Search Documents

By A. R. Trusiak, C. Bazin, D. Hodouin

"The On-Stream Analyzer has proven, and is still proving to be, an extremely valuable tool for the control of flotation circuits. This paper examines the possibility of going further in the utilization of the analyzer results by using a real-time material balance program to calculate the mass flow rates within the streams of a circuit for process control, analysis and optimization as well as for the detection of sensor malfunction. The material balance algorithm improves measurements and give good estimates of unmeasured variables during normal operating conditions. The potential and limitations of the method are illustrated using results of an application to the Line 3 flotation circuit of Brunswick Mining and Smelting, New Brunswick.IntroductionThe On-Stream Analyzer (OSA) has gained, over the last two decades, the status of key instrument for the control of flotation plants. The OSA is currently used in almost every sulphide flotation plant and its application in oxide mineral flotation plants isincreasing. Commercial OSAs include a sampling system that collectssamples from various streams of a circuit and exposes thesamples to an X-ray beam. The sample then generates X-ray pulsesthat are converted into assays using multilinear regression programs(Leroux and Franklin, 1994). The estimated assays arepassed to the flotation operators via operator consoles, or to controlloops of various degrees of complexity (McKee, 1991), whichmanipulate reagent addition, aeration rate or froth thickness toensure an adequate operation of the plant. Although this applicationrapidly pays for the purchase of the OSA, one should ask:""Could we get more from the OSA?"""

Jan 1, 1995

By Y. X. Li

In order to control the pulp level in flotation column which has long time-delay and great inertia, and is difficult to establish an accurate mathematical model, a Fuzzy-PID dual mode controller is applied. This method takes advantage of both strong flexibility and good reliability of fuzzy logic and high precision of PID. With the very stable and reliable process instruments, the control system has a good control performance and results the stable control of flotation column.

Jan 1, 2014

By B. Salt, R. Sullivan, E. Roos, M. de Vries, K. McKenna, L. R. Nelson, M. J. N. Uys, P. Jacobs, E. Munnik, P. Lewarne

A review is presented of the furnace cooling system design and operating developments since the initial furnace start up in January 2001 of the Chambishi DC-arc furnace to produce ferrocobalt alloy. The furnace operating conditions present a number of challenges to the furnace designers and operators in particular to provide a sustainable lining that can withstand the highly corrosive siliceous slag (roughly 50% SiO2). This is exacerbated by the dual requirement to operate this slag in a relatively highly superheated state (?T ? 400°C when tapped at 1550°C) to ensure that the ferroalloy can be tapped (estimated 1370°C alloy liquidus temperature) prior to atomisation. The core design challenge occurs at the alloy-slag interface, where a single refractory type cannot simultaneously meet the dual requirement of both adequate alloy and slag corrosion compatibility. Progression by the fourth furnace campaign to involve a retrofit of the bath sidewall to Hatch high-intensity, water-cooled, copper Waffle coolers and tapblocks is described. This involves operation of the entire slag bath, the slag-alloy interface and a portion of the alloy within a more chemically compatible refractory/slag (and possibly at times even an alloy) freeze-lining. Due to the high furnace power density employed (up to 500 kW/m2 hearth area) coupled with the large slag superheat needed, the resulting imposed peak bath sidewall heat fluxes are substantial, especially in the alloy (typically averaging 170 kW/m2), and on occasion have exceeded 1000 kW/m2. The performance of the Hatch Copper Waffle coolers under such aggressive process pyrometallurgical conditions is described. Aspects of the design and especially operating strategies to effect sustained production at up to 38 MW power input are also discussed.

Jan 1, 2004

By L. R. Nelson

A review is presented of the furnace cooling system design and operating developments since the initial furnace start-up in January 2001 of the Chambishi DC-arc furnace to produce ferrocobalt alloy. The furnace operating conditions present a number of challenges to the furnace designers and operators, in particular, to provide a sustainable lining that can withstand the highly corrosive siliceous slag (roughly 50% SiO2). This is exacerbated by the dual requirement of operating this slag in a relatively highly superheated state (?T ?400°C when tapped at 1550°C) to ensure that the ferroalloy can be tapped (estimated 1370°C alloy liquidus temperature) prior to atomization. The core design challenge occurs at the alloy-slag interface, where a single refractory type cannot simultaneously meet the dual requirement of both adequate alloy and slag corrosion compatibility. Progression by the fourth furnace campaign to involve a retrofit of the bath sidewall to HATCH high-intensity, water-cooled, copper Waffle coolers and tap blocks is described. This involves operation of the entire slag bath, the slag-alloy interface and a portion of the alloy within a more chemically compatible refractory/slag (and possibly at times even an alloy) freeze-lining. Due to the high furnace power density employed (up to 500 kW/m2 hearth area), coupled with the large slag superheat needed, the resulting imposed peak bath sidewall heat fluxes are substantial, especially in the alloy (typically averaging 170 kW/m2), and on occasion have exceeded 1000 kW/m2. The performance of the HATCH Copper Waffle coolers under such aggressive process pyrometallurgical conditions is described. Aspects of the design and especially operating strategies to effect sustained production at up to 38 MW power input are also discussed.

