Search Documents

By S. Simukanga, J. Masinja, P. A. Bepswa, A. N. Mwale

Screening is an important operation in dry and wet closed circuit mineral processing applications. Most of the operations have seen the benefits of using fine wet screens in primary grinding sections where the product is relatively coarse, and screens can handle the tonnage. Although many models are available, most of them are confined to dry and wet coarse screening applications. Most models do not give good predictions for wet fine screening. To facilitate model development, pilot plant tests were performed using a single deck Derrick Screen at aperture sizes of 45µm, 75µm, 106µm, and 150µm. Experimental data for a range of feed solids concentrations and flowrates was collected and used to calibrate the performance of the screen under the different operating conditions. The results showed that the performance of the screen deteriorated with an increase in feed solids concentration, and, similarly with an increasing feed flow rate. The percentage of fines carryover also increased with the reduction aperture size. The carryover was even more pronounced at higher feed rates and solids concentrations. A model was developed for predicting the performance of fine wet screening. Data from the test work was used to compare the actual and predicted results, and to calibrate the model. Additional independent data was used to validate the model and assess its limitations. This model will give an expected screen performance based on the initial feed conditions. It incorporates screen design parameters that allow for an estimation of a suitable screen size for a desired performance based on given feed conditions. This paper discusses; the results from the test work performed, the model development and its limitations.

Jan 1, 2016

By H. J. McQueen

Friction stir welding (FSW) generates complex microstructures in the nugget and thermomechamically affected zones (TMAZ) that have led to diverse explanations. However, consideration as an extension of microstructure development in piercing/extrusion can lead to a better understanding of flow patterns and microstructures during FSW. Furthermore, the intense strains in the nugget/TMAZ can be interpreted by comparison with ultra-high strains in torsion that exhibit dynamic recovery as extending into regions with high angle boundaries strongly serrated with spacing near the subgrain size defined by the high temperature. The subgrain diameter and extrusion pressure have been shown to be dependent on the Zener-Hollomon parameter Z (=t exp(Q/RT), Q activation energy, R gas constant) combining temperature T and strain rate E in the same mariner as for torsion and compression. Moreover, such tests exhibited steady state stress os above a certain strain Es (rising as Z increases) in which T, , subgrain diameter and dislocation spacing remain constant. Due to the subgrain boundaries (SGB, low angle <5°, i.e.,LAB), continually rearranging by migration, decomposition and reformation, the subgrains remain equiaxed in elongating grains, whose boundaries (GB) become serrated as they migrate to absorb SGB. As confirmed by straining in torsion at constant T and Z. , more subgrains bordering the greater GB area have several high-angle facets (HAB) ; these small cellular regions with a mixture of HAB and LAB facets are designated HAB-subgrains. As serrations in neighboring grain boundaries impinge, the grains pinch off becoming shorter; this occurs at higher strains as subgrain size decreases with rising Z. The pinching-off makes the neighbors thicker so the grains never become thinner than about two subgrain diameters. In conjunction with such geometric dynamic recrystallization (gDRX), GB&apos;s undergo a net migration into grains with denser substructure, enlarging those with larger subgrains. Consequently a better name has been proposed: grain dimensioning dynamic recovery (gDRV).

Jan 1, 2007

By Viswanathan N. Nurni

In metallurgical processes, more and more usage of hydrocarbons is encouragedto bring down the carbon emissions. In this regard, numerous investigations onreduction of oxides by C-O-H-N gas mixture have been reported. Attempts to simulatethese reduction processes using shrinking core model, one of the commonmodels used for such studies, have under predicted the reduction rates. This maybe owing to the fact that the homogeneous reaction in the gas phase is not being considered. If the reaction temperatures are above 1000K, generally so for manyreduction processes, the homogeneous gas reaction rates are expected to be sohigh that one can assume local equilibrium in the gas phase. In the present study,reduction of wustite in a C-O-H-N gas mixture has been modeled using shrinkingcore model considering the water gas shift equilibrium in the gas while it diffusesthrough the product layer.

