Search Documents

By Stefan Mutz, Thomas Pretz, Jörg Julius, Frank Hilbk-Kortenbruck, Aachen, Hartmut Harbeck

"The annual global production of primary aluminium was about 22 million tons in the year 2000. This represents the second largest production rate in the world market after steel. Experts are predicting that the global demand for aluminium will rise in the near future.In contrast to other metals such as copper, the recycling of aluminium is considered relatively difficult. There are two main alloying groups, namely cast and wrought. In the case of aluminium it is important that the constituents within the scrap aluminium to be smelted are clearly identified. Within the secondary remelting process alloying does take place, but a removal of unwanted metal constituents is problematic. Therefore mixed aluminium scrap (not genuine in-house collected material) can only be used to produce cast aluminium. Thus, in other words a down-cycling of wrought (extrusions and rolled) products occurs. However, in spite of these difficulties there is an increasing trend to produce aluminium from secondary scrap (approx. 35% in Germany). Similar trends have been identified in the USA and other industrialised countries. This development is mainly caused by the high energy consumption for the production of aluminium from bauxite. Producing aluminium from scrap is saving about 95 % of the energy in comparison to the processing of primary raw materials.An improvement in sorting technology is necessary to separate wrought and cast alloys of aluminium as the absolute amount of recovery from old scrap is higher than the demand for cast.This contribution presents some alternative methods, utilizing automatic picking technologies, which will be used to separate secondary aluminium in the future. During the recent years, some major progresses have been made in automatic picking technologies for different applications in the processing of primary and secondary materials.For the sorting of aluminium scrap a special additional sensor will be necessary. The best detection will ultimately be based on online elemental analysis of single particles. First investigations in the field of laser based systems for these applications have already been performed several years ago. One of the laser based techniques is the laserinduced breakdown spectrometry (LIBS)."

Jan 1, 2003

By Douglas Fuerstenau

"Comminution not only is the major energy cost in mineral processing but it also can limit the grade of ores that can economically be processed. Rittinger first postulated that the energy for comminution is proportional to the new surface produced. Because most of the strain energy needed to initiate and propagate cracks in particles cannot be recovered, researchers reported woefully low comminution efficiencies. Recognizing this, Schoenert proposed that the baseline for assessing comminution efficiency should be the mechanical energy needed to break single particles. After extensive single-particle studies, Schoenert turned to comminuting beds of particle in piston dies. Finding particle-bed comminution to be energy efficient, he had the brilliant conception of making the process continuous, namely his invention of the choke-fed high-pressure roll mill (HPGR). Practical comminution involves masses of particles, which can lead to energy efficiencies far less than that of single-particle breakage. A three-way classification of grinding modes in terms of the mobility of the particle mass provides the necessary insight into the low efficiency of different comminution machines: namely, single-particle breakage, confined particle-bed grinding (HPGR), and loose-bed grinding (ball mill). Examples are given for determining and comparing comminution efficiencies. The size distribution of comminuted products become self-similar, and a single parameter, the median size, can be used to quantify the extent of comminution. The reduction ratio, expressed in terms of the mean size of the feed and product, varies linearly with expended energy and the slope of such plots is a measure of the grindability of the mineral. This approach clearly gives the order of energy efficiency as single-particle > particle-bed > loose-bed comminution. HPGRs are more efficient at low reduction ratios but lose that advantage at higher reduction ratios. Extensive hybrid grinding experiments carried out with our instrumented ball mill and HPGR showed significant energy savings by first comminuting the material in the HPGR and followed by the ball mill. At Cerro Verde an industrial-scale hybrid HPGR/ball mill system is reported to have a 13 percent energy savings over a SAG mill/ball mill system."

