Search Documents

By MáAndrusiewicz, M M. Hilden, M S. Powell, P D. Kay, B Foggiatto

Circuits with unconventional and potentially more complex designs are predicted to become more common in the future as plant designers and operators look to process ores more efficiently. Ideas have been presented on how circuit design may be changed to utilise todayÆs technology to address tomorrowÆs challenges (Powell and Bye, 2009; Powell, Benzer and Mainza, 2011a). High pressure grinding roll (HPGR) installations in various configurations; application of preconcentration plants eg sorting, magnetic or gravity separation; and preweakening devices eg microwave processing; are all examples of technologies having the potential to deliver a significant reduction in comminution energy for ores amenable to these processes. Not only will future novel circuits contain equipment for which models are not as well developed (if at all), many will likely require the ability to respond to variations in feed, and interact with geometallurgical models requiring process adjustments when the ore characteristics change.In order to confidently simulate more complex circuits and accommodate variations in feed type, a significant advancement in modelling and simulation capability is required. This includes dealing with multiple components in the feed, accommodating different competencies of ore, utilising physical separation opportunities and tracking the grade reporting along the different potential processing routes. Developing the means to simulate complex circuits has been a key priority of a collaborative group of comminution researchers working to solve these challenges.Simulations are presented utilising a new configurable simulation interface called the model developers kit (MDK). The MDK is designed to accommodate multi-components and incorporate new models under development. These will be used to demonstrate the potential of flexible circuit design, based on measured ore properties and equipment performance.CITATION:Foggiatto, B, Hilden, M M, Powell, M S, Kay, P D and Andrusiewicz, M, 2012. Simulation of flexible circuits, in Proceedings 11th AusIMM Mill OperatorsÆ Conference, pp 291-298 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Oct 29, 2012

By Sylvain Duchesne, Claude Bazin, Robert Grant

Grinding simulators are now widely available to assist in the selection of circuit operating conditions or flowsheet designs that yield a given product size distribution. These simulators do not inform of the impact of the grinding product size distribution on subsequent flotation stages. This paper presents a method to predict the effect of fineness of grind on a flotation bank operation. Application of the method is illustrated with data from two Canadian concentrators.

Jan 1, 1995

By J Arola, B Kettunen, W M. Bond

Separation of fine and coarse particles as separate streams prior to flotation, and their differential treatments, could substantially increase recovery and reduce collector consumption. A series of simulations was carried out to quantify the benefits from plant scale implementation of such a treatment at the Ispat Inland flotation circuit (Minorca mine). The study involved detailed plant sampling and sample analysis to generate the data needed for simulation studies. Samples were taken from all the streams from flotation feed to final concentrate, as well as cell lips and cell pulp. Sampling was repeated for the second ore blend that was processed by the plant during different periods of a year. Sample analysis involved sieve analysis and size by size chemical analysis for total iron, magnetic iron and silica. Raw data were mass balanced, and flow rates within the circuit were calculated. Mass balanced data were used to calculate model parameters representing the existing conditions. Usim Pac was used for both mass balancing and simulation studies. Available pilot and bench scale data were utilised to define the modified flow sheet and expected performance. The simulated flow sheet included classification of flotation circuit feed by hydrocyclones. While the coarse fraction followed the existing flow sheet, the fine fraction was sent to a magnetic separator. Magnetic and rougher flotation concentrates were combined to form the final concentrate. Two simulations for each blend were carried out. The first one assumed that there would be no change in size by size kinetic behaviours of both gangue and magnetite due to this modification. This was named as the worst case scenario, in which particles had a longer residence time due to reduced flow rates within the flotation cells. The second scenario anticipated that there would be an increase in recoveries of gangue and magnetite going into froth launder due to more efficient use of collector in absence of fines. Results showed that pre-classification of flotation feed could improve the flotation recovery by 0.3 - 0.5 per cent, with the major benefit being significant reduction (40 - 60 per cent) in the amount of collector used.