Sep 1, 2012

By D. J. Varnes

Rocks in mines, quarries, and many outcrops commonly show evidence of being under high stress. Saw cuts and drillholes close in, partly mined coal bursts violently, and pillars crush and rock spalls in mines even at moderate depths. Similarly, strong and massive rocks such as granite and sandstone naturally divide themselves into sheets that lie more or less parallel to their outward topographic form. The sheets may be either convex or concave. Thin plates of bare rock surfaces bow up and buckle. Such effects often cannot be explained by the obvious loads now acting on the rock, that is, by loads resulting from overburden, topographic irregularities, or stress concentration around openings. The stresses exceed those that would result from such loads. Two interpretations of such excess stresses have been made. In one, the stresses are assigned in origin to now active tectonic forces. In the other, they are regarded as leftover and locked in from some previous state at which higher pressures prevailed. These two sources may both operate and they generally cannot be distinguished easily in the field. There are other complicating sources of stress such as temperature gradients and chemical alteration. Nevertheless, some bodies of rock, by exfoliating, show evidence of high internal stress even though they are practically unweathered and so isolated topographically that the presence in them of significant stress due to exterior loads or active tectonic forces seems unlikely (Fig. 1). Moreover, completely isolated rocks are known to change density, shape, or size 1,2 to expand under constant compressive stress;3 and even to disintegrate 4,5 without the intervention of weathering processes. The stresses involved here must be truly residual in the sense long used by metallurgists; that is, residual stresses in a body are those that remain.

Jan 1, 1970

By Young-Sool Jin, Heung Nam Han, Hong Joon Kim, Jae Kon Lee

A mathematical model was developed considering non-symmetric cooling' and stress distribution in both thickness and width direction of strip on a run-out table of hot strip mill. In order to solve a transient heat transfer equation including the heat evolved from phase transformation, a finite element method coupled with thermodynamic and kinetic analyses was applied. The heat capacity of each phase and heat evolution due to phase transformation were obtained from the thermodynamic analysis of the Fe-C-Mn-Si system based on a sub lattice model. The phase transformation kinetics of the steels was derived by using an inverse additivity technique with continuous cooling experiments. Heat transfer coefficients of strips on the run-out table were, by applying an inverse method, determined from the actual mill data under various cooling conditions. As for the stress analysis, the density change of strip due to cooling and phase transformation, and the transformation superplasticity were considered. A 90nstitutive equation for the transformation superplasticity, which is related to the phase transformation kinetics and the internal stress due to the density change of strip, was applied. A finite element method was adopted to calculate the deformation behaviour of strip on a run-out table of hot strip mill. Using the models, the temperature-phase-deformation history and distribution of a strip on the run-outtable was calculated.

Jan 1, 2001

By D. R. Zhu, H. S. Chiang, Y. M. Jiang, S. S. Peng

Two types of underground tests on three types of shield supports, 2-leg, 4-leg, and chock shield, were con- ducted to measure the horizontal movement of the canopy against the immediate roof. In no friction tests, steel rollers were placed between roof and canopy top, while in friction tests normal direct contact was used. Results indicate that the canopy for 2-leg shields always moves forward during setting, whereas there was no forward movement for 4-leg or chock shields. Based on the test results, a model for active horizontal force was proposed.

Jan 1, 1990

By R. P. King

"A model of the flotation process can be used effectively for design and control. For design purposes it can be used to predict the performance of a large plant from laboratory and pilot plant data. It can be used for control purposes to assess the utility of any proposed operating modification to an existing plant, and it can also be used to determine optimal control settings for best operating performance.A model has been developed that will allow for variations in feed properties and for variations in fineness of grind. The model can be used to assess the effects of reagent addition rate, aeration rate and pulp level on plant performance and these are the control variables that are used individually or together in all operating flotation plants. The model provides for a distribution in the values for flotation rate constants, particle size and mineral fraction, and, because of its complexity, requires a large computer when it is to be applied to a typical plant configuration. The computational method that is used to apply this model to such configurations is described in this paper, and particular attention is paid to the problem of allowing for feedback of tailings from cleaners and recleaners."

