acorga is a well-known and establishedprocess. As the process advances, Cu is extracted to the solvent and is replaced by H+in the aqueous phase. At high acidity the distribution coefficient of Cu between the solvent and the aqueous phase decreases, and eventually a relatively high concentration of Cu remains in the raffinate, regardless of the number of equilibrium stages. In many processes this phenomenon is immaterial, as the raffinate returns to the leaching and no copper is lost. But when the copper raffinate is used for processing of additional metals, minimal concentration of Cu in the raffinate is essential for an effective process. The recommended process includes three steps: SX-1?standard extraction of Cu with oxime. The resulting raffinate contains about 6?8 g/lCu and 50 g/lH2SO4. SX-N?Extraction of most of the acid into an amine-based solvent, and stripping it with a base, e.g. aqueous ammonia, caustic, Na2CO3etc. The extraction removes no metals, but decreases the concentration of acid from 50 to10 g/l. SX-2?Extraction of the neutralized solution once more with oxime, to remove the residual copper. Such a process enables one to get down to 100?150 ppm Cu in the final raffinate. This process was developed and tested in the R&D facility at Chambishi Metals plc, by a joint team of Chambishi & Bateman, using Cognis LIX 984N and real solutions from the plant. In the laboratory the equilibrium isotherms for SX-1, SX-2 and the neutralization were developed. Computer simulation of the results showed that the optimal configuration is two stages for both extraction and stripping for SX-1. For SX2, 2?3 stages of extraction and two stages of stripping are needed. The stripping stages may be common to both SX-1 and SX-2. The neutralization of SX1 raffinate was tested both by conventional precipitation with lime and by extraction with alamine-based solvent extraction and stripping with a solution of 10% NaOH. For the SX-N only one stage was needed for both the extraction and stripping. The bench scale pilot plant was run for both SX-1 and SX-2. The results were very similar to the simulated ones, and the configuration of 2E2S for SX-1 and 3E2S for SX 2 delivered the required raffinate with less than 150 ppm Cu.

Jan 1, 2005

By B. Ndlovu

The world mineral industry has over the years experienced growth in the production of nickel (Ni) due to its use not only in the catalytic industry but mainly also to its demand in the stainless steel production industry. Hence there is a need to produce more pure and marketable grades of Ni through the use of efficient purification routes such as solvent extraction. The V10-LIX®84-IC system resulted in nickel (Ni) extraction isotherm shifting to lower pH levels. The pH window of opportunity for the extraction of Ni subsequently increased in solution containing magnesium (Mg) and calcium (Ca) even though the system also resulted in slight shifts to less positive pH values for Mg and Ca curves. Results showed a synergistic effect of LIX®84-IC when added to V10 for all elements, Ni, Mg and Ca. The pH50 values for elements Ni, Mg and Ca in the V10-LIX®84-IC system were 2.95, 8.10 and 7.70 respectively. Comparing results obtained when using V10 alone produced ?pH50 for Ni, Mg and Ca being 3.35, 0.15 and 0.70 respectively, which was a confirmation of the synergistic shifts for all the elements. The ?pH50 (Mg-Ni)value was 1.95 with no synergism and 5.15 with synergism and that for ?pH50 (Ca-Ni)was 2.10 with no synergism and 4.75 with synergism. This showed that the Ni-Mg separation factor was relatively larger than the Ni-Ca. Improved loading and stripping kinetics were also observed. Keywords: Nickel solvent extraction separation, magnesium, calcium, versatic 10 acid and LIX®84-IC.

Jan 1, 2008

By H. Hu, T. S. Golosinski

The paper reports on use of neural network algorithms to model performance and condition of mining trucks. The model inputs are selected digital data streams collected by truck Vital Information Management System, VIMS, during its operation in a mine. The models built as the result of this work allow for projection of truck condition and performance into future. As such they facilitate more efficient management of mine truck fleets. Keywords: Modelling, modeling, neural networks, mining equipment, Vital Information Management System, Intelligent Miner, mining truck

