Search Documents

By M. Kargar Razi

According to the unique properties and many applications of nano gold powder, it was prepared from an acid leaching tail solution. The low value residual gold ions were existed in acid leaching tail solution in Mouteh gold mine in Iran with fellow heavy metals as pollutants (FeP2+P, Cu P2+P, and ZnP2+P). Preparation of nano gold particles is achieved by two steps. First, by addition of ammonia as separation step and second using suitable reagents for precipitation/extraction step. Conditions control includes pH, residence time, temperature, agitation, and molar ratio of reducing agents to reagents. The optimum results show that nano gold powder was prepared by temperature of 250 °C, urea to gold molar ratio of 3:1, and reduction time of 20 min. Also, heat treatment improves dispersion of nano sized gold particles. The properties of gold nano powders were evaluated by SEM, A.A, and UV- Vis spectroscopy methods. The size of prepared particles were about 80nm to 90nm.

Jan 1, 2014

By S. M. Saghaeian

The studies on the product finer than 12 mm due to crushing and sizing plant of Gol-e-Gohar Iron Ore mine showed which this material contains significant amount of valuable iron minerals. Based on the carried out experimental studies on representative samples of fine hematite iron ore, a processing flow sheet was designed to upgrade iron content. This plant is known as first hematite iron ore enrichment plant in Iran which the dry medium intensity magnetic separators (with magnetic field intensity 4500 gauss and high gradient) has been used. The main objective of the plant is production of high iron product with iron content more than 61% and weight recovery more than 55% from the feed with iron content of 54%. In this plant, at first material are screened by a single deck screen (6mm), then oversize particles is separated at one stage by magnetic separator, while undersize ones at two stages. Based on the results of experimental tests, a final high iron product with iron content of 62.4% and weight recovery 63.9% can be achieved. It should mention which the iron content can be enhanced through increasing the speed of magnetic separators especially at the part of separation of finer iron ore.

Jan 1, 2014

By Chen Jianhua, Shen Zhenchang, Zhang Ming, Chen Feifei

"With flotation cells scale-up, the performance of flotation cells is increasingly variated for short-coming and dead zones existing. The residence time distribution which shows the movement of tracers can assess the unexpected flows such as short circuit and dead zone in a flotation cell. In this paper, the residence time distribution of pilot flotation cell is investigated with CFD simulation. Based on the theoretical model, the short-coming probability and dead zone probability are quantitative. The ideal average residence time is similar with the actual average residence time. The mixing effect of the flotation cell approaches the ideal continuous stirred tank reactor. The operating parameters such as speed and feed velocity have complicated influence on residence time distribution. The research would be helpful to understanding flow in flotation cell and improving performances of flotation cell.INTRODUCTIONFroth flotation is a method which purifies minerals according to difference of physical and chemical surface characters of different mineral particles (Shen, Z., Chen J, 2012). Flotation cell is a key equipment for froth flotation method. The flotation cell performances have great influences on flotation index.It is difficult to evaluate flotation cell how to improve flotation procedure to complexity.The main flotation cell performances are aeration volume, air dispersion, solid suspension and residence time distribution, etc (Shen, Z., Lu, S., 2008). Residence time distribution is an important fluid hydrodynamics which could reveal the flow of pulp, mixing of particles and air bubbles. In the flotation cell, the ideal average residence time of pulp could obtain easily. But the fluid differential units would be different flow paths. The residence time of the fluid differential units is variation and would be longer or less than the average residence time. The residence time distribution would reveal the mixing effect and predict the not ideal flow in the flotation cell such as shorting cutting, dead zone, etc. As we know, the shorting cutting and dead zone would be decreased the real flotation time and it is disadvantage to better flotation effect, especially for large flotation cell. It is impossible to eliminate the not ideal flow totally. It is very important to reveal and decrease the shorting cutting and dead zone probability in the flotation cell."