Jan 1, 2014

By M. Pérez-Tello

A mathematical model to represent the expansion and fragmentation of copper matte particles oxidized under flash converting conditions is presented. The model assumes that the particles are initially nonporous, have a constant mass prior to fragmentation and travel at a constant velocity throughout the reaction chamber. The model requires the specification of the following five parameters: the particle expansion rate, a fragmentation diameter factor, a fragmentation size distribution parameter and the fractions of the finest and the coarsest particles in the feed that undergo fragmentation. The model predictions show good agreement with the experimental data collected in a laboratory furnace over a wide range of experimental conditions. The evolution of the size distribution of the particles along the reactor length was computed, and the model parameters were correlated with the experimental operating variables. Model predictions indicate that particle residence time is an important factor in the generation of dust. The presence of two maxima in the particle density function may be attributed to turbulent conditions prevailing in the furnace, which cause particles to follow different trajectories within the furnace even if they are injected at the same location.

Jan 1, 2008

By Ralph W. M. Lai, D. W. Fuerstenau

In aqueous solution, an oxide surface is considered to consist of positively charged surface sites, MOH, neutral surface sites MOH, and negatively charged surface sites MO-. From the mass action law, equations for the distribution of these sites have been derived in terms of the hydrogen ion concentration in solution and the point of zero charge of the oxide. Experiments on the flotation of futile were carried out with sulfonates and amines which are collectors that physisorb, and with oleate, which is a collector that chemisorbs. Flotation response was found to correlate with the site distribution model. A similar correlation was found with published data on iron oxide potation.

Jan 1, 1977

By C. B. Manula, R. L. Sanford, R. A. Rivell

Data input is divided into three divisions for discussion purposes below. These divisions are: a) Information Retrieval and Control Options; b) Gaming Options; and c) Descriptive Parameters. The data input for model application, which is listed in Appendix B, is segmented to show the location of these divisions. In actuality, however, these divisions form a continuous data file. Each 80 character record of data input is edited by an alpha¬numeric editing routine which permits the format free numerical speci¬fication in most data divisions. The editing routine searches each record for numeric types and when found, converts these numbers into computational forms. The Fortran convention of decimal and integer numbers is not used by the editor so that a number may appear on the data record in either form desired by the user. Since the editor seeks only numeric types, the input records may be freely commented and identified with alphabetic letters (A - Z). One or more input records may contain only commentary information or be entirely blank. On the other extreme, any data record may contain up to 40 numeric elements, each separated by a blank space. The application of this capability can be seen by reference to the data. Information Retrieval and Control Options The information retrieval options specify the detail of infor¬mation required to be printed. These have been placed in a separate data division, identified on the data record by option name for use

Jan 1, 1974

By C. B. Manula, R. L. Sanford, R. A. Rivell

A generalized flow diagram is presented in Figure 1, which shows the relationship of major model functions. This chart does not necessarily represent the anatomy of the computer program, but is designed to illustrate the informational interchange between major model functions. The related flow-charts-and computer program are listed in Parts I and II, Appendix A, respectively. Mine Car Stations The mine car stations are the activity centers for the model. The locomotive logic structure is designed to provide these stations with adequate cars and to clear them of car buildup. The model is designed to simulate a three stage haulage situation which suggests the need to differentiate station levels by type numbers. These are: a) TYPE 1 stations - mine cars are loaded here; b) TYPE 2 stations - intermediate stop and storage for mine cars; c) TYPE 3 stations - all mine cars that are stored here leave mine; and d) TYPE 4 stations - mine portal. TYPES 1, 2, and 3 have all loaded cars assigned to an ascending TYPE as their next destination. In input, if single stage haulage is designated, TYPES 1 and 3 do not exist, and if two stage haulage is called for, TYPES 2 and 3 are absent. A motor is not permitted to exchange cars between stations of the same type. Station Status Update. One major model function is to continuously evaluate the status of each mine car station. The major evaluations