Jan 1, 2003

By M. Lu, E. Broch

Abstract: The formation and mechanism of rock cavern roof arches are studied and effects of rockbolts reinforcement are evaluated. It is concluded that a self-supported roof arch may be formed by the excavation. To stabilize such a roof arch, fully grouted rockbolts are effective. Tensioned cable anchors may have a negative influence on the arch stability. Zusammenfassung: Die Formation und der Mechanismus von Deckenbögen in Bergkavernen wird studiert und der Effekt von Verstärkung mit Bergbolzen wird bewertet. Es wird festgestellt, dass die Exkavation einen selbst-stützenden / freitragenden Deckenbogen formen kann. Eingegossene Bergbolzen sind effektiv um einen solchen Deckenbogen zu stabilisieren. Spannkabel können unter Umständen eine negative Einwirkung auf die Stabilität des Bogens haben. Résumé: Le mécanisme et la formation de voûtes dans les cavernes rocheures sont étudiés et l'effet du renforcement par boulonnage est évalué. Nous arrivons à la conclusion qu'une voûte autosupportêe peut être formée par excavation. Des boulons entièrement scellés par coulis de bêton s'avèrent efficaces pour stabiliser de telles voûtes. Les tirants d'ancrage peuvent quant à eux nuire à la stabilité de la voûte.

Jan 1, 2003

By L. R. C. Cairncross

In this study, the general case of semi-batch precipitation of nickel hydroxide, Ni(OH)2, is modelled. In practical terms, this precipitation may occur at high ionic strengths, and therefore the model for solution chemistry requires a thermodynamic model that is applicable for these non-ideal systems. The activity coefficient model used is the ELEC-NRTL. The simulation of the solution chemistry is performed using AspenPlusTM, which is a steady-state process simulation program. This allows for rigorous evaluation of the solution speciation, and hence the degree of supersaturation which dictates the precipitation kinetics. The solution chemistry provides the basis for predicting the formation of solids. The model consists of a solution chemistry part and a solids formation part. The inputs to the model include the initial concentrations of the streams entering the reactor, kinetic parameters for the precipitation processes, and a particle size distribution obtained from experiments. The solids formation part of the model employs the method of moments for the evaluation of the population balance. The population balance accounts for total particle number, and the formation and increase in size of particles across the entire particle size range. This balance requires kinetic parameters for the processes of nucleation, growth and agglomeration. Using optimization routines inherent in AspenPlusTM, parameters for the rate of nucleation, growth and agglomeration may be extracted from experimental particle size distributions. These are then used in the model to extrapolate to other conditions. Once an understanding of the kinetics of the precipitation processes is gained, the model provides a predictive tool for optimization of these kinetics, in order to increase the overall particle size and therefore improve solid-liquid separation. The work is deemed to be significant as it sheds some light on solids formation processes involved in precipitation?including nucleation, growth and aggregation?and develops a predictive modelling tool applicable to solutions of high ionic strength. We argue that this dewatering of suspensions is affected significantly by particle size distribution and surface properties of the solids formed by precipitation. This modelling tool should facilitate the design of solids formation processes where downstream dewatering is enhanced.

Jan 1, 1997

By X. J. Lyu, Q. Liu, M. J. Lian, X. Q. Cao, X. F. You, B. O. Erdenekhuyag, H. Y. Zang

The slag-tailings based artificial reef concrete (ARC) is made of CaO-CaSO4 activated granulated blast furnace slag (GBFS) as cementitious material and deslimed gold ore tailings as aggregate. In order to investigate the potential application of the material as ARC, the strength development, hydration characteristics and application performance were investigated under different preparation conditions. Results indicated that using seawater as the mixing water can effectively promote the hydration of slag and generate a large amount of hydration products such as ettringite, CSH gel, hydrotalcite and Friedel ’s salt in the early curing period and so render the material with good early compressive strength. Seawater curing can also accelerate the hydration reaction in the early stage of hydration. The hydration reaction of slag and the compressive strength of ARC can be effectively promoted by increasing the specific surface area of slag. The 28-day compressive strength of the slag-tailings based ARC prepared with 30-50% CaO-CaSO4 activated slag and 50-70% tailings can reach to 19.88-42.87 MPa. Also, the material has a low surface pH value, good resistance to chloride ion penetration, and sulfate corrosion. Keywords: Artificial reef concrete, granulated blast furnace slag, deslimed gold ore tailings, compressive strength, hydration characteristics