Jan 1, 2014

By Anatoly N. Vinogradov, Vyacheslav A. Tsukerman

The North-West Russia for 70 years is one the main producers of Cu, Ni, Co, Al, P, REE, Zr, Ti, Ta, Nb, Fe ores and other minerals in the Europe. The largest mining companies locates in the Murmansk region, and they extracts about 100 mln. tons of ores and damps up to 250 mln. tons of waste annually. Nevertheless an environmental situation in the region is stable due to rational management and treatment of wastes on a base of the Noosphere Concept for Nature-using suggested by Vladimir Vernadsky. The Concept was performed practically in the Fersman’ scheme of “non-waste mining and metallurgical complex”, published in the monograph “Minerals of the Kola Peninsula: current status – analyses – prognoses” (1941) and served till now as a sort of “Bible” for industry managers lightening a strategy for an industrial “trophic chain”. The recent status of this scheme is shown on the picture below, and one can see that our concept is coincide with modern ideology known as the “Industrial Ecology”. Moving along this mainstream in the Kola Peninsula since 1929, when the first mining company “Apatit” was established, we provided the water close circuit in mineral dressing plants, hydrometallurgical treatment of slag and tailings, productions of building materials and adsorbents of wastes. The innovation set of technologies designed by the Kola Science Centre allow recently to involve tailing damps in re-exploitation as the technogenic deposits of valuable raw materials for fertilizers, tanning and cleaning agents, pigments, concretes and etc.

Jan 1, 2003

By Markus A. Reuter, Bernardete Castro, Antoinette van Schaik, Han Remmerswaal

"INTRODUCTIONThis paper describes the inseparable relation between product design, liberation, and metallurgical recovery in the optimisation of the recycling of passenger vehicles. This is illustrated by the use of an optimisation model that uses among others a material combination matrix based on metallurgical process data, which depicts the consequences of the quality of intermediate recycling products on the metallurgical material recovery and final alloy metal quality. Various simulations will illustrate the effect of changes in product design and hence liberation on the recycling of end-of-life vehicles. Although this paper focuses on cars, the discussion is equally valid for any end-of-life product. Calibration of the model will be discussed with reference to experiments carried out in order to study the optimal recovery of aluminium applications used in passenger vehicles.OPTIMISATION MODEL FOR LINKING DESIGN TO RECYCLINGImproving the efficiency of the material cycle of the car is essential to ensure a sustainable development of our society as well as e.g. to comply with the recycling targets imposed by EU legislation. In the design of the car various materials are combined and connected in many different ways, making up the complex 'mineralogy' of the car and its components, which is changing rapidly and continuously over time. The liberation of materials during shredding plays an important role in the composition (and quality) of the intermediate recycling streams. In order to include the influence of product design (mineralogy), liberation, and particle composition (quality of the intermediate recycling streams) in the optimisation of the recycling of end-of-life vehicles, a systems engineering tool for optimisation of the recycling of end-of-life vehicles was developed, including a liberation model. The optimisation model predicts the recovery rate as a function of product design (in this case a car), liberation, efficiency of the different process steps, economy and legislation based on a developed recycling flow sheet for end-of-life vehicles. Optimisation can only be achieved through the combination of the knowledge areas of product design, shredding, mechanical separation technology linked to pyro- and hydrometallurgy, composing a complex interconnected network of unit operations."

Jan 1, 2003

By Jan Cilliers

The demand for minerals has never been higher. Moreover, the rate at which this demand is growing is unprecedented, and expansions and new mines will be required to fulfill these growing needs. This, in turn requires a skilled engineering workforce that can develop and operate minerals extraction processes to utilise effectively more complex ores. The International Mineral Processing Council (IMPC) has established a Commission on Mineral Processing Education to consider the global challenge of ensuring an adequate supply of well-trained mineral processing engineers. The Commission has, for the past four years, been collecting international information of the recent mineral process engineering supply position and with a view into the near future. The data collected by the international Commission members are as follows: ? Undergraduate student numbers in minerals engineering ? A breakdown of their male/female ratio ? The fraction of the students from each course that enter the minerals industry ? Teaching staff numbers and age profile This data has been collated and analysed, and the findings are presented. It is clear from the analysis that the supply situation is very different in different regions of the world, and that great opportunities exist for targeted recruitment into the industry to meet the future talent demand. Further, an attempt at correlate the supply of minerals engineers based on the local mineral production rate is shown, which shows that there are clear outliers.

Sep 1, 2012

By Katsunori Okaya, Toshio Inoue

"Movement of mill charge in the centrifugal mill is essentially dissimilar from that in the vibration mill with circular path despite the apparently similar mode in the motion of mill body. This dissimilarity is the reason for the substantial advantage regarding the comminution intensity of the centrifugal mill over the conventional vibration mill accounting approximately to one order of magnitude.Bradley et al. (1983) developed a math model regarding the mill charge as a lump. This model is presented in a simple equation with which motion of the mill charge can be predicted in terms of the parameters describing the specification and the operation of the centrifugal mill. This equation can be used to explain the power draft of the mill as well as the stability of mill operation. More recently, on the other hand, a numerical simulation technology called discrete element method (DEM) has become a vital vehicle for estimating the motion of the grinding media in the mill. The authors developed a mill simulator called ‘Mill Solver’. Mill Solver is designed to simulate various types of mills including centrifugal and vibration mills. The principle of this simulator is to calculate individual kinetic events of interaction of balls in the mill in a minute time interval, hence the mill performance can be constructed for estimation."