Jan 1, 1974

A mathematical model of a decision is represented by the table in Figure 4. The numbers in the matrix are a measure of the effectiveness of each alternate for an objective. In this case, it is the per cent of desired profit that is satisfied by each alternate for each objective. By adding all of the values in a row, we have a measure of which alternative is best for both objectives. However, if you think that alternate 1 is necessarily better than alternate 2, you are wrong. Why? Because the weight of the objectives may not be equal. It could just be that this plant can sell more than it can produce, so that the loss in profit represented by increased sales is more than twice as important as reductions in costs. (See Figure 5). In that case, let us weigh objective 1 as 30 per cent and objective 2 as 70 per cent. It must also be obvious that the sum of all the "objective weights" must equal 100 per cent. Now, by evaluating our alternates according to their weights. we find that alternate 2 is better than alternate

Jan 1, 1964

By Valery M. Yubko

"The conceptual basis of the developed model is based on the ideas that evolution of geological-geomorphological structure of the CCZ is controlled by four types of geologic factors: geodynamic, sedimentary, volcano-tectonic and erosion-lithodynamic. In terms of time span two factors (geodynamic and sedimentary) can be considered as permanent and the other two as sporadic.The geodynamic factor is responsible for two interrelated geological processes. One is going under conditions of the axial zones of spreading centers and results in formation of the CCZ basement. Historically this process is considered as a starting point for each of the heterochronous fragments of the CCZ. The second one controls gradual shifting towards the Pacific Plate (Hawaiian direction) and subsidence to greater depths of the neogenic fragments (basement). Thus, those processes control basic topography, geomorphology and the seabed depth within the CCZ.Volcanogenic-hydrothermal activity of the EPR is one of the main sources of the ore components, which finally compose manganese nodules as a result of complicated processes of their dissemination and transportation to the bottom in dissolved and organically bound forms.The role of the other constant factor (sedimentary) is displayed in two ways. Firstly, initial relief is smoothed under the influence of this factor, which, subsequently, makes favorable conditions for manganese nodules formation. Secondly, it favors transportation of the majority of the ore components, participating in manganese nodules formation into zones of active nodules formation."

Jan 1, 2017

By Masahiro Hirasawa

"As a fundamental study on pyrometallurgical process that involves recycling steel-making dusts, of which major components are oxides of Zn and Fe, and waste polyvinyl chloride simultaneously ·at. a temperature below 1273 K, the reactions between the mixture of ZnO + Fe20 3 and PVC or the compound ZnFez04 and PVC are investigated in a series of thermogravimetric and small scale reactor experiments under inert atmosphere at below 1273 K. The reaction mechanisms are discussed· on the .basis of the thermogravimetric data. From the results of reactor experiments, it is shown that the chlorination of Zn in the oxides by Cl in PVC, the vaporization of ZnC12, the reduction of iron oxide to metal Fe by C and H in PVC, and, thereby, the separation of Zn and Fe in the oxides take place simultaneously at the temperature ~= 1200 K.IntroductionSteelmaking dusts are metallurgical wastes consisting mainly of oxides of Zn and Fe with smaller amounts of other heavy metal elements, such as Pb. In the recent investigation, [1] the present authors have investigated the fundamental aspects of a potential pyrometallurgical process· of recycling steel-making dust with polyvinyl chloride (PVC) at temperature T < 1273 K under Ar atmosphere. The performance of a similar process, in which HCl in the combustion gas of PVC is used to chlorinate Zn in the EAF dust under oxidizing atmosphere, has been investigated by Tailoka and Fray [2] with fundamental experiments. Figure 1 shows an example of the potential ·diagram of Zrt-Cl-0 and Fe-Cl-0 systems. In the figure, we can find two regions where ZnCl2(g) can be separated from Fe containing compounds. One is ZnCiz-Fe20 3 coexistence region at higher oxygen and chlorine potentials and the other is ZnCl2-Fe coexistence region at lower oxygen and chlorine potentials. The process adopted by Fray and coworkers is aimed mainly at the performance of Zn/Fe separation in the higher oxygen potential region. In the present study, the lower oxygen potential region is focused on."

