Search Documents

By A. K. Majumder

In an effort to test the extent to which fundamental fluid mechanics can provide a prediction of particle separation, a study on a simple parallel-sided sluice has been carried out using glass bead and ilmenite mixtures at various particle concentrations and flow rates. Great care was taken to devise a flow splitter at the end of the sluice that permitted accurate measurement of the fluid flow profile and collection of slurry samples, allowing evaluation of the particle species concentration profiles. The actual particle separation taking place was compared to predictions made by first describing the fluid flow by a modified ?law of the wall? model. The hydrostatic pressure variation in the flow was taken into account and the eddy viscosity was calculated from a mixing length expression incorporating Van Driest?s damping function. Using an average density and viscosity provides an explicit expression for the velocity gradient that can be numerically integrated for the flow profile. The particle concentration distributions were then modelled with Hunt?s sediment transport model assuming the particle eddy diffusion coefficient to be the same as that for the water. Hindered settling velocities of the particles were calculated using Brauer and Thiele?s formula, which takes the particle size and density distributions and solids volumetric concentration into account. The resulting predictions of particle separation are shown to be a strong function of the settling velocities of the particles used. Keywords: flowing film concentrators, fluid velocity

Sep 1, 2012

By Nicholas Enraght – Moony, Alvin Johns, Zhonghan Zhang, Zhang Xianhua, T. Choo

"The tungsten (WO3)-rich ores from King Island Scheelite (KIS) Mine, Australia, containing scheelite(CaWO4) were previously recovered by gravity separation and flotation of fine scheelite (1978-92) with an overall recovery of 76% when both circuits were operating. Recently extensive test work has been conducted both in China and in Australia on KIS ores, assaying from 0.50%-1.34%WO3, and tailings assaying 0.2%WO3(GZRINM 2009; 2011a; 2011b). A new approach, which aimed at improving both recovery and grade of concentrate by means of an innovative all flotation route WOF (Whole Ore Flotation), achieved excellent results: 90%-92% recovery to concentrate grade of 66%-70%WO3for the ores. The process is characterized by the use of selective reagents in flotation, including a newly developed collector for scheelite, GYWA. The WOF process also embraces the "Improved Petrov Method" in the cleaning of scheelite rougher concentrate. Earlier investigations found that the removal of sulfides by flotation prior to scheelite flotation was not necessary. It was also found that recycling of process water did not influenceresults. Two main ore types were tested as well as tailings from 2009to 2001(0.20%WO3). In addition a flow sheet using wet high gradient magnetic separation (WHGMS) in tandem with flotation was also investigated for the tailings."

Jan 1, 2014

By J. Jirestig

Poor grade iron ore, suffering from interlocked particles, is challenging for any process technology. Extensive grinding is required to attain liberation and to yield concentrates of desired grade. Fine particles are commonly very difficult to retrieve, resulting in value losses to tailings. In this paper, results from HGMS tests on fine to ultra fine hematite are reported. Special emphasis is put on slurry flow velocity. If hydrodynamic drag force, i.e. flow velocity is not controlled and kept constant during the feed and/or rinse phase, fine particles will not be captured and retained. Flow velocity control and constant flow during separation are achieved by discharge valves/slots and parallel magnetic flux and pulp flow configuration. In tests, increased slurry velocity from 178 to 268mm/s caused Fe-recovery to drop from 78 to 48%. Proving that flow velocity control is essential to optimise grade and recovery. HGMS tests at the Metso mineral laboratory on natural (weathered) hematite fines clearly show a large potential to recover valuables, currently lost to tailings. Even at particle size distributions as low as D80=35µm concentrate grades and recovery are very satisfying. Two samples, in-plant desliming rejects (D80=35µm and 50µm) where significantly upgraded and recovered, from 45,3% to 63,6% and 53,0% to 63,4% Fe, at 73,8% and 81,1% Fe-recovery respectively. Other types of in plant rejects, such as gravity separation or WHIMS tailings, also benefits from HGMS separation. Gravity separation reject, 55,8% Fe, at D80=500µm was tested, yielding a magnetic concentrate of 60,8% Fe at 50,7% Fe-recovery. The results indicated that the hematite was poorly liberated. HGMS separation was repeated after a regrind to D80 150µm, resulting in a magnetic concentrate of 62,8% Fe at 75,9% Fe-recovery. The tests prove that hematite, presently lost to tailings streams, can be recovered by HGMS and thereby significantly contribute to improved overall economy. Keywords: HGMS, magnetic separation, recovery of ultra-fine hematite, Indian hematite ores, hematite recovery from tailings