Jan 1, 2018

By Y. M. Zhu, G. C. Gong

"The influences of citric acid (CA) on the flotation behavior of cassiterite using styrene phosphonic acid (SPA) as the collector were studied via single mineral flotation tests, contact angle measurements, zeta potential determinations and solution chemistry calculations in this paper. The results of single mineral flotation tests revealed under the conditions that pH=4 and the dosage of SPA 100 mg/L, the recovery of cassiterite reached 76.78%. The addition of 20 mg/L AC decreased the recovery by 12.42% at pH=4. Contact angle measurements demonstrated the adsorption of CA prior to SPA made the cassiterite more hydrophilic. The results of zeta potential determinations and solution chemistry calculations indicated the chemisorption of C6H7O7- and C6H6O72- filled parts of the Sn site on cassiterite surface and prevented the SPA from getting adsorbed. The hydrophilic effect of AC was stronger than the hydrophobic effect of SPA and the flotation of cassiterite was crippled. INTRODUCTION Tin is an important metal which is widely used in medical instruments, marine engineering materials, electronic and electrical devices, welding, weapon manufacturing and organotin compounds industries (Angadi et al., 2015). Cassiterite is the only mineral from which the tin can be extracted, thus it is of great importance to efficiently exploit the cassiterite mineral. Flotation is an indispensable method for the recovery of cassiterite (Li et al., 2015). However, cassiterite is usually in association with many other gangue minerals which are of similar floatability, such as quartz, calcite, fluorite, iron oxide, wolframite, scheelite and tourmaline (Sreevinas, 1997). So some modifiers are necessary in the cassiterite flotation process. The amounts of the modifiers used in the flotation must be judicious and the overmuch modifiers may cripple the recovery of cassiterite (Parsonage et al., 1987). Nevertheless, the research about the effects of the modifiers on the flotation behavior of cassiterite has been rarely reported. In this paper, the influences of citric acid (CA) on the flotation of cassiterite using styrene phosphonic acid (SPA) as the collector were investigated by flotation tests, contact angle measurements, zeta potential determinations and solution chemistry calculations."

Jan 1, 2018

By Chen Jianhua, Shen Zhenchang, Zhang Ming, Chen Feifei

"With flotation cells scale-up, the performance of flotation cells is increasingly variated for short-coming and dead zones existing. The residence time distribution which shows the movement of tracers can assess the unexpected flows such as short circuit and dead zone in a flotation cell. In this paper, the residence time distribution of pilot flotation cell is investigated with CFD simulation. Based on the theoretical model, the short-coming probability and dead zone probability are quantitative. The ideal average residence time is similar with the actual average residence time. The mixing effect of the flotation cell approaches the ideal continuous stirred tank reactor. The operating parameters such as speed and feed velocity have complicated influence on residence time distribution. The research would be helpful to understanding flow in flotation cell and improving performances of flotation cell.INTRODUCTION Froth flotation is a method which purifies minerals according to difference of physical and chemical surface characters of different mineral particles (Shen, Z., Chen J, 2012). Flotation cell is a key equipment for froth flotation method. The flotation cell performances have great influences on flotation index.It is difficult to evaluate flotation cell how to improve flotation procedure to complexity. The main flotation cell performances are aeration volume, air dispersion, solid suspension and residence time distribution, etc (Shen, Z., Lu, S., 2008). Residence time distribution is an important fluid hydrodynamics which could reveal the flow of pulp, mixing of particles and air bubbles. In the flotation cell, the ideal average residence time of pulp could obtain easily. But the fluid differential units would be different flow paths. The residence time of the fluid differential units is variation and would be longer or less than the average residence time. The residence time distribution would reveal the mixing effect and predict the not ideal flow in the flotation cell such as shorting cutting, dead zone, etc. As we know, the shorting cutting and dead zone would be decreased the real flotation time and it is disadvantage to better flotation effect, especially for large flotation cell. It is impossible to eliminate the not ideal flow totally. It is very important to reveal and decrease the shorting cutting and dead zone probability in the flotation cell."

Jan 1, 2018

By C. Y. Wang

Based on Slurry Electrolysis, and combined with the problems of extraction antimony, lead and their separation from complicated antimony-lead concentrate, a series of tests have been done in this paper. Using Slurry Electrolysis to treat complicated antimony-lead concentrate, antimony and lead can be separated in one step. And 2# metal antimony can be achieved in cathode directly at the same time. Lead in concentrate is restrained still in the residue at the form of PbCl2. After converting the residue with ammonium carbonate and removing sulfur with kerosene, the high quality of lead concentrate (Pb is above 45%, S is below 15%) and pure sulfur can be achieved, respectively. The recovery of all in Slurry Electrolysis process is as: Sb 95%, Pb 95%, Ag 80%, S 60%.