Jan 1, 2000

By J. Dubinski

Underground mining of copper ore in the Legnicko-Glogowski Copper District basin generates a number of seismic events, whose energy often exceeds 108 J (Butra, 2010). The number of tremors and their location shows that the area of rock mass involved in the deformation and failure processes significantly exceeds the area of mining operations. Locations of seismic foci of high energy tremors, as well as the aftershocks that accompany them, extend hundreds of metres, confirming that processes to restore the energy balance of rock mass occur in a short amount of time and over a wide area. Thus, an area of similar size should be considered in predictive analyses of the tremor?s position. It should be emphasized that a large portion of the rock mass still has a level of freedom resulting from the way the rock mass is held together and its geometric continuity. This level of freedom is dependent on the scale and duration of mining, the number and size of natural structures, and rock mass pressure, including gravitational pressure constricting individual fragments of rock mass in the area (Burtan, 2012). It can therefore be assumed that in real conditions, areas where mechanical strength of the layers has been exceeded may be located a considerable distance from each other and do not necessarily need to coincide with the working area location. In this sense, a rock mass influenced by mining exploitation is essentially a rock mass in which numerous fragments with concentrated energy exist simultaneously. Energy will be released within these rock mass areas each time rock mass strengths are exceeded by the induced stress changes related to mining operations. This paper presents a discussion on how to predict the location of seismic activity areas based on a change of position of seismic energy equivalents. This parameter is calculated for a given set of seismic tremors with defined foci coordinates. For a given set of seismic tremors, the location of seismic energy equivalents may be used for predicting the location of high-energy seismic events.

Aug 1, 2013

By A. G. Pentland

MODELS of mines may be classified under three headings: (a) sheet models, (b) solid models, and (c) skeleton models. Sheet Models are made of sheets of glass, celluloid, or a transparent plastic material. The sheets are mounted in a frame which holds them in a parallel position, either horizontally or vertically. Mine plans are traced on the horizontal sheets; sections on the vertical sheets. The model may be illuminated by electric lights which are separated from the sheets by an opalescent screen. It is viewed from above or from the side in such a way that the plans or sections are superimposed. Sheet models are very useful for the illustration of mines with narrow, steep-dipping veins, but they do not give a satisfactory representation of a large mine with a flat-dipping vein. Large sheets of glass are difficult to handle and to support in a frame.? It is difficult to see through a considerable number of them. Solid Models are made of plaster, concrete, wood, or other material usually in the form of slabs in which the mine workings are indicated by properly dimensioned cavities in the model. The slabs are coloured to show the veins, country rock, faults, etc. The model is divided along selected sections in order to give access to the interior by the temporary removal of parts. Solid models may be used to illustrate large orebodies of the Sullivan type, but they are cumbersome because parts must be removed to study details of structure. Skeleton Models are ones in which the underground cavities are represented by metal, wood, plaster, or wax proportionately shaped and held in place by a frame. They illustrate the work which has been done but do not show the position of the veins, dykes, faults, etc., in the unworked parts. It is necessary to project by eye between workings in order to determine the location of these features. The essential difference between the skeleton model and the solid model is that the skeleton model represents the mine workings by solid forms and does not show the unworked parts of the vein, whereas the solid model shows the vein as a solid and the mine workings as cavities in it. The skeleton model lends itself well to the representation of a large mine with large stopes.

Jan 1, 1944

By Florian Kongoli

Non-ferrous smelting slags .can contain magnesia, which substantially increases the liquidus temperature. One option to counter this increase is to flux with calcia. In this work, experimental liquidus measurements including new measurements on industrial slags have been compared with a FACT-based thermodynamic model, which predicts the liquidus surface of multi-component iron silicate smelting slags containing both magnesia and calcia. The predictions, which agree well with experimental results, are summarized in both isothermal and non-isothermal diagrams to give a quantitative description of the liquidus surface.