Sep 1, 2012

By Osvaldo A. Bascur

One of the most vigorous of the continuous improvement methodologies is Six Sigma. While usually associated with improving manufacturing and product quality, leading manufacturers are using it to improve their extended supply chain and logistics capabilities. At the same time they are improving reliability, some companies have cut hundreds of millions of dollars of fat out of their supply chain. Mine to Mill optimization is a transformation of culture from processing tons to processing a quality feed size distribution which reduces the overall operational costs and add the highest value to the mine. This is not a constant endeavor. As such, a dynamic performance monitoring and diagnosis is necessary. We will review a systemic way to performance improvements and avoiding metal losses from the blast to the metals. We show the requirements of using raw data to analysis the data to define operational maps which provide assistance to run the mine and mill. A case study is provided with the mine and mill data was integrated combining process control and quality information resulting in reduction of water and energy consumption while improving the overall value of the mine. The results of reengineering the current system with this SIX SIGMA Strategy had the following results: an increase of ore milling: 4.6% decrease of mil power: 3.9%, decrease of fresh water consumption: 6.8%, with a total of net profits: US$ 31.8 million, the PI System infrastructure contribution: US$ 7.95 million. These estimates were for the period 2009/04/04 to 2009/12/31.

Sep 1, 2012

By Y. Q. He

The dynamic behaviors of the air bubbles in the dense gas-solid fluidized bed have a significant effect on the performance of dry coal separation. The combined method of experimental measurement and numerical simulation were conducted to conduct the research of the mean diameters and rising velocities of the bubbles in the fluidized bed. The calculation models of mean diameter and rising velocity of the single bubble were proposed. The results indicate that the models of the mean diameter raised by Rowe (1976) and the rising velocity raised in this work could be used to predict the bubble dynamic parameters in gas-solid fluidized bed. With the Geldart B type of magnetite powder as the medium solids and while the gas velocity of 1.5Umf=U=2.2Umf, the mean bubble diameter along the bed height is 36mm<Db<45mm, and the mean rising velocity of the single bubble is distributed in 40~65cm/s. The results of experimental measurement, numerical simulation and model predication provide well agreements finally. Based on the uniform fluctuations of air bubbles diameters and their rising velocities, the separation experiments of 50~6mm dry coal using the mixture of 0.3~0.15mm magnetite powder and <1mm fine coal as medium solids were carried out. The results show that the probable error, E is 0.06, which indicates a good separation performance by using the gas-solid fluidized bed. Keywords: dense gas-solid fluidized bed, bubble dynamic behaviors, numerical simulation, separation performance

Sep 1, 2012

By Xiaojing Li, Yueyuan Zhou, Shengyao Ma

The paper presented the acnalysis of field intensity of rare-earth permanent magnet with high BH product and the research of magnet configuration with narrow pole pitch on open magnet surface, both using ANSYS. The possibility to separate micro-disseminated feebly magnetic minerals by improving the field intensity and the capture line distribution rate on open magnet surface were verified. The research on the method to pile up magnets to increase peak value of field intensity on the open magnet to 1.0T was done through correlation analysis between shape and dimensions of magnetic material and field intensity. And the research on technology method to improve capture line distribution rate in the separating field to 1.25×105/m2 was done through optimization design of magnet configuration of the open magnet with narrow pole pitch. Laboratory tests were performed with hematite sample from ore dressing plant such as Hainan mine, Qidong iron ore, etc. The separation results acquired were at the same level as using the electromagnetic separator, verifying the possibility to separate micro-disseminated feebly magnetic minerals with open magnet. The research suggested a new idea about the development of magnetic separator with high efficiency and low costs to separate feebly magnetic minerals and non-magnetic gangues.