Jan 1, 2005

By R. P. King

The two-phase distributed-parameter kinetic model for flotation has been wed as the basis for a digital simulation of flotation plants of arbitrary configuration. Ores that contain any number of minerals can be handled. The effects of incomplete liberation and particle-size distribution are specifically allowed for. The simulator allows the investigation of variations in throughput and in the slurry density of the feed, and of variations in stage volumes and aeration rates. It can handle cell behavior even under conditions of hindered flotation. The requirements for data preparation are completely documented so that the computer program is suitable for general-purpose use. The program is written in Fortran. The output from the simulator consists of the complete material flows in the plant, as well as the grade, recovery, and particle-size distribution in each stream.

Jan 1, 1976

By J. B. Rubinstein, V. J. Badenicov, S. T. Hall

A simulation technique of the flotation process and models for the solution of various problems are developed. A modular model of column flotation is constructed taking into account the kinetic relationship of the process and the hydrodynamic regime in the cell. Based on this, a technique of calculating the column parameters based on laboratory tests is developed (scaling-up), and applied for calcetest results is developed (scaling-up), and applied for calculating the design of commercial-scale columns.

Jan 1, 1996

By A. M. Engelbrecht

Paper written on project work carried out in partial fulfilment of B Eng. (Chemical) Membrane processes are vital to a diverse range of technical and biological systems, which also feature in mineral processing applications such as the treatment of mine effluent and extraction of metals. In this study, the use of cellular automata to model a semi-permeable membrane that separates two compartments is considered. Cellular automata are simple models that enable the analyst to represent well-understood local effects, such as the chemical or steric interaction between molecules themselves and between molecules or particles and their environment, which can lead to very complex behaviour on a macroscopic scale. Various physical phenomena, including reverse osmosis, were simulated. Initial results for the effect of temperature, membrane structure and chemical properties of the solutes on the performance of the membrane appear promising. Keywords: cellular automata, dynamic modelling, filtration, membrane fouling, osmosis.

Jan 1, 2007

By M. W. Hildyard, J. A. L. Napier

The improved design of support systems and stoping strategies when changes in depth, stratigraphy, or mine layouts are considered must rely on the designer's fundamental understanding of the deformation mechanisms adjacent to underground openings so that he can extrapolate and predict expected mining conditions. This paper presents a numerical approach to the modeling of the growth and interaction of mining-induced fractures near the edges of deep tabular excavations in brittle rock. Fractures are represented by displacement-discontinuity elements that are allowed to propagate according to prescribed growth rules. This procedure is justified by the reproduction of several features observed in laboratory simulations of fracturing around openings in blocks of rock subjected to uniaxial and biaxial compression. An attempt is then made to explain some of the attributes of the fracture zone surrounding tabular stopes. Particular questions of interest are the sequence in which fractures form and are mobilized in front of the mining face, the effect of parting planes parallel to the plane of the excavation, and the distribution of stress parallel to the hangingwall and footwall.

Jan 1, 1992

By Alan P. Prosser

The heap leaching of gold ores by cyanide solution was simulated by computer program. The program uses only basic m1neralogical and chemical data; no process testwork data, such as column leach results, are required. The simulation can be used to forecast the consequences of various changes before and during the testwork, design and operation stages of a particular ore-process system. Also, data about the state of different parts of the heap at various times help create an understanding of what is happening. The effects of mineralogical and process variables on the overall performance of the ?heap leaching process are illustrated. Comparisons are made to column leach tests and production data. The potential of the simulation strategy is discussed for making better use of operating experience and for suggesting specific research objectives. It has become clear that more attention should be paid to other metals that dissolve in the solution.

Jan 1, 1989

A universal grinding circuit model suitable for an arbitrary structure of flowsheet has been suggested. In this model classifiers and mills are described in a most general form. What puts the model in a class by itself is the automatic mass balancing in all streams of the circuit. In order to provide this a special classifier model is adopted. A previously published continuous steady state process model is used as a grinding model. Non-linearity of classifier model leads to a resultant circuit model which is equivalent to a non-linear integral equation. It complicates somewhat calculation algorithm in comparison to that of linear model, but provides a physically meaningful solution.