Jan 1, 2020

By X. H. Zhang, J. G. Cai

This paper discusses the behaviour of large-span tunnels and caverns in hard rock in a major underground storage project. In-situ stresses were measured using hydraulic fracturing during site investigation and 3-D overcoring during construction, with both showing a horizontal stress ratio of 2-3 times the vertical stress. Rock reinforcement design was based on the Q-system combined with numerical modelling for special design cases. Tunnel design and performance were checked and monitored by in-situ instrumentation during rock excavation, including measurements of deformation, convergence, and load on rock bolts. Deformation measurements using borehole extensometers included one borehole drilled into the virgin rock before rock excavation, which provided a direct and very insightful comparison with the various deformation measurements made during rock excavation. These measurements coupled with numerical calculations have proven that the high horizontal stresses to be very favourable to the stability of the large openings.

Jan 1, 2003

By B. D. Paphane

Since the commissioning of the Tati Hydrometallurgical Demonstration Plant (HDP), which produces Ni and Cu cathodes as well as Co salt from sulphide concentrate employing Activox® technology for leaching, no kinetic data has been generated from the plant under normal operating conditions, from start-up to steady state. Therefore, this study seeks to determine the rate of reactions in the leaching circuit with a view to obtaining kinetic data for the system under the current operating conditions. Within the five-compartment autoclave operating at temperatures of 105°C to 110°C and 1100 kPa pressure, the extraction of Ni, Cu, and Co ions into solution is achieved through the oxidation of the sulphide concentrate using approximately 50 kg H2SO4 per ton, 10 kg NaCl per ton, and 1.21 kg O2 per kilogram of S2-. The autoclave reactions showed complex kinetics, with the reactions proceeding via some intermediate(s) before the formation of products. The results suggest that the reactions were dependent on the size of the metal ion of interest and the concentrations of the metal ions in the Tati concentrate as well as in chloride leaching. The order of the leaching rate constants was Fe2+ > Cu2+ > Ni2+ > Co2+ in all the compartments studied. The rates of leaching were also found to decrease as the concentrate moves from compartment 1 to compartment 4. For compartments 1 to 3, leaching was observed to be preceded by dilution of the repulp liquor (copper raffinate). The leaching rate for compartment 5 was slightly lower than that of compartment 1 but higher than the rates of compartments 2, 3, and 4 for all the metals studied. Keywords kinetics, hydrometallurgical plant, Activox

Jun 1, 2013

By G. Westcott

Impala Platinum, the second largest platinum producer in the world, is embarking on a major expansion strategy to increase its platinum output from 1.95 moz to 2.8 moz by the year 2008. The smelter complex, situated in Rustenburg, will undergo substantial modifications, and will require a solution to the capture and treatment of the smelter off-gases and dust produced during smelting and converting operations. One of the solutions explored was an integrated approach, where all the off-gases from both the primary smelting electric furnaces and the converter operations are routed to a large central gas cleaning and double catalysis, double absorption acid plant. In-line sulphur burning, utilizing a highly flexible LURO1spin cup burner with a turn down ratio of 50%, ensures that the acid plant operates independently of operating conditions at the smelter complex.

Jan 1, 2007

Mr D. Wilson was born in Vereeniging, where he also completed his schooling, matriculating from General Smuts High School in 1977. Mr K.A. van Gessel was born in 1937. His final studies were conducted at Delft University, where he graduated with an M.Sc. degree in mining engineering. Mr J.G. McCaUum was born in 1938 and matriculated at Jeppe Boys' High School, Johannesburg. Mr H. Scot RusseU started his mining career at Randfontein Estates Gold Mining Co. Witwatersrand, Limited, where he served as temporary Geological Assistant, Sampler, Surveyor, Shift Boss, and Mine Overseer. Mr A.M.P. Henderson was born in 1950, in Kiel, West Germany. Mr J.L. Nel matriculated in 1957. This was folowed by his atainment ofthe folowing degrees: B.Sc. (Chemistry, Physics) in 1961, B.Sc. Hons. (Metalurgy) in 1965, M.Sc. (Metalurgy) in 1967, and MDD (student of the year) in 1981