Jan 1, 2004

By P. H. Brace, N. A. Ziegler

THE study of the relations between gases and metals is one of perennial interest to all who are connected with the production of high-grade metallurgical products. The data reported here are the outcome of work of which the object is to determine the effects of dissolved gases upon the properties of metals and to gain a more detailed knowledge of the reactions between gases and molten metals, particularly iron, under definitely controlled conditions. We have attempted to reduce our margin of error by dealing with relatively large quantities of metal, from 4 to 13 lb. (2 to 6 kg.), and keeping the ratio of refractories to metal as small as practicable. The ingots produced are large enough to provide adequate material for making determinations of magnetic and other physical properties. In choosing the bell-jar type of furnace we were influenced by the results of considerable experimentation along the same general line which indicated that, for the scale on which we were to work, it was important to have the furnace structure independent of the vacuum enclosure. The use of a bell-jar surrounding the entire furnace accomplishes this and also gives great freedom of observation during operation and ready accessibility for the replacements which are necessary after each run. Our choice of analytical method was made in the light of the results secured by H. M. Ryder, who used a method similar in principle for the analysis of very small quantities of gas extracted from strips of metal heated electrically in high vacuum. We have presented some selected preliminary experimental data as being illustrative of the quantity and nature of the gases which occur in iron of different origins, but do not urge any hypothesis at this time. THE FURNACE Our furnace may be described as "a high-frequency induction furnace of the bell-jar type." It is similar in a general way to that described by Cain and Peterson' though we have introduced certain refinements of

Jan 1, 1928

By K. Maruyama, T. Nagasaka, R. d. Chairaksa-Fujimoto, T. Miki

Electric arc furnace dust or EAF dust is considered as a hazardous waste but contains a significant amount of economically attractive non-ferrous metals especially zinc as zinc ferrite (ZnFe2O4) and zincite (ZnO). The hydrometallurgical process is one alternative choice because it has the advantage of low energy consumption. The leaching of ZnO from the dust is very easy. However, the complex major zinc compound in EAF dust is ZnFe2O4, which is very stable and insoluble in alkaline or acidic media under delicate conditions, and causes the low zinc recovery yield by hydrometallurgical process. Moreover, halogens and heavy metals in the dust are also problematic. In order to develop a new effective EAF dust treatment process, the authors have proposed "CaO addition process" as pre-dust treatment to transform all of ZnFe2O4 in the EAF dust to ZnO and Ca2Fe2O5. In this study, the selective zinc leaching from the CaO treated EAF dust with NaOH and NH4Cl solutions was carried out. The effect of temperature, reaction time and solute concentration were systematically investigated. All of zinc in CaO treated EAF dust was successfully leached into both solutions.

Jan 1, 2014

By F. P. Hassani, H. B. Poorooshasb

"A kinematically admissible velocity field is applied to study the arching phenomenon in a cohesionless granular backfill placed between two rough vertical inflexible walls. The study is different from the previous works which are almost entirely based on statically admissible stress fields. It is shown that the development of arches is strongly influenced by the previous loading history of the backfill which controls the nature of the constitutive laws that govern the mode of stress deformation of the backfill. This preliminary study employs simple models in order to show the potential use of such an approach in evaluating and understanding mine backfill behaviour in the underground excavation. The pressure exerted on the floor supporting the backfill is also studied in some detail.INTRODUCTIONAlthough the importance of the design of mine backfill has been discussed extensively by a number of authors, i.e. Hassani et al. (1985), such theories have not as yet been incorporated in design criteria for underground mine backfilling operation . One factor that such theories must incorporate is the so-called arching action .The phenomenon of arching has been the subject of study by many researchers in the field of civil engineering and mining engineering. Among these scholars the works of Terzaghi (1943), Krynine (1945), Lusher and Hoeg (1964) and Handy (1985) are noteworthy. All these authors use a statically admissible stress field as a starting point to develop their theories. The approach taken by the authors of the present paper is completely different because they employ a kinematicallyadmissible velocity field. It is their belief that the arching phenomenon is closely linked to the interlocking of grains as manifested by the degree of dilatation of the granular mass when subjected to a shearing process.This paper is written in the following sequence. To define the interlocking quantitatively it is necessary to state the constitutive law governing the stress deformation characateristics of the granular mass. This is done in the next section as the law proposed by Poorooshasb and Pietruszazck (1985, 1986) is stated and simplified to suit the present analysis. Next the mode of formation of an arch in a simple system consisting of a mass of backfill sandwiched between two vertical rough walls of infinite extent is examined. The velocity field obtained from the above example is used to obtain solutions to two more complicated situations. The first situation assumes that the backfill is supported, in addition to the vertical rough walls, by a Winkler foundation at its base. The second situation assumes that the Winkler foundation has been replaced by a horizontal beam of uniform section and simply supported at its extremities. The findings of the study are summarized and their practical implications discussed in the concluding remarks."