Jan 1, 1993

By B. Sonmez, B. Aksanl, S. Ersaym, I. F. Erkal, S. L. Ergun

The comminution circuit at the plant has experienced severe problems arising from a feed containing much larger amounts of clay and having a higher moisture content than expected. Three solutions were proposed: replacement of the crushing circuit by tooth roll-type crushers; using washing drums; and introducing primary autogenous/semi-autogenous millling instead of crushing. Computer simulations studied the effects of the first two solutions on the circuit capacity. Laboratory and industrial scale ball mill batch grinding tests were carried out. The parameters of a kinetic model were determined. The ball mill circuit was sampled and operating model parameters were calculated. The ability of batch grinding parameters to represent the operating conditions was studied. Tests failed to find acceptable model parameters, but parameters from an industrial size batch mill were found to be suitable for simulation studies. The simulations indicated that a continuous feed of -12 mm provided by tooth roll crushers would maintain the desired capacity. Increasing the feed size to -25 mm would reduce the capacity by 8-15%. The use of washing drums could also solve the capacity problem, but small fluctuations in water flow rates might result in problems of excess water in the grinding circuit.

Jan 4, 1993

By L. G. Austin, P. T. Luckie, R. R. Klimpel, R. S. C. Rogers

1. INTRODUCTION The Bond method(1.2) for sizing tumbling ball mills has been used successfully for many years. The primary purpose of the Bond calculation is to predict the mill size and mill power to give a desired capacity (Q tph) from a wet overflow ball mill in normal closed circuit operated with a circulation ratio of 2.5. The input to the calculation is the make-up feed size characterized by the 80%-passing size in micrometers (xG), the desired circuit product size characterized by the 80%-passing size (xQ), and an empirical grindability number (the standard Bond Work Index) determined by a standard test. A series of empirical correcting factors(2) are used to convert the results to open-circuit or dry grinding in grate discharge mills. The Bond method has two major engineering advantages: (i) it is very simple; (ii) experience has shown that it works for many circumstances, as a reasonable approximation, because it is based on direct conversion of test data to results measured on industrial mills. There are four main disadvantages of the Bond method. First, there will be a range of error for any specific mill and material and set of operating conditions because the method does not explicitly include several variables which are obviously important: (i) recycle ratio and classifier efficiency (ii) mixture of ball sizes in the mill (iii) variations of residence time distributions with mill geometry and slurry density (iv) lifter design (v) slurry rheology (vi) different underfilling or overfilling of the mill as flow rate is changed. Second, the Bond method explicitly assumes that specific grinding energy E is not a function of ball load, whereas industrial practice and laboratory tests(3) show that the specific grinding energy (to go from a specified feed to a specified product) is less for lower ball loadings than the ball loading for maximum mill capacity. Third, the method uses only the 80% passing size of circuit feed and product to characterize the size distributions, whereas mill capacity in general must depend on the shape of the feed size distribution and the product size distribution. The prime example of this is the use of the reverse closed circuit which removes some fraction of the make-up feed before it is

Jan 1, 1982

By Yusuf R

Many heap and dump leaching processes involve competitive, complementary and sequential reactions between multiple minerals and reagents. A comprehensive model has been developed to deal with these complex chemical interactions which is based on diffusion control of the reaction rates. All of the mineralogical and chemical data are obtained from representative samples of the ore or the chemical literature. These properties have been incorporated in a general way, making the model applicable to a range of ore-process systems. The simulation is used to predict the absolute reaction rates and can be used before any leach tests are carried out on the ore itself. The applicability of the model and some tests of its validity are presented.

Jan 1, 1984

By C. J. Barbee, P. J. Bethell, C. J. Wood

Heavy medium cyclones (HMCs) are commonly used to clean run-of-mine coal in the 50 x 0.5 mm size range. Unfortunately, field studies suggest that industrial HMC circuits are often poorly designed or improperly operated, resulting in significant losses of recoverable clean coal. To assist plant engineers in overcoming these problems, a spreadsheet-based simulation package has been developed to predict HMC performance based on mathematical expressions originally developed by Wood (1990). A case study is provided to illustrate the predictive capabilities of this simulation package for an industrial HMC.