Jan 1, 2014

By P. Prusti

An air dense medium fluidised bed separator (ADMFBS) is the efficient process for dry beneficiation of coal. Magnetite powder is used as medium to make required pseudo-fluid density in the fluidised bed. The dynamics behavior of coal particle with reference to its position changes from ideal solution to pseudo-fluid condition in different bed densities depending on its shape, size and density. In this study, three set of projected areas of coal particles were taken. Each set contains four different shapes i.e., cubical, rectangular prism, spherical and triangular prism with different density starting from 1.3 to 1.8 gm/cc. It is observed that coal particles having density less than the bed density, floats in different positions in the dense medium fluidised bed. Higher projected area particles float in deeper height in comparison with smaller projected area. The difference in position of coal particles in the bed increases (nearly 1 to 4 cm) as the coal density approaches towards bed density irrespective of the different shapes. Different shapes of coal particles behave in different ways based on the operating parameters like bed density, coal density and projected area of coal particle. Empirical model equations are proposed to predict the position of coal particle of each shape in the air dense medium fluidised bed system. Keywords: ADMFBS, particle dynamics, projected area, shape

Sep 1, 2012

By C. Igathinathane

Coal is the most widely used fossil fuel in the world, providing more than 25% of world?s electrical power. Since, the behavior of coal particles in crushing and grinding circuits, concentration operations, and solid-liquid separations is strongly dependent on size, a reliable and accurate measurement of particle size and particle size distribution (PSD) is a vital aspect in coal cleaning. This paper presents a method for the characterization of ground coal particles using image analysis. We utilized a machine vision (MV) approach that used a document scanner as the imaging device and a user-coded ImageJ plug-in that processed the image and automated the PSD analysis and output results in textual and graphical forms. This approach used a sum of volumes (SVolume) as weighting factor for particle length as the primary dimension of analysis and was successfully applied to ground coal. The plug-in was further improved for computational performance in this study. Mechanical sieving (MS) was also used to compare the MV results. MV results showed that the PSD of ground coal followed a uni-modal normal distribution, and the log-normal plot against particle size exhibited a linear trend for most of the range. MS results, however, had three linear segments, when the width-based sieving results were transformed to lengths by applying the shape factor (width/length), but a definite deviation between the PSD plots was observed. This study demonstrates successful application of the MV SVolume approach for PSD analysis of ground coal, which can be extended to similar particulate minerals and products. Keywords: dimension, fiji, image j, image analysis, coal, particle size, plug-in

Sep 1, 2012

By Carina Ulsen

The existing construction and demolition (C and D) waste recycling technologies and standards have been essentially focused on the production and use of coarse recycled aggregates. Very few papers focus on the production of recycled sand, although some previous results show that fine aggregates fraction represents around 40 to 60% (in mass) of the Brazilian waste and it is usually used as road sub-base or disposed in landfills. The quality of the recycled aggregate is strictly related to the content of porous and low strength phases, mainly the patches of cement remaining on recycled aggregate surface. Some technologies have already been described in the literature, even though, to the moment none of these technologies has clearly succeeded in reaching the large market available. This paper presents a summary of the main properties of the sand produced from C and D waste by tertiary crushing at vertical shaft impact crusher (VSI). Additionally mineral processing technologies were applied on the attained product, such as gravity concentration by shaking table and spirals and removal of cement paste by attrition. The main properties of recycled sand are discussed and compared to the previous C and D waste, respectively: particle size distribution, particle shape, chemical composition, cement paste content and porosity. The results demonstrates that the crushing allowed producing a VSI-sand with low porosity and water absorption due to the increasing on the content of silicates from natural rock phases and increasing liberation between these phases and porous cement paste; the particles from the attained product presented also a better shape (sphericity and aspect ratio). The attrition tests assayed on VSI recycled fine aggregates accomplished a further selective disaggregation of cement paste and other porous materials - the porosity was reduced from 10% to 7%. The concentration by shaking table at narrow sieve fractions led to a product with low porosity and reduced content of cement paste, however, concentration on spirals did not succeed since separation was performed essentially by the size of the particles. The heavy product from shaking table represented almost 75% in weight and has less than 60% recovery of CaO+LOI, and expressive reduction on porosity. The aim of producing sand from construction and demolition waste through mineral processing unit operations changes the recycling approach and contributes for up cycling the recycled sand. The exhaustion of sand deposits on Brazilian large centers, the increasing price and the construction boom encourages the development of alternative materials as construction raw material. Keywords: recycled sand, construction and demolition waste, technological characterization