Jan 1, 2014

By R. J. Batterham

The science and practice of mineral processing has been driven for many years by the same set of challenges that shape the mining industry ? demand and supply, grades and their decline with time, mineralogy and the need to handle more complex ores, disposal and minimising of wastes and the availability of water, power, skilled labour and the like together with increasing expectations from the community, concerns on sustainability and safety and, as well, climate change forcing a more targeted approach on energy. In this paper, the author builds on previous reviews to the IMPC and suggests that while step changes are long overdue, the unprecedented demand mitigates against drastic changes. That said, the ?Mine of the Future? is already requiring some different approaches.

Jan 1, 2014

The status quo of ecological rehabilitation in metal mine in China and progress in the technology for ecological rehabilitation are presented, including soil reconstruction, soil improvement, plant selection. The present and future research fields are suggested: basic theory of ecological rehabilitation, operative technology of ecological rehabilitation, new technology application research and ecological rehabilitation plan, management and policy.

Jan 1, 2014

By James D. Bittner

Separation Technologies, LLC (ST) has developed a processing system based on triboelectric charging and electrostatic separation that provides the mineral processing industry a means to beneficiate fine materials with an entirely dry technology. The environmentally friendly process can eliminate wet processing and required drying of the final material. The process requires little if any pre-treatment of the material other than grinding and operates at high capacity ? up to 40 tonnes per hour by a compact machine. Energy consumption is low, approximately 1 kWh/tonnes of material processed. Since the only potential emission of the process is dust, permitting is typically relatively easy. In contrast to the other available electrostatic separation processes that are typically limited to particles greater than 75 µm in size, the ST belt separator is ideally suited for separation of very fine (<1 µm) to moderately coarse (500 µm) materials with very high throughputs. The triboelectric particle charging is effective for a wide range of materials and only requires particle ? particle contact. The small gap, high electric field, counter current flow, vigorous particle-particle agitation and self-cleaning action of the belt on the electrodes are the critical features of the ST separator. The high efficiency multi-stage separation through charging / recharging and internal recycle results in far superior separations and is effective on fine materials that cannot be separated at all by the conventional electrostatic techniques. Since 1995, this triboelectric process has been extensively used for the beneficiation of coal fly ash with eighteen separators in place and over 100 machine-years of operation at locations in the North America and Europe. The technology has been also successfully applied to the beneficiation of a variety of minerals including calcium carbonates, talc, and potash.

Sep 1, 2012

Certainly, one of the core processes for the recovery of fine magnetic susceptible iron ore particles, mainly hematite iron ores, is the wet high intensity magnetic separation or WHIMS. This paper outlines a variety of recent developments and innovative process designs in the JONES® type WHIMS. It highlights the latest examples of flow sheets in terms of the diversity of combined rougher, scavenger, cleaner stages and adds to this the new more advanced JONES® type WHIMS equipment technologies like the HIGH FLOW ? HF, the QUATTRO ? Q and the SPLITT OPERATION ? SP. As a result of the increase of the capacity and magnetic intensity, the specific production cost could be decreased. The high worldwide acceptance for this continuous development testifies for the success and state of the art technology.

Sep 1, 2012

By A. N. Zagorodnyaya, T. N. Bukurov, Z. S. Abisheva, A. S. Sharipova

"This work presents the results of the development of a flowchart for the processing of copper production dusts resulting in the production of of lead sulphate, tribasic lead sulfate (TBLS), ammonium perrhenate, zinc, copper-cadmium cake.Dust having the following composition (wt. %): Pb 39.3, Zn 6.5, Cu 5.5, Stotal 10.8, As 1.05, SiO2 0.6, Cd 0.8, Bi 0.04, Ag 0.02, Re 0.03 was used in the work. Chemical and X-ray analyses indicated that lead, zinc and copper are generally present in dusts in the form of sulfates, various amounts of oxides, sulfides, silicates; sulfur (according to chemical analysis data) is present in the form of sulfates (~ 90%) and in the form of sulfides (~ 10%).Studies have been made on two-stage stripping of copper production dusts containing lead, zinc, copper, cadmium, rhenium in the form of various compounds, at first with soda solution, then with nitric acid solution. It has been indicated that the combination of the operations allows the recovery of 98% rhenium, 82% lead, 92% copper, and 96% zinc and cadmium each in the solution. Furthermore, at the stage of dust leaching with soda, 80% of rhenium and 0.01% of nonferrous metals each are recovered in the solution. Remaining quantities of all metals are recovered in nitric acid solution up to maximum values. Such behaviour of metals is caused by metal phase content in dusts and properties being formed during leaching of compounds. At the stage of dust leaching with soda solutions, carbonates with various compositions are formed due to the exchange reaction between nonferrous metal sulfates and sodium carbonate. The conversion of PbSO4 into PbCO3 is apparently occurs in the solid phase and the conversion of readily watersoluble zinc, copper, and cadmium sulfates into that of carbonate salts occurs in the liquid phase. Rhenium passes into solution due to the presence of rhenium greater quantity in dust in the form of readily soluble perrhenate. Some portion of the perrhenate is, probable, blocked by lead sulfate and liberated in the conversion of PbSO4 into PbCO3."