Jan 1, 1989

A single aggregate mix design was developed based on a slag/Portland binder blend and crusher sand. The primary design criterion, after performance, was the fluidity of the shotcrete and concrete product to ensure compatibility with existing manufacturing and shotcrete equipment as well as the pipeline transport system Cullinan Diamond Mine was investigating. Initial work had determined an optimum slag/Portland to aggregate ratio while retaining a water to slag/Portland binder ratio of approximately 0.4 to maintain the flowability of the product. A super-plasticizer was used to further increase the flowability of the product as opposed to increasing the water content. Finally a polypropylene reinforcement fibre was included in the wet shotcrete product to improve the energy absorption characteristics of the shotcrete. Previous work on the strength development of the slag/Portland based shotcretes and concretes suggests that strength class based on the EN 206 specification is C40/50 while the toughness classification based on the EFNARC plate test is c, or greater than 1000 joules1. Typically, Cullinan Diamond Mine wanted a shotcrete with a uniaxial compressive cube strength of 45 MPa and an energy absorption, based on the EFNARC plate test, in excess of 700 joules. This is an EN 206 strength class of C36/45 requiring a uniaxial compressive core strength of 30.5 MPa and a toughness classification of b1. In addition to the shotcrete development, Cullinan Diamond Mine required concrete products with a compressive strength range of between 25 and 60 Mpa for differing construction operations underground to be manufactured from the same mix design and raw materials as for the shotcrete. Finally, Cullinan Diamond Mine carries out extensive dry shotcrete operations and an investigation into the possibility of using the single aggregate mix design for dry shotcreting was also entered into. This paper presents data that indicates that, under the terms of reference, a wet and dry shotcrete and concrete product appears to be possible from a single aggregate less than 6.7 mm in size, and that a slag/Portland binder is suitable for use as the hydraulic binder in all the products, and that the necessary performance parameters of strength and energy absorption can be achieved.

Jan 1, 2006

By Ravichandar. D., K. B. Nagashanmugam, M. S. Pillai

"Blast furnace performance depends strongly on the coke reactivity index(CRI) and coke strength after reaction (CSR) properties. An innovative and cost-effective method, known as the Salem Box Test, has been developed to prevent the mass production of inferior coke unsuitable for blast furnace use. This method consists of coal carbonization on a micro-scale and involves charging approximately 18 kg of coal blend in a stainless steel box, carbonizing it together with coal cake in the plant coke ovens, and testing the coke produced for CRI and CSR to determine its suitability for blast furnace use. Only coal blends that yield coke with CRI <25% and CSR >64% are permitted for mass production, and other coal blends are either rejected or the blending ratios adjusted in an attempt to upgrade them. The experimental results reveal that, for a given coal blend, the quality of coke produced by the Salem Box Test is comparable with that produced by bulk production, indicating that the test is acceptable as a screening tool for regular use. The present paper describes the methodology and application of Salem Box Test to predict the suitability of coke for blast furnace use at JSW Steel Limited, Salem Works (JSWSL), and illustrates its advantages in adjusting the coal blending ratio to produce superior coke, in detecting coal contamination, and in preventing bulk production of inferior coke."

Jan 1, 2015

After the closure of collieries, they naturally start to fill up with water. As a result, hydraulic gradients develop between them and different hydraulic pressures are exerted onto peripheral areas or compartments within mines. This results in water flow between mines, or onto the surface. This flow is referred to as intermine flow. The collieries in the Witbank Coalfield have geometries such that there are several areas where this intermine flow is possible. Since the Department of Water Affairs and Forestry has declared intermine flow as one of its greatest concerns for granting closure to South African mines, much research into this phenomenon is required. The challenges in determining intermine flow are numerous, and attention has focused on identification of areas where these flows can take place. The quantification of these flows is problematic due to the uncertainties in exact geometric configurations and the variation in site-specific hydraulic properties of thecoal and overlying lithilogical layers. During this study, use has been made of numerical flow modelling and several analytic solutions to test the applicability of the flow models as well as to predict groundwater flow directions, filling times of voids and flow volumes. The numerical modelling methodology entailed a downscaling approach starting with a broad regional model covering the entire area, followed by modelling the interactions between interconnected mines, and finally looking in detail at the major areas of interaction.