Jan 1, 2003

By Tatyana A. Ivanova, Tamara N. Matveyeva, Anatoliy A. Fedorov, Tatyana V. Nedosekina, Valentin A. Chanturiya

"As native gold resources are depleted, gold-sulfide ores consisting essentially of pyrite, arsenopyrite, chalcopyrite and pyrrhotite become one of the major precious metal sources, being presently estimated to account for more than 40 % of total world gold reserves.The prime objective of flotation of gold-bearing sulfide minerals is to increase precious metal recovery and to reduce the product cost through achieving maximum selectivity of the separation process. Separation of Fe and As sulfides in selective flotation of gold-bearing ore products helps reduce the amount of ore mass being sent to hydrometallurgical processing, and permits of recovering gold from As-free gold-containing pyrite concentrate by using a lower-cost pyrometallurgical procedure.An ingenious methodology purposively developed by the authors for comprehensive assessment of flotation properties, electrophysical characteristics and crystal chemistry features of natural gold-bearing sulfide minerals of different origin has been used in this study.Comparison of the elemental composition, non-stoichiometry and flotation activity of different monomineralic pyrite varieties revealed that samples with negligible impurity content are nearly stoichiometric, and show similar flotation properties. Samples with high Cu, As and Au contents are highly non-stoichiometric and readily floatable, even in a highly alkaline environment. Samples with high surface sulfur content and the most defective crystal lattices, with iron atoms substituting for sulfide vacancies, are most sensitive to depressive action of the alkali.A correlation between the type of conductivity, impurity content and amount of xanthate compounds on the surface and in the liquid phase of monomineralic pyrite and arsenopyrite suspensions detected by the extractive spectrophotometric analysis has been identified and analyzed. Experiments showed that the proportion of cuprous xanthate in sorptive collector layers on sample surfaces increases with increasing Cu content of the samples. In general, pH for pH, overall xanthate sorption on pyrite is greater than on arsenopyrite, with the difference increasing in passing from an acid environment to an alkaline one. A significant decrease in xanthate sorption on arsenopyrite is observed at pH greater than 9, whereas sorption on pyrite markedly decreases at pH greater than 11."

Jan 1, 2003

By M. Brennan, A. Majumder

"The mineral processing research community has devoted enormous efforts to the modelling of mineral separators over the last 50 years. While many efforts have been made to use only fundamental physics as a basis for the models, fluid flow patterns in separators are not simple while the motion of particle swarms through fluids is extremely complex.Sluices are the simplest forms of the flowing film concentrators and have been used in the mineral industry since at least the 15th century (Agricola, 1556). Flowing film concentrators, in general, still have widespread applications in the mineral industry and their designs vary according to the specific need.Particle separation in a sluice appears deceptively simple, but the separation mechanism has proven complex and difficult to quantify. Separation in the sluice takes place in a flowing film a few millimetres thick. The particles are suspended by turbulent eddies and motion relative to the fluid takes place in socalled hindered settling mode. The difficulty arises because the presence of the particles influences the flow pattern and when the particles are fully transported by the flow, the flowing slurry may not have the same viscosity as water and does not have the same density. For this reason a universal theory which can explain the behaviour of a suspension in a sluice or any other gravity equipment has still not been developed. Few attempts have been made to analyse the mechanisms involved in the concentrating processes of pinched sluices and Reichert cones.Fluid flow behaviour and slurry transportation characteristics in open channels have, however, been active research topics in the disciplines of fluid mechanics, chemical engineering and civil engineering for many years. As a result, an extensive literature is available, which has led to the development of the model described in this paper.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 beads at various particle concentrations and flow rates, and the actual particle separation taking place compared to predictions from modelling.The modelling strategy was divided into two parts; the first was to predict the horizontal slurry flow velocity profile and the second to predict the solid particle concentration over the depth of the flow at different flow conditions. The predictions were then validated with measured data sets. This can easily be done if the slurry flow is split at different fractional flow depths so that the mass flow of solids over different flow heights is measured accurately over a fixed period. Mass flow splits can then be calculated by integrating the product of the predicted slurry velocity profile and the predicted solids concentration profiles from the base of the sluice bed to the slurry free surface."