Jan 1, 2003

By Jia Muxin

The adsorption behaviors of metal ions on a silicate mineral and the flotability of a metalized silicate mineral with sodium oleate as collector are closely related to the structure of the silicate mineral. The adsorption isothermals of the metal ions on different silicate minerals are studied, the parameters of the isothermal give the information how many surface oxygen anions a metal ions is attached to, from this information and the structure of the mineral the site where a metal ion is attached can be guessed (The adsorption isothermals and adsorption behaviors with pH are fitted with Temkin, Frenudlich and Kurbatov equations respectively). Metal ions as Cu2+, Ni2+, Zn2+ are attached to the oxygen anions of surface species –Si-O– on the quartz surface, most of them coordinate to only one surface oxygen, their adsorption ability on quartz is low, but this metal ions on quartz surface protrude out of the surface perfectly, so their ability to combine with oleate acid radical is excellent. Quartz can be activated well by such metal ions. The isothermal of Fe3+ on quartz indicates these cations coordinates with more than two surface oxygen anions. This means the Fe3+ nucleates on quartz surface in the form of Fe(OH)3 colloid particles. The isothermals of Cu2+, Ni2+, Zn2+ Fe3+ on spodumene shows such ions coordinate with more than two surface oxygen anions, it indicate these ions may attach to the position where Li+ is formerly exists, such metal ions stay in a cave on the mineral surface, their adsorption abilities on spodumene are quite well, but they are affect by the oxygen anions of the two Si tetrahedron chains, the can almost not activate this mineral. These cations may attach on the site where Al3+ exists on the muscovite surface, just like spodumene, muscovite can not be activated well by such cations. Pt2+ has a large ion radius, as Pauling’s rules depicts polyhedron can not attach to Si tetrahedron well, the adsorption ability of Pt2+ on quartz is low and quartz is activated badly by Pt2+. On the other Pt2+ can adsorb well on spodumene and muscovite well and activated them well because it is not affected so bad by the tetrahedron oxygen anions of the minerals for their its large radius. The isothermal of Ca2+ on quartz indicates a Ca2+ coordinates with less than one surface oxygen anion, this means Ca2+ attaches on quartz mainly by physical adsorption. The adsorption abilities on mineral surface and the interaction abilities with oleate acid radical of the transition metal ions with the same valence are guessed to be determined by their d electron number, the rules should be d1<d2<d3<d4>d5<d6<d7<d8<d9>d10. Hydrogen bond may play an important rule in the activation of the mineral by HF and the interaction of the HF activated silicate mineral with collectors. The general rule is high bond valence and small metal ion radius surface species –Me-O can be activated well by HF with sodium oleate as collector. For example – SiIV-OH is neutral while –SiIV-FH+ and –SiVI-FH+0.33 are positively charged, the positively charged surface species are more likely to connect oleate acid radical via hydrogen bonds.

Jan 1, 2003

By P. T. Nguyen

The analysis of particle capture by an air bubble by two separate phenomena of collision and attachment has been difficult due to the uncertainty in determining when and where the two phenomena begin and terminate at the bubble surface, respectively. Therefore, a unified analysis was developed by solving the motion equation for the particle capture as one process. No distinction be-tween collision and attachment was made. Both the liquid film described by the micro-hydrodynamic resistance functions and the surface forces were included in the new model. The model predictions with the DLVO (Derjaguin-Landau-Verwey-Overbeek) surface forces did not agree with the experimental trajectories of latex and silica spheres. The model predictions with the inclusion of the non-DLVO attractive surface force agreed with the experimental data. The non-DLVO force reduced the repulsion between the non-attaching particle and the bubble, and significantly increased the attraction between the attaching particle and the bubble. The unified analysis is capable of predicting particle capture without the use of the induction time. The unified analysis also shows significant influence of the short-range hydrodynamic interactions on the collision which has been previously ignored.