Sep 1, 2012

By Y. M. Zhao

The dry coal preparation technology based on air dense medium fluidized bed is an innovative dry highly efficient technology of coal by introducing fluidization technology into beneficiation.. A modularized 40-60 tones/hr system (KZX-40) with an air dense medium fluidized bed separator was established. The construction cost of the modularized system is only 1/3 of the traditional wet heavy-medium coal preparation plant construction cost with the same capacity, and the consumption of dense medium can be reduced by 50%-60%. The footprint and volume of the system are decreased by 75% and by 89.81%, respectively. These facts show the improved intensity, efficiency and economy of modularized coal preparation technique. The modularized KZX-40 system can be used to separated 50-6 mm coal over a separation density range of 1.3-2.2g/cm3 depending on the variation of <1mm fine coal content. The separation efficiency of the system was investigated at low and high separation densities by float-sink experiments. The partition curve of the products at low separation density shows that the separation density (d) is 1.33g/cm3, and the probable error (E) value is 0.05g/cm3. Using 50-6mm coal with an ash content of 55.35% as feedstock, at a high separation density of 1.61g/cm3, the clean coal with an ash content of 14.67% and the tailing with an ash content of 71.26% are produced with an E value of 0.06g/cm3. It is clear that the modularized KZX-40 system is suitable for separating 50-6mm coal effectively. During the dry coal preparation process, the dust is removed by the bag collector and recycled back into the separator. So the system has the advantages to work without environmental pollution and water. The modularized system greatly reduces the construction and operating costs and decreases the area and volume required, compared with wet dense medium system. The system is applicable to the coals located in arid regions and also to the coals that tend to slime during wet separation. Keywords: modularized, KZX-40, fluidized bed, beneficiation

Sep 1, 2012

Over the last four years, a major collaboration between CSIRO-MIS, UCT and JKMRC has produced two significant advances in the modelling of comminution. The Universal Comminution Model (UCM) uses parametric SPH/DEM studies of comminution and separation devices to calibrate mathematical models which are suitable for industrial use. The Virtual Comminution Machine (VCM) uses the same computationally intensive techniques based on discrete element modelling (DEM) and smoothed particle hydrodynamics (SPH) to directly model any comminution device by combining a CAD file describing the machine design, sophisticated ore characterisation methods and advanced particle scale flow and breakage simulation. The UCM should be useful for process analysis and optimisation of existing equipment and circuits while the VCM should make a major contribution to the evaluation and development of new machines. Progress to date is encouraging with several processes now well described using DEM. This paper also reports on progress towards combining DEM for particles with a realistic slurry model based on SPH for different types of comminution devices. In due course, this combination should allow accurate modelling of particle transport in wet mills to be combined with DEM models.