Jan 1, 1998

By D. G. Almond

Nkomati Nickel Mine in South Africa processes one of the world?s hardest ores in a 375k TPM autogenous grinding circuit. The plant employs an unconventional processing circuit, comprising a primary autogenous mill (AG or FAG) and pebble crushing circuit followed by a secondary hybrid pebble/ball mill and classification cyclone cluster. Advanced Process Control (APC) or Expert Control is now generally accepted as an advantage in most modern high capacity concentrator grinding circuits. This paper reviews the installation of model predictive control at Nkomati and examines the results and benefits of installation in an autogenous mill during plant ramp-up. The paper also demonstrates how predefining process control objectives together with good plant and supplier cooperation can achieve a robust and reliable APC implementation on a very short site program. The results conclusively demonstrate operational and economic benefits during circuit operation with the advanced process control system ON. Production performance and circuit stability were significantly better than during manual control.

Jan 1, 2012

By K. Becerra, D. G. Almond, T. D. Maru

"Nkomati Nickel Mine in South Africa processes one of the world’s hardest ores in a 375k TPM autogenous grinding circuit. The plant employs an unconventional processing circuit, comprising a primary autogenous mill (AG or FAG) and pebble crushing circuit followed by a secondary hybrid pebble/ball mill and classification cyclone cluster.Advanced Process Control (APC) or Expert Control is now generally accepted as an advantage in most modern high capacity concentrator grinding circuits. This paper reviews the installation of model predictive control at Nkomati and examines the results and benefits of installation in an autogenous mill during plant ramp-up. The paper also demonstrates how predefining process control objectives together with good plant and supplier cooperation can achieve a robust and reliable APC implementation on a very short site program.The results conclusively demonstrate operational and economic benefits during circuit operation with the advanced process control system ON. Production performance and circuit stability were significantly better than during manual control."

Jan 1, 2012

By Vincent C. Smith, Agit Singh, Daniel de V. van der Spuy, Bernard Muller

"Process control in the minerals industry is made difficult by factors such as the multivariable nature of the processes, noise, many disturbances, the changing nature of the processes, nonlinearities, process constraints, slow responses and large deadtimes. A good example of such a process is a milling circuit, where all these problems are experienced. Since model predictive control (MPC) effectively handles most of these difficulties, Mintek decided to develop, customise and test an MPC controller for milling control.Conventional MPC uses a dynamic model of the plant to calculate the future (predicted) plant outputs based on the movements of the manipulated variables in the past. The future moves of the manipulated variables can then be calculated using an optimisation algorithm that will minimise the difference between the setpoints and the predicted outputs. The least squares optimisation method is the quickest, but it does not cater for process constraints. Constraints can, however, be handled explicitly by using a quadratic programming optimisation method. This method ensures that the variables stay within the constraints at all times, but the calculation procedure is much more computationally intensive.• modifying MPC for the Minerals Processing IndustryIt was found that in order to obtain good, robust control using MPC in the minerals processing industry, the conventional MPC algorithm had to be adapted. The MPC controller was first tested on an extensive milling simulator, which was specifically developed for that purpose. Problem areas were identified and the MPC controller was specially adapted to accommodate for these difficulties. Some of the adaptations to the conventional MPC algorithm and special features added were as follows:Since milling plants contain many integrating processes (e.g. sumps), and integrating errors can cause offsets and poor control, it was essential that integrators be handled explicitly. This problem was solved by adapting the algorithm and also using an integrating error portion to compensate for integrating errors.Constraints on the inputs and outputs adversely affected the control. The controller using simple least squares optimisation was tuned with the constraints in place, while the maximum expected setpoint changes were made. This ensured that the process would be able to handle most setpoint changes up to these expected values without violating the constraints. A later improvement on this method was the incorporation of a quadratic programming algorithm to perform the optimisation. This method handles limits explicitly and was shown to perform better than the least squares methods (at the expense of a much higher computational cost)"

Jan 1, 2003