Jan 1, 2005

The results of a series of physical model studies of the stability of jointed rock slopes carried out some years ago are described. These models were tested under centrifugal loading. Four ?geological? structures were tested; all had the same joint orientations and strengths, but the relative spacing of the joints was different. The results showed that the mechanisms of failure of the slopes varied for the different models, and the gravitational loads at which failure developed also varied. The following conclusions are drawn from the results of the physical model tests: ?the mechanism of slope failure is by progressive deformations throughout the slope ?the ratio of the spacings between the different sets of discontinuities exerts considerable influence on the slope failure ?in the models tested, the intact material strength had no noticeable influence on the stability of the jointed rock slopes(note that for hard rock, similitude conditions indicate a prototype slope height of 300 to 400 m, and therefore intact rock failure might not be expected under simple gravity conditions, i.e. no tectonic stresses) ?rock slope failures are three-dimensional and in general are not amenable to two-dimensional simplification ?knowledge of the orientations of discontinuities in the slopes does not allow prediction of a unique failure surface, nor of a volume of failure. The volume of failure is determined to a large extent by the plan configuration of the slope.

Jan 1, 2006

By M. M. D. S. Vemba

Paper written on project work carried out in partial fulfilment of B.Sc. Eng (Mining Engineering) degree Engineering refers to the application of mathematics and other sciences to logically develop solutions to problems. However, in opencast mining, due to the randomness of operations, the use of mathematics is often neglected. Instead, practices that have proven successful elsewhere tend to apply. At the other end of the scale, mathematical models have been developed to identify and quantify all the operational parameters involved in the loading and haulage of ore. These are then used to optimize the operation. Optimization in this context, refers to the reduction of cost through effective utilization of the available time and resources. Considering these mathematical models, the loading and transport system at SMC (Catoca Mining Society) will be assessed with a view to its optimization. The discussion will be introduced by considering the mine background and other general information, which will highlight the main operational variables.

Jan 1, 2004

By G. van Aswegen

Large structures in mines (faults, dykes, pillars, abutments) fail when the load exceeds the strength. If the loading system stiffness (LSS, also referred to as the ‘local mine stiffness’) is lower than the stiffness of the structure the failure will be unstable. The greater the difference between these stiffnesses, the greater the energy release during unstable failure. A method has been formulated by which the LSS is established through cost-effective numerical modelling while insight into the structure stiffness (source stiffness - SS) is provided by apparent stiffness, a parameter based on the empirical relation between seismic moment and radiated seismic energy of small seismic events from the area/volume of interest. By plotting LSS vs SS a ‘stiffness diagram’ is constructed that is useful for seismic hazard ranking of different structures. A comparative back-analysis of the Dagbreek fault in Welkom, South Africa, (source of the 1990 Matjhabeng earthquake) and two other nearby faults confirm the applicability of the method. An extension of the method, where ‘hazard magnitude’ is used as a measure of the LSS, allows a ‘stiffness diagram’ to be constructed using seismic data only. Les larges structures dans les exploitations minières (failles, dikes, piliers, culées) s’écroulent lorsque la charge dépasse la résistance. Si la raideur du système de chargement (LSS, aussi appelé la « raideur de la mine locale ») est plus basse que la raideur de la structure, l’éboulement sera instable. Plus la différence est grande entre ces deux raideurs, plus l’énergie dégagée pendant l’éboulement instable est grande. Une méthode a été formulée selon laquelle le LSS est établi par modélisation numérique rentable, tandis que la raideur de la structure (SS – raideur de la source) est établie par la raideur apparente, un paramètre basé sur la relation empirique entre le moment sismique et l’énergie sismique radiée de petits moments sismiques dans l’aire/volume en question. On trace le « diagramme de raideur » en relevant les deux mesures (LSS et SS), ce qui est utile pour classer les risques sismiques de structures différentes. On vérifie l’applicabilité de la méthode en comparant la faille de Welkom, Afrique du Sud (foyer du séisme de Matjhabeng en 1990) avec deux autres failles proches. Une extension de cette méthode, dans laquelle on utilise la « magnitude du risque » comme mesure du LSS, permet de construire le « diagramme de raideur » avec des données sismiques seulement. Große Strukturen in den Gruben (Verwerfungen, Dämme, Pfeiler, Auflager) scheitern, wenn die Last die Stärke übersteigt. Wenn die Steifheit des Ladesystems (LSS, auch bezeichnet als `lokale Grubensteifheit’) niedriger ist als die Steifheit der Struktur wird der Bruch instabil. Je grösser der Unterschied zwischen diesen Steifheiten, desto grösser die Energiefreigabe während des instabilen Bruches. Es wurde eine Methode formuliert, durch die das LSS durch eine kosteneffektive numerisches Modell erstellt wird, während ein Einblick von der offensichtlichen Steifheit in die Struktursteifheit (Quellensteifheit - SS) gegeben ist, ein Parameter der auf der emipirischen Relation zwischen seismischem Moment und ausgestrahlter seismischer Energie von kleinen seismischen Ereignissen vom Interessens-/Volumenbereich basiert. Durch das Kartograpahieren der LSS in Bezug auf die SS wird ein `Steifheitsdiagramm &apos; konstruiert, welches für seismische Gefahrklassifizierung der unterschiedlichen Strukturen nützlich ist. Eine vergleichende Postanalyse der Dagbreek Verwerfung in Welkom, Südafrika, (Zentrum des Matjhabeng Erdbebens 1990) und in zwei anderen nahe gelegenen Verwerfungen bestätigen die Anwendbarkeit der Methode. Eine Verlängerung der Methode, in der die `Gefahrengröße&apos; als Maß für das LSS verwendet wird, gestattet, daß ein `Steifheitsdiagramm&apos; konstruiert wird, welches nur seismische Daten verwendet.