Jan 1, 2014

Producing robust agglomerates with desirable structural properties is the key to success of the heap leaching of complex, low-grade nickel laterite ores. Mechanically strong but porous agglomerates with narrow size distribution (e.g., 5 ? 40 mm) are required to build permeable and perennially stable heaps of 4 ? 10 m height. The characterization of granules? microstructure is essential for greater understanding and correlation with the concomitant mechanical properties and robustness. In this study, we used non-destructive, computed X-Ray tomography analyses and scanning electron microscopy (SEM) in tandem with conventional compressive strength,and integrity/stability tests to draw links between nickel (Ni) laterite agglomerate strength/stability, internal microstructure (e.g.,porosity, inter-particle bridges) and binder type. It is shown that binder composition (e.g., tap water vs. 30% w/w H2SO4) and post-agglomeration drying both significantly impact on granules? physicochemical and microstructural properties. Acidic binder, in contrast with tap water, leads to significant incongruent leaching of different species from particles during ore-binder mixing and agglomeration. Upon wet agglomerate curing, solidified/recrystallized leached species form strong intra-granular solid bridges which, lead to greater compressive strength of dry agglomerates. Structural stability/integrity of cured agglomerates in acidic solution, however, decreases dramatically due to greater porosity and acid solubility of intra-granular solid bridges. These findings are useful in shedding more light on agglomerate structure characteristics and their optimization for enhanced column and heap leaching with concentrated sulphuric acid.

Sep 1, 2012

By M. Roy

Coke combustion during sintering process is significantly governed by coke breeze size distribution, which in turn affect sinter quality i.e. sinter granulometry, Tumbler index, Reduction Degradation Index and Reducibility Index of sinter. To know the effect of microfines of coke breeze on sintering process parameters, pot sintering tests were conducted and it was found that when microfines are eliminated and optimum coke size (1-3mm) is used yield of sinter increased by more than 7%. Similarly productivity increased by 9%, the strength of sinter i.e. ISO Tumbler Index increased by 2.3% and +10mm sinter size increased by 5.5%. Controlling of microfines during crushing and by improving crushing index will result in improved quality and productivity of sinter with lesser amount of coke breeze. This paper describes that how to optimize solid fuel i.e. coke breeze consumption in sintering process through granulometry control. Keywords: iron ore sintering, coke breeze granulometry, productivity of sinter, quality of sinter

Sep 1, 2012

According to the Darcy?s law, filtration rate should increase with increasing pressure drop (?P) across a filter cake. In centrifugal filtration, ?P is created by the free water present on the surface of a filter cake. Thus, ?P can become negative when the free water disappears from the cake surface, which makes it difficult to achieve low cake moistures. In the present work, we have developed methods of overcoming this problem by either increasing the air pressure inside or applying vacuum outside a centrifuge. Laboratory dewatering tests conducted using these provisions showed that the moistures of various minerals and coal fines can be reduced to the level that can obviate the need for thermal drying. The new concept has been successfully tested both at pilot- and full-scale, which lead to commercial deployment through Decanter Machine, Inc., under the trade name Centribaric® centrifuge. Keywords: dewatering, hyperbaric centrifuge, pressure drop, Darcy?s law, Centribaric® centrifuge

Sep 1, 2012

By K. Anand Rao

Extensive exploration for uranium metal by the Department of Atomic Energy in India led to the discovery of new carbonate host rock uranium deposits Tummalapalle and Gogi, which are located in southern part of India. Alkaline leaching at elevated temperature and pressure using autoclave is the central unit process in the overall strategy for extraction of uranium from these new ore deposits. Autoclave leaching is affected largely by intrinsic nature of the ore and by the process variables and the equipment settings. In order to understand such a complex process and also to be able to predict the process performance under randomly (ore grade, uranium mineralogy, gangue mineralogy etc.) and consciously perturbed variable settings (partial pressure of oxygen, temperature, concentration of leachant, stirring speed and time), sound and robust mathematical models are needed with analytical and/or numerical solutions. Present investigation attempts to develop a variant of kinetic correlation based on shrinking core model for batch alkaline leaching process in an autoclave. The leaching model incorporates particle size distribution according to popular Gates-Gaudin-Schuhmann equation, rather than assuming a single mean size for all the particles participating in the leaching reaction. Kinetic experimental data generated in autoclave leaching of Gogi uranium ore was used to evaluate various constants of the model developed. The ore sample used for experiments assayed 0.21 per cent U3O8. The partial pressure of oxygen was varied between 0.15 to 6.5 atm, temperature from 343 K to 438 K, concentration of sodium carbonate leachant in the range 0.47 g mol/L to 0.94 g mol/L, and the stirring speed from 573 rev/min to 900 rev/min in the experiments. The goodness of fit of the leaching process model is found to be >0.95. Morphological studies carried out on leached and unleached particulate matter using Scanning Electron Microscopy with Energy Dispersive X ray analyses are included in the present study.