Jan 1, 2003

By Yimin Zhu, Dezhou Wei, Haijun Liang, Zhifu Zhou

Pyrite is one of sulfide minerals most widely distributed in the earth. During the flotating separation of metal sulfide and the floating desulfation of coal, it is often requaired to restrain pyrite. However, the result is not very satisfactory because of the little difference of floatability of minerals to be separated. The problem is extremely outstanding for the floating desulfation of coal. Researchers carried out a number of experiments with T. ferrooxidans (Yang et al, 2001; Zhou et al, 2000; Zhou et al, 2001; Wei et al, 2001) for an enhanced restraint to pyrite. The results showed that the T. ferrooxidans after being cultured works the best in restraining pyrite. After 10 min pretreatment, the removal rate of pyrite from coal is up to 80%~90%. The research results by Yuan Xin et al (Yuan et al, 2000) show that T. ferrooxidans can restrain effectively the floatability of pyrite and can reduce the recovery of pyrite from 95% to 8%. Though the restraining properties of T. ferrooxidans to pyrite have been proved by many tests and experiments, there are very limited reports on its restraining mechanism. Thus, a systematic experimental research is carried out in this paper.

Jan 1, 2003

By Graeme Jameson

"In current flotation circuits it is customary to grind the whole of the mill feed to an initial size, which is determined by the liberation characteristics of the ore, and the ability of the flotation machines to float the fully liberated particles. Usually, the product of the primary mill will be ground further in a ball mill, to improve the liberation of the values and hence the product grade, before passing to rougher flotation cells. The final grind size is ore-dependent, and is a balance of factors including the relation between surface liberation and grind size, the grain size of the valuable particles within the host rock, and the feed grade. The P80 for the feed to flotation may be as high as 300 µm for free-milling copper ores, and as low as 53 µm for finely disseminated complex ores. The grinding of the feed ore requires considerable input of energy. The rate of replacement of wear components such as mill liners and media, is proportional to the energy input. The costs of these two factors - energy and wear - are usually comparable and together they form the major part of the running costs of a concentrator, and indeed, of the whole mining operation. The energy required increases as the final grind size decreases. This talk will focus on the reduction of milling costs that could be achieved, using a fluidized bed flotation technology that can give high recoveries at coarse particle sizes. A number of different circuit designs will be presented, using the new technology. Calculations show that reductions in the running costs of a mine and associated concentrator of the order of ten to twenty percent could be achieved."

Jan 1, 2014

By Khoi Nguyen, Keeney. Luke

An outcome of geometallurgical mapping and modeling is the generation of continuous down hole profiles of quantitative geometallurgical attributes at assay scale. These attributes vary from traditional routine data collection methods such as assays and geotechnical logging (e.g. RQD, Fracture Frequency), to new measurements including petrophysical attributes, EQUOtip hardness data and modelled estimates of metallurgical performance indices (e.g. A*b, BMWi). Highly variable continuous geometallurgical data with high spatial resolution can make it difficult to identify spatially continuous domains. An automated method based on the well known time series analysis technique of cumulative summation (CuSum) has been developed which uses statistical analysis to identify domain boundaries in down hole profiles. A bootstrap analysis automatically identifies potential domain boundary locations with associated confidence levels and a series of t-tests determine the significance of defined populations. A hierarchical clustering algorithm is applied to results to enable data sensitivity to be taken into account. The method produces statistically significant populations and is suitable for application on either single or multiple drill hole datasets. Domains defined preserve the inherent variability of the continuous down hole data and provide a method for scaling up variability data (i.e. assay interval scale) to mine scale (i.e. bench scale, block model). A major advantage of the method is its ability to provide suitable input parameters for use in spatial modeling of non-additive and non-linear geometallurgical attributes through conditional simulation techniques.