Jan 1, 2014

By J. J. Lui

Surfactant plays a critical role in many industrial processes such as petroleum, mining, environment and food processing, and daily life. For ionic surfactant systems, adsorption of surfactant at oil-water interface imposes a net charge to the interface. The nature and magnitude of the charges at the oil-water interface depend on charge characteristics of surfactant molecules, concentration of surfactant and solution pH. In our previous study, zeta potential measurement of hexadecane drop-lets in industrial process water was shown to be a simple, yet powerful method for identifying various types of ionic surfactants present in the aqueous system. In this paper, the site-binding model was applied to prediction of zeta potential of oil droplets in aqueous solutions containing a mixture of cationic and anionic surfactant molecules. The total number of interfacial sites for various types of surfactant involved was determined by applying the Gibbs adsorption equation to the measured interfacial tension-surfactant concentration profiles. The distribution of various types of surfactant molecules at the interfaces was evaluated by Langmuir isotherm constants. With the ionization constant of surfactant taken from the literature, the predicted zeta potentials of oil droplets in surfactant solutions from site binding model were shown in a reasonable agreement with the measured zeta potential values.

Jan 1, 2014

By Zhou Junwu, Han Long, Song Zhenguo, Sun Chuanyao

The deposit genesis, the ore property, and the mineral characteristics, which has a close relationship with the beneficiablity, could be regarded as the “Genetic Properties”, actually are the determining factors for beneficiation process development. However, in conventional concentrator’s process development and engineering, there have been lack of systematic research and test on those genetic properties, also lack of well-established collecting, statistics and analysis system for those huge amount of data such as the historical mineral processing technique research, process mineralogy study, practical operating and engineering data. Obviously it is far less than enough for profound integration combining modern IT together with traditional mineral processing technology, as a result there are a number of disadvantages in mineral processing R&D and engineering, such as long term, high cost, low efficiency as well as repetitive testwork not making full advantage of previous experiences. Aiming to make utmost of those Genetic Properties and overcome above disadvantages in mineral processing R&D and engineering, the concept of “Genetic Mineral Processing Engineering (GMPE)” is put forwarded, and the strategical study has been conducted on this regard, with the main focus on following aspects: a) Testing and extraction of the “Genetic Properties”; b) Establishment of data base and data warehouse of GMPE; c) Intelligent decision making system for beneficiation process; d) Verification testwork; e) Virtual concentrator. The technical route proposed is to firstly conduct research and test on those Genetic Properties mentioned above, based on that the specific data base will be set up and applied for intelligent process recommendation and simulation with a few verification testwork conducted in the meantime, finally achieve a fast, high efficiency and precise determination of technology and process for ore beneficiation. The systematic study on the three-in-one GMPE would make an innovative breakthrough in conventional mineral process testwork research and engineering methodology, and provide technical support for the concentrator’s design and upgrading.

Jan 1, 2018

By Nilmara Dias, Maria Margarida Martins, Maria Teresa Carvalho

"The heavy residual fraction of the mechanical and biological treatment (MBT) plants treating undifferentiated municipal solid waste is mostly composed by inert materials, such as glass, stones, ceramics and brick. Glass, appearing in significant grade and amount, is not recycled today due to the high content in the other materials, mainly stones. All these materials have approximately the same density but stones have significantly different shape from other materials particles. In this paper it is presented the research developed to eliminate stones appearing in the fine fraction (2.8-8 mm) of the MBT plants heavy residual fraction (MBTr) using a shaking table operated in dry conditions. Two different samples from two different MBT plants were used. Firstly, the samples were characterized (materials content and particle size analysis). Then, laboratory work was carried out to optimize the operation of the shaking table, by changing the machine parameters, deck slope and movement frequency. The study showed that it is possible to upgrade the product in glass. The best results, obtained with one single stage of processing, were 82% content in glass with 84% in recovery, in the glass concentrate, when the coarser fraction (-8+5.6 mm) of one of the MBT plants was used. With the intermediate particle size fraction (-5.6 + 4mm), the glass concentrate attained 74% content in glass with 91% recovery."