Jan 1, 2004

By Ji Hoe Kwon

Hindered-settling column process has been used for the separation of mineral particles for many years. In this study, a model based on DEM (Discrete Element Method) was used to simulate the movements of particles in a settling column separation process. DEM model is capable of describing the motion of particles and calculating the impact forces between particle and particle, and particle and wall contact under various operating conditions. Drag force, rotation induced Magnus lift force, contact force and the gravitational force acting on individual particles were considered for establishing the mathematic models. The overall motion of the particles in water is satisfactorily simulated by the DEM, and the result shows good agreements with the result of lap-scale experiment.

Jan 1, 2014

By Harushige Kusumi, Yuzuru Ashida, Toshiki Watanabe, Toshifumi Matsuoka

ABSTRACT: The technique of various kinds of numerical analysis offered remarkable results to explication of many rock mechanical phenomena. These techniques regarded a rock stratum as a continuum body and solved the governing equation of continuum mechanics by using finite differ-ence method, finite element method and/or others. Using numerical analysis based on continuum mechanics, however, the simulations of destruction phenomena caused by elastic waves and on the other hand, the elastic wave propagation which generated by fracturing events are quite difficult. In this paper, we express a rock stratum by a granular body model and develop a simulation tech-nique can simulate both wave phenomena and destruction phenomena at once. The DEM (discrete element method) simulation results are compared with finite difference solutions of elastic wave equation. Furthermore, we simulate Hopkinson effect by this proposed method. Hopkinson effect is famous as a destruction phenomenon caused by elastic wave propagation. ABRISS: In dieser Studie drücken wird eine Felsschicht durch ein Granularkörpermodell aus und entwickeln eine Simulationstechnik, die sowohl Wellenphänomene als auch Zerstörungsphänomene simulieren kann. Die Ergebnisse der Simulation mit DEM (Diskrete-Elemente-Methode) werden mit denen der Finit-Differenz-Lösung von Gleichungen elastischer Wellen verglichen. Desweiteren simulierten wir den Hopkinson-Effekt mit dieser vorgeschlagenen Methode. Der Hopkinson-Effekt ist als Zerstörungsphänomen berühmt, verursacht durch Fortpflanzung elastischer Wellen. RESUME: Dans cette étude, nous exprimons un lit de roches par un modèle à corps granulaire et développons une technique, permettant de simuler d'un coup le phénomène des ondes et le phé-nomène de destruction. Les résultats de la simulation DEM (méthode d'éléments discrets) sont comparés aux solutions à différence finie de l'équation d'ondes élastiques. En outre, nous simulons l'effet d'Hopkinson par la méthode proposée. L'effet d'Hopkinson est connu comme phénomène de destruction, causé par la propagation d'ondes élastiques.

Jan 1, 2003

By Alexander Siefert, Steffen Pankoke

In the development of commercial vehicles as well as of passenger cars, the effects of vehicle vibrations on operating safety, health and comfort can only be predicted by numerical simulation when appropriate occupant models are available. Such models must be based on human anatomy and have dynamic properties of real humans in order to achieve realistic results. Since human dynamic behavior depend on posture and percentile, the occupant model needs to be adjustable to these parameters with respect to geometry and dynamic properties [1,2]. Dynamic Human Finite-Element-Model CASIMIR CASIMIR is a non-linear, dynamic finite-element-model of the human body. It consists of a dynamic model of the upper torso with head, neck, shoulders and arms as well as of a dynamic model of the lower extremity with pelvis and legs. The most important part is the lumbar area with dynamic non-linear models of the lumbar spine and of back and abdominal musculature. The frequency-dependent characteristics of the intervertebral discs and the effects of muscle activation and non-linear frequency-dependent muscle properties are included. In the latest stage of development, CASIMIR has been equipped with a compliant model of the body surface in the contact areas to the seat. This results in a very realistic transmission of static and vibrational forces into the human body, see fig. 1. Intense model verification and validation has been performed in all stages of model development, starting with validation of small components like intervertebral disc, ending with validation of whole-body-vibrations using measurements of the dynamic mass / mechanical impedance [4]. For an in-detail examination of stresses in the vertebral bodies and discs, a non-linear submodel of the lumbar spine with an increased number of degrees of freedom can be coupled to the whole-body-model, enabling the researcher to examine local effects of vibrations and single shocks on the lumbar materials. [ ] Since it is well known that human dynamic behavior is significantly affected by anthropometric data and posture, CASIMIR can be individualized to the anthropometric status of single individuals or to the mean values for specific percentile groups. Furthermore, posture can be adjusted to the seating conditions applicable to a specific vehicle. Posture modification capabilities include the variation of the lumbar lordosis [3].