Sep 1, 2012

By M. F. He

There are three polymorphic modifications in pyrrhotite (monoclinic, hexagonal and orthorhombic), with monoclinic and hexagonal polymorph being more abundant. The floatability of different crystalline structures of pyrrhotite (monoclinic and hexagonal) is studied in this paper. It is shown that the floatability difference of monoclinic and hexagonal is obvious and that the flotation recovery of monoclinic pyrrhotite is greater than that of hexagonal pyrrhotite using different collectors . Using butyl dithiophosphate as collector, the flotation recovery of pyrrhotite is higher than when sodium diethyl dithiocarbamate and sodium butyl xanthate are used as the collectors. Monoclinic and hexagonal pyrrhotites are more easily activated by Cu2+ in acidic conditions than in alkaline conditions, while Cu2+ can not activate hexagonal pyrrhotite when sodium diethyldithiocarbamate was used as the collector. By the measurement of contact angle, it is indicated that monoclinic and hex-agonal pyrrhotite float well and are easily activated by Cu2+when dithiophosphate was used as the collector. Using sodium diethyl dithiocarbamate as a collector, the flotation behavior of pyrrhotite was investigated. The relationship between potential and pH range for pyrrhotite flotation was established. The results show that the flotation of pyrrhotite is dependent on pulp potential at certain pH values.

Jan 1, 2014

By M. Saghaeian, R. Hejazi, R. Barreiro, M. Tavousi, M. Ahadi, O. Asghari

"Banded iron formations are the major iron source for producing high grade concentrate. BIF are generally lower grade and more difficult to extract. In order to know how best to process BIF ore, it require comprehensive knowledge on ore bodies even in the early stages of the mining process. Geometallurgy combines geological and mineral processing information and allows for higher flexibility in response to changes in the plant feed when mining different parts of the ore. This can lead to economic benefits, owing to improved plant operation, yet is time consuming and correspondingly costly. The main focus of this paper work has been to develop a comprehensive geometallurgical testing framework for developing Yogi BIF low grade magnetite deposit located at Western Australia. In order to provide successful geometallurgical mapping, was collected data from several different disciplines (geology, chemistry, mineralogy) and parameters (physical properties, metallurgical response and geotechnical measurements) during drilling activities and test work. The investigation shown magnetic separation behavior of material had high variation and is most important element in the proper design of ore beneficiation plant. This paper provides an improved and efficient magnetic separation test methods used for the investigation of separation behavior and variability of the Yogi Ore deposit and evaluates their applicability and potential in the geometallurgical context. In this paper a small-scale magnetic separation has been developed that allows estimating from single pass separation test using small amounts of sample material. Verification of the evaluation method and validation was done with different mineral systems."

Jan 1, 2018

By T. M. Bhatti

The main purpose of present study was to characterize the dissolution of uranium from low-grade black shale with indigenous isolated strain of Acidithiobacillus ferrooxidans (BSAF-01). The nature of shale is quartzite-schistose and contains uranium content of 0.005% U3O8 (50 ppm U3O8). Uranium is present in the tetravalent oxidation state (U4+) in the shale matrix. The main minerals identified are graphite, quartz, pyrite, mica minerals (phlogopite, biotite, sericite, and chlorite), microcline, K-feldspar and kerogen (hydrocarbon-compounds). Bacterial oxidation of pyrite was an acid-generating system and produced sulfuric acid and ferric sulfate during this biological phenomenon. A series of bioleaching experiments were conducted in shake flasks for leaching of uranium from black shale. The leaching experiments were performed at 50% pulp density using indigenous Fe- and S-oxidizing bacterium (Acidithiobacillus ferrooxidans). Bioleaching data revealed that about 90% ~ 95% U3O8 was leached out from black shale ore. Uranium dissolution from shale depends on pH (pH 1.5~1.9), redox potential (=500 mev) and the concentration of Fe2+ and Fe3+ ions in the leaching environment. The oxidative leaching of U4+ involves a redox reaction with Fe3+ as an oxidant and bacterial re-oxidation of Fe2+ in acidic leaching environment. Uranium leaching efficiency was attributed to the sulfuric acid and ferric sulfate production during bacterial oxidation of pyrite.