Jan 1, 2014

By C. Suazo, N. Mora

"The bubble size distribution plays a fundamental role in flotation kinetics. Nowadays it is possible to measure the bubble size distribution, gas holdup and superficial gas velocity by using specialized equipment developed for this purpose. In this article a methodology is presented for estimating the kinetic rate constant using a known particle and bubble size distributions. The attachment and detachment probabilities were modelled from three matrixes. The first matrix represents particle size distribution, the second was named the probability matrix, and the third is the bubble size distribution. Using laboratory data, it has been possible to estimate the flotation rate constant as a function of operating conditions and both the bubble and particle size distributions. The model tested at laboratory scale achieved a quadratic correlation coefficient equal to 0.97. This methodology could be used in the operation and control of the flotation, when in the near future we have industrial equipment that allows in-line measurement, in real time, of gas dispersion and bubble size distribution. knowing particle size distribution, it will be possible to manipulate the operating flotation cell conditions, in order to modify the bubble size distribution and find the bubble distribution that maximizes flotation kinetics. Furthermore, knowing the bubble size distribution, the grinding operator will be able to look for the particle size distribution that produces the maximum flotation kinetics. In order to do this, the operator will have to modify the solids percentage fed to the hydrocyclone cluster, modify the apex/vortex ratio, etc. Finally, this article shows that the maximum flotation speed is reached when the particle size distibution slope is equal to the bubble size distribution slope."

Jan 1, 2014

By Andrei Kolesnikov

"The results of CFD modelling of complex heat transfer and fluid flow in DC electric arc furnace are presented. The model includes energy and momentum conservation equations for gas phase (freeboard), liquid phase (molten bath) and solid phase (walls and graphite electrode). Simulation revealed the temperature distribution in all elements of the furnace. The influence of the arc and melt resistance on the temperature distribution is discussed.Computational Fluid Dynamics (CFD) becomes a powerful tool for better understanding of the heta and mass transfer processes in industrial equipment. In current work commercial CFD software STAR-CD was used to analyse complex heat transfer in DC EAF, including forced and natural convection, conduction and radiation. The finite volume solver using body-fitted grid was used. The grids are non-staggered hexahedral grids and all variables are evaluated at cell centers. Steady-state calculations were performed with PISO algorithm. For the convective terms in equations the hybrid differencing scheme was used. An upwind differencing scheme was used for other terms. Conjugative heat transfer was taken into account in heat transfer calculations on gas-liquid and gas-solid interfaces. Radiation from the arc and inner surfaces of the furnace was calculated by a “discrete beam” approach. Source term in energy conservation equation was used to represent heat release in the zone occupied by the DC electric arc. The volume of this zone was calculated from semi-empirical equations. The Joule heat and heat due to chemical reaction in the melt were calculated through the source term expression in the energy balance equations for the slag and metal in the molten bath. The gas acceleration in the electric arc zone and stirring of the slag due to gas bubbles motion were accounted for by momentum source terms in the momentum conservation equations.All simulations were carried out using axisymmetric two-dimensional grid. The grid resolution was 20 cm per cell in most part of the furnace, except the arc region, where resolution was 5 mm per cell."

Jan 1, 2003

By N. Panteleeva, E. Konoplev, I. Michailova, S. Titkov, T. Gurkova, T. Chumakova, Sylvinit Co.

"INTRODUCTIONFlotation of mineral particles by cationic surfactants – namely alkyl amines and their derivatives – is finding ever growing application in processing of various ores and in quality refinement of the benefication end products, iron concentrates for example. In the recent years application of cationic collectors - alkylmorpholines - enabled the development of new highly efficient procedure of water-soluble mineral processing, being also the one with low energy consumption. This procedure is a reverse flotation of carnallite from bay salts of salt-pans and from carnallite ores.The efficiency of cationic SAS influence depends upon colloid-chemical state of the collector in flotation liquid phase, especially if the latter comprises mineral salts. Availability of clayey impurities or slime particles of iron oxide in cation flotation feed, which actively sorb the cationic collector, deteriorate considerably the cation flotation performances. In Halurgy Institute the investigations of cation flotation under potash and carnallite ore processing have been carried out, and new reagent modes, enhancing cationic surfactants sorption at mineral particles, have been developed. Application of new reagents enables one to decrease considerably the consumption of expensive cationic collectors and to enhance the floatability of coarse mineral particles."

Jan 1, 2003

By Thomas Pretz, Alexandra Pehlken

Sampling of municipal solid waste is known to be difficult to sample if it has to be representative. When dealing with municipal solid waste, the usual method is to study separately waste material below 40 mm and particles bigger than 40 mm. The material < 40 mm cannot be sorted manual and is termed the “fine fraction”. The particles > 40 mm can be separated manually and are usually sorted into pre-defined groups of materials. These groups can be defined depending on the demand of quality assurance. But standardisation of procedures is not far progressed, even for the widely practised sorting methods.