Jan 1, 2003

By C. Maree

Paper written on project work carried out in partial fulfilment of B. Eng (Chem. Eng.) This paper describes the development and use of a process simulator to optimize gypsum recovery at Afmine, a small plant near Yzerfontein on the South African West Coast. The simulator consisted of models of the major plant equipment, viz. hydrocyclones, surge bins, settling tanks, a primary sieve and a dewatering screen. Experimental data suggested that the equipment could be modelled satisfactorily from data published in the open literature. Simulation studies have indicated that gypsum losses on the plant could be reduced substantially by means of some relatively minor modifications to the plant layout. Moreover, sensitivity analyses of the plant have indicated that optimization of the settling tanks should receive priority over the other process units.

Jan 1, 2005

By S. Montes

Mineral floatability is strongly connected with mineral structure and composition. The complexity of mineral surfaces has meant that few attempts have been made to understand the effect of impurities (trace elements) on the flotation process. Zeta potential technique has been extensively used to evaluate the surface charge of mineral particles. Nevertheless, those measurements consider only an average of mineral particles surface charge. The presence of fine particles and inclusions covering a mineral may influence the local reagent adsorption process, which will ultimately be reflected in the subsequent global response of the mineral froth flotation efficiency. The current work assesses the iron oxide mineral, haematite (Fe2O3). Zeta potential and streaming potential techniques were used to determine the surface charge of fine and coarser particles of haematite. Further analysis was performed to determine the point of zero charge (PZC) of the mineral. In addition, adsorption isotherms of alkyl ammonium chloride reagents with different lengths of carbon chains were carried out. It was found haematite floatability depends strongly on the isoelectric point (IEP) value, which is affected by other mineral species present at the mineral surface. Haematite floatability became significant only at relatively high tetradecyl ammonium chloride collector concentrations (concentration greater than 4 10-4M), which does not happen with other oxides such as quartz. Unlike quartz, haematite develops a low surface charge over a wide range of pH, disabling all long-range attractive interactions between the mineral and the collector. Therefore, hydrophobic chain interaction is probably the most likely mechanism acting to promote collector adsorption and enhance further floatability. Diffused reflectance infrared Fourier transform spectroscopy (DRIFTS) was used to suggest a possible mechanism to understand the sequence of collector adsorption and the effect of trace species on it. Keywords: floatability, haematite, adsorption, electrochemistry, zeta potential.

Jan 1, 2007

By Y. He

The nature of the density distribution in a gas-solid phase fluidized bed for dry coal separation is an important factor in influencing the performance. However, during coal processing, fine slime would mix with dense medium of magnetite powder and accumulate in the bed, and its influence will affect the density of the bed. The mathematical model of the density distribution with fine slime behaviour was studied in an experimental laboratory system. The functional relationship between the density of the air dense medium fluidized bed and the coal slime content were fitted, applying the principle of least-squares method with damping factors. With the model, by analysing the trial testing data, the mechanism of the density distribution of the bed was studied and applied to a coal separation test with a pilot scale separator. Keywords: dry coal separation, gas-solid phase, model fitting.

Jan 1, 2002