Jan 1, 2014

By Atul Kumar Bhatnagar

TATA STEEL is the pioneer in coal beneficiation in India. The first coal beneficiation plant was set up in 1951 at West Bokaro and subsequently in 1953 in Jharia. Over the years more beneficiation capacity was added by setting up new coal washeries in both West Bokaro and Jharia, the coal divisions of Tata Steel. currently the beneficiation capacity available is 8.5 mtp aappx Being a captive supplier supplying the coking coal as per demands of our steel works is challenging in the sense that the intent is to leverage the captive raw material advantage and select the best strategy for maximizing value up to steel making. The beneficiation plants were designed to produce clean coal up to 17% ash. With the demand of increasing productivity at the blast furnace the clean coal ash was lowered. subsequently the processes were audited and efficiency improved. The next challenge was to improve the throughputs in the existing plants to maximize the clean coal production. The upgradation was designed in house and implemented to upgrade the plant from nameplate capacity of 2.1 mtpa to 4.5 mtpa while the yearly production targets were also achieved. The presentation will detail the different phases of improvements, process followed, challenges faced, and the results as achieved in the process. Keywords: coal beneficiation, India, coal washeries, beneficiation plants XXVI INTERNATIONAL

Sep 1, 2012

By ?. Kostadinov

Theoretical study of the behavior of grinding medium presented by TBP (tetraballpebs) was conducted by computer modeling of the position of grinding medium at different speeds and different drum fill factor. For the modeling were used known equations of motion derived for the ball load. Algorithm was created by SolidWorks software, allowing determination of center of gravity of the load in the mill and energy for grinding. These results were compared with similar results obtained by ball grinding medium. The capabilities of both grinding media are demonstrated by graphics under the following conditions: 1.Equal dimensions of the experimental model of the mill; 2.Equal mass of the grinding loads; 3.Comparison of grinding bodies with the same number, mass and density. Keywords: simulation, behavior, grinding medium, balls, tetraballpebs

Sep 1, 2012

By H. J. Wu

The optimum performance in mineral processing plants usually demands the optimum size reduction, which is frequently achieved by a closed grinding circuit. Due to the difficulty associated with fine size classification, the size separation efficiency in closed grinding circuit is usually very low, which leads to huge circulating load and significant overgrinding. Recent development of Der-rick Stack Sizer has made an efficient fine size separation down to 200 mesh with a satisfactory screening capacity. With the efficient fine size separation with Derrick Stack sizer, the problems commonly associated with typical closed grinding circuit, including high circulating load and over-grinding, have been eliminated. Feedback from numerous Stack Sizer applications in iron ore industry indicated that, compared with conventional fine size classification equipment, the circulating load achieved by Derrick screen was usually decreased from approximate 300% down to 60% or so, while the iron recovery has been increased by 6% to 8% only by more efficient size separation in the closed grinding circuit. To study this phenomenon, this paper discusses the relationship between size classification, circulating load and mineral recovery in the closed grinding circuit. A simulation model to predict the improvement in the grinding circuit from more efficient screening is also introduced with this paper.

Jan 1, 2014

By Markus A. Reuter, Jacques J. Eksteen

"The modelling of bath type furnaces tends to be either fundamental in nature or total empirical ""black box models"". While models of the first kind are robust, they tend to lack accuracy, in cases where they are to be used for process control, and may take an overly long in development time. On the other hand, black box models tend to lack robustness of prediction for all circumstances and tend to be limited to interpolate based on historical data. While this may very often provide short-term accuracy, the models may have problems with stability in the long term. Moreover, the model structure may not reflect any of the fundamental properties of the system and remains opaque to fundamental analysis. The aim of the model development is in obtaining a model that have robustness, high accuracy, a short development time and which is adaptive to a changing environment. The models are specifically developed for on-line decision support or control of furnaces. The modelling approach is evaluated for a chromite bath smelting furnace. The role of systems analysis, mass balance smoothing, thermodynamic analysis are integrated into one generic algorithm for bath type furnaces. Not only are fundamental process relationships such the mass and energy balances satisfied through a scientific method of data reconciliation; the proposed modelling approach allows adaptive prediction, the first step towards predictive control or metallurgical decision support.It is shown that neither an ARMAX model only, nor a purely equilibrium model only, could predict the dynamics of the process very well. However, a combination of the two approaches more than doubled the accuracy as measured by the R2. It was noted for carbon dissolved in the alloy, that the improvement in R2 in using a hybrid modelling approach was more than the sum of the R2 of the individual traditional approaches. Similar improvements in modelling performance were noted for all other species.It shows further that process dynamics may totally overshadow the effect of thermochemical equilibrium (even when the thermochemical equilibrium prediction includes the recycled alloy and slag heels as additional feeds). The exclusion of process dynamics therefore leads to erroneous results."

Jan 1, 2003