Jan 6, 2006

By Kenji Matsuzaki

A mathematical simulation model of the Imperial Smelting Furnace operation has been developed for purpose of improving the furnace operation. The model consists of top, shaft and bottom regions, Analysis in the shaft region is based on differential heat and materials balances taking account of reaction kinetics and heat transfer rates, and those in the top and the bottom regions are based on heat and materials balances under a steady-state condition, Longitudinal distributions of process variables such as temperatures of gas, sinter and coke, gas composition and fractional conversion under a practical operating condition were calculated, With this model, the furnace performance under various operating conditions also can be shown reasonably predicted,

Jan 1, 2003

By Malsuzaki. Kenji

A mathematical simulation model of the Imperial Smelting Furnace operation has been developed for purpose of improving the furnace operation. The model consists of top, shaft and bottom regions. Analysis in the shaft region is based on differential heat and materials balances taking account of reaction kinetics and heat transfer rates, and those in the top and the bottom regions are based on heat and materials balances under a steady-state condition. Longitudinal distributions of process variables such as temperatures of gas, sinter and coke, gas composition and fractional conversion under a practical operating condition were calculated. With this model, the furnace performance under various operating conditions also can be shown reasonably predicted.

Jan 1, 2003

By M Phil Schwarz

"A CFD model for the gas-side of an impinging jet iron bath reactor has been developed to assist in the interpretation and design of oxidizing gas-melt reaction rate experiments. Heat transfer and reaction at the surface have been included. A free surface model, which solves for the liquid motion within the bath as well as the gas motion above the bath, has also been developed. This model also predicts extent and shape of the dimple formed on the liquid surface by the impinging jet. Both models have been run for conditions typical of jet impingement experiments using CO2 , H2O and O2• The main conclusions are• Results from the model for gas-side mass transport are in good agreement with the empirical correlation of Belton and Belton (1980)• Experimentally determined rates of decarburization by CO2 should be corrected for mass transfer because the experiments are often carried out under mixed rate control, eg for gas flow rate 10 L/min when the sulphur concentration is lower than 0.045• If the rate constant for oxygen reaction is greater than about 5 mol/m2.s.atm, it cannot be determined using a gas flow rate of 10 L/min or lower because of mass transfer control.• A cavity of depth 4.5 mm is predicted for a gas flow rate of 10 L/min and this substantially changes the flow field (and hence reaction rate) compared with a simulation in which the surface is assumed to be flat."

Jan 1, 1999

By E. Topuz, M. J. McPherson, I. M. Konduri

The research is carried out to investigate the flow mechanics and effectiveness of face ventilation using induction fans, as an effort to provide a more efficient alternative to conventional auxiliary duct or brattice ventilation systems. The induction effects and plume longevity of fluid jets are being studied with the help of a plexi-glass model, using water as the fluid medium and creating a system that is dynamically similar to the flow in mine headings. This scaled model represents a typical 1.5m x 6m x 25m heading connecting to a crosscut. Reynolds Number criteria are used to obtain dynamic similarity of the flow in the designed model to simulate real mining conditions. The role of various parameters such as nozzle position in three dimensions, airway roughness, nozzle geometry, and flow rate in the jet are studied. During the laboratory tests, flow visualization techniques, coupled with still and video photography, were employed to identify the flow patterns. The results indicate that in typical coal mine headings, jet fans can produce effective face ventilation for distance up to 30 meters.

Jan 1, 1995