Jan 1, 2003

By P. K. N. Raghavan

Bauxite is the principal source for the production of aluminum metal which has gained tremendous importance over the years due to its versatility as a metal capable of substituting almost all of the non-ferrous metals, non metals like wood, plastics, fibre glass, etc. and even steel in many applications. 90% of the high purity alumina needed for the production of this wonder metal in the world comes from bauxite. Even though alternate methods have been developed to produce alumina from non bauxite sources, owing to the high capital and operating costs involved, Bayer process still retains the superiority among all the methods by virtue of its simplicity and favourable economics. Bauxite continues to be the perfect ore for alumina production and its remarkable utilization makes it the most important non-ferrous ore. Most of the Alumina Refineries in the world process different types of bauxite and that too from varied sources. The technology employed in the Alumina Refineries depends on the Bauxite processed, with an ultimate view to produce alumina of required quality in the most efficient manner. Bureau of Indian Standards (BIS) and Indian Bureau of Mines (IBM) had from time to time put forward bauxite specifications for metallurgical and other applications. However, these are found no fully based on alumina technology, present developments and availability of different types of bauxite in the country. The aim of this paper is to provide scheme of metallurgical grading of Indian Bauxite based on the background of alumina processing technologies and availability of various types of bauxite in the country. Efforts are made here to evaluate beneficiation techniques for central Indian Bauxite from the point of view of alumina production. Keywords: bauxite beneficiation, digestion, alumina, gibbsite, bohemite, diaspore

Sep 1, 2012

By Heribert F. Broicher, Arne Volland

"The analysis of laser-induced fluorescence (LIF) combines remote sensing techniques with laser spectroscopy. Over variable distances different rock types are identified or changes in the quality of one rock type are recognised. LIF Analysers operate non-destructive, non-contact, on-line and in real-time. Installed over conveyors or over driveways the analysers ""look"" at the moving bulk solids 20 times per second and provide in real-time information for downstream sorting, blending or optimisation of processing. LIF Analysers recognise rock types (hardness information or dilution control), degrees of alteration, degrees of oxidation and –semi-quantitative- content of certain minerals in an ore. Only under favourable conditions quantitative assay information can be obtained, as e.g. the phosphorus content of Kiruna type iron ore.The main application of LIF Analysers is seen in big open pit operations, where the muck is trucked to an in-pit crusher and the crushed rock conveyed out of the pit to its respective destination: e.g. the primary high grade ore to the mill, the low grade or oxidised ore to the leach pads and the internal waste to the dumps.In underground mining and especially in block caving operations dilution control is of utmost importance. Here a LIF Analyser mounted on the back of a pick-up truck can be moved from drawpoint to drawpoint and quickly assess the degree of dilution.Closer to the mill the stockpile management based on LIF analysis will lead to stockpiles sorted according to rock type (hardness), degrees of alteration and oxidation. In case of direct feeding rom ore to the mill the knowledge of the type and quality of all the material between the in-pit crusher and the mill facilitates optimisation of e.g. reagent dosage."

Jan 1, 2003

By M. H. Moys, M. M. Bwalya

The use of the Discrete Element Method (DEM) to simulate the interaction of particles in dynamic environments is growing rapidly. This technique is also being applied in the study of milling extensively. However, to date more focus has been on the validation of the model. In spite of the advantages of the detailed information that is available from the simulator, appropriate application of this information remains a daunting task. In this paper a technique that combines particle breakage probability data with the energy spectra generated from the DEM is discussed.

Jan 1, 2003

By R. Peter King, Claudio L. Schneider, Reiner Neumann

Predicting the liberation spectra that is produced from breakage of two-phase ores under the random fracture assumption can be accomplished by measuring the conditional, on length, linear grade distribution directly from images obtained from a representative sample of the unbroken texture. The technique tremendously simplifies the prediction problem by avoiding the multiple convolutions that arise from the traditional method based on a description of texture from a measurement of the chord length distributions associated with each phase in the unbroken texture. For a 3-phase system, the phase transition probabilities associated with the textural information make the traditional calculation an almost insurmountable mathematical problem. The direct measurement procedure on the other hand can be expanded to three phases comparatively easily.

Jan 1, 2003

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