Jan 1, 2014

By E. Barsky

The composition of multi-component materials is proved to be unambiguously determined by the composition uncertainty parameter (composition entropy). At the same time, every component, depending on its content, introduces a certain value into this parameter. Quality criterion of multi-component separation is determined by the total change of this parameter for all products obtained as a result of the separation process. This parameter can unambiguously estimate both multi-product separation and complicated schemes of modern beneficiation plants. It can be used to estimate binary mixtures separation, too. During the optimization of multi-product separation, another problem arises, which is not peculiar to binary mixtures. In this case, the total efficacy depends on the order of separation boundaries, as well. In the present paper, a specific example of initial product separation into seven products by six boundaries is investigated. It is clear that with respect to a single boundary, two products only can be obtained. In this case, the separation optimum is shown to correspond to the separation of boundary-class particles equally into both products. We have also established that from the point of view of entropy, the maximal possible separation efficacy corresponds to equiprobable composition of the initial mixture, i.e., the fine product quantity in the material to be separated is equal to that of the coarse product. Three groups of experiments were carried out. In the first group, the boundaries were consecutively specified from the finest to the coarsest one. In the second group, the boundaries were set vice-versa, i.e., consecutively from the coarsest to the finest one. In the third group, no sequence was observed. At every separation stage, a boundary was specified in such a way that the separated material composition was as close as possible to equiprobable one relative to this boundary. Total efficacy was different in all the three cases. In third group of experiments, its value is the highest, which confirms the advanced supposition. The obtained results are phenomenological, ? they can be applied to any processes of solid, liquid and gaseous materials separation.

Jan 1, 2014

By E. Scheepers

Within a phosphorus producing submerged arc furnace it was found that the continuous fluctuation of the furnace between a flowrate-driven state (high throughput) and a thermodynamic-driven state (low throughput) caused both techniques to have similar overall, predictive abilities and resulted in a linear, ARX-type, adaptive prediction model as the model of choice. Secondly, this pre-diction model was developed, tested and then shown to have a reasonable, 8-hour-ahead predictive accuracy (R2, coefficient of determination) of 30% (±6%) on future Pslag values. This inherent relationship exists because, at the precise moment of a Pslag prediction, the furnace already contains the metallurgical memory (input variables to the linear model) needed to ensure that some predictive possibilities will always exists - all as a result of the long residence times in the furnace. Residence time, however, is not a directly adjustable variable but rather a fully dependent variable and a function of an array of interconnected and interactive variables. In fact, this applies to virtually all input variables, reemphasising the importance of innate metallurgical memory. Thirdly, predictive control possibilities were explored by simulating the set-points of two fully independent variables used by operators to control the process: the ratio of fixed carbon-to-P2O5 and the ratio of silica gravel-to-pellets. This linear, predictive control model showed only a slight improvement with an 8-hour-ahead predictive accuracy of 35% (±7%). This highlights how ineffective current adjustments are in optimally steering the process and how difficult even incremental improvements in feed-forward and predictive control can be. Finally, it is shown that fundamental design-, sampling- and process restrictions currently associated with the process will always limit the predictive and especially control accuracy or meaningfulness of any dynamic model. These restrictions include the size of the furnace resulting in long residence times, 8-hours sampling intervals, an extremely complex and interactive process and a 16% spatial analyses variation on the Pslag values ? the very value that the model is to predict. The point is made that, given the current status quo, even the perfect dynamic prediction model can not improve on an 8-hour-ahead prediction of 30% (±6%). This barrier can only be pierced with e.g. tidier and more frequent sampling regimes and other upfront capital investments, a decision that becomes a cost accounting exercise and that can only be taken by the management structure. An investment demanding ever-increasing attention is CFD software and it?s potential to shed more light on the complex interactions within the furnace.

Jan 1, 2014