Jan 1, 1989

At the exploration stage for new mining projects or for medium- and long-term areas of existing mines (typical of South African gold mines), drilling data is on a relatively large grid. This grid is normally larger than the selective mining units (SMUs). Direct estimates for the SMUs and also of much larger block units will then be smoothed due to the information effect and the high error variance. Any capital-intensive project decisions made on the basis of any of the smoothed estimates will tend to misrepresent the economic value of the project or operation, i.e. the average grade of blocks above cut-off will be underestimated and the tonnage overestimated for cut-off grades below the mean grade of the orebody. This paper presents a direct approach technique for deriving recoverable resources, referred to as localized direct conditioning (LDC), designed to correct smoothing effects and also to provide support corrections.

Jan 1, 2014

By Hal Palmer

A number of Articles contained in the United Nations Convention on the Law of the Sea III (UNCLOS) bear directly on legal issues related to the recovery of mineral deposits on the ocean floor. Mineral resources of the continental margin other than oil, gas and hydrates include, sand, gravel, shell, coral, gemstones and phosphorite. Jurisdiction over the exploration and exploitation of these deposits is determined either by the Territorial Sea claims of a coastal state or their recognized sovereign rights in the Exclusive Economic Zone (EEZ). At least 33 coastal states have the right under Article 76 (Continental Shelf regime) to claim seabed jurisdiction beyond 200 nautical miles (EEZ limit). In addition, several states have "Pioneer Investor" status with regard to seabed resources in international waters (The Area). A global maritime boundary database (GMBD) has been developed as a basic resource identifying the current claims of coastal states. It provides detailed information on existing boundaries and limits determined by the UNCLOS accords or by multi-lateral agreements which affect mineral exploration and exploitation. This paper will describe the GMBD and its construction and maintenance.

Jan 1, 2000

By T. J. Carew

INTRODUCTION The application of micro-computers to the evaluation of an advanced exploration project is illustrated by reference to a large project in the Sudbury Basin region of Canada. The deposit in question is polymetalic (Cu, Pb, Zn, Au, Ag) with a strike length of 3048m (10000ft) and has been explored to a depth of 457m (1500ft) by approximately 1350 diamond drillhole of between 9 - 305m (30-1000ft) in length. The large amount of data involved and the ability to rapidly recalculate geological reserves in response to fluctuating metal prices were key factors in the application of a micro-computer based evaluation system. The evaluation software was implemented at two levels, with data capture implemented mainly on - an entry level workstation comprising an IBM-PC compatible (PC-XT) micro-computer equipped with a 20Mb hard disc, printer and RS-232 serial communication facilities. Data analysis, display, geo- logical modelling and processing was carried out on a high-level workstation based on a Hewlett-Packard 9000 super- micro computer and peripherals, including a 40Mb hard disc, tablet digitizer and plotter. Basic drillhole data was entered either directly at the super- micro or transferred from the PC via the RS-232 link. The evaluation process proceeds once the data is edited and in a "clean" format within a master project database. The software is comprised of a suite of modular, integrated programs based on ease of use, interactive graphics and optimized for the computer workstation. Drillhole data was displayed and plotted as hard copy plots on which geological interpretations were developed. The data was copied into small manageable projects consisting of 100 to 150 drillholes, within which a sectional model was created such that all the models are interconnected along the entire strike with no overlaps. Geological reserves were derived by developing a set of function variables, including Net Smelter Return, within the database. The values of these functions can be displayed on the model sections and "ore” zones defined in accordance with pre-determined cutoff values. Interpolation of the modelled data was then carried out and geological reserves calculated and reported. DATA CAPTURE Drillhole data for the project consisted mainly of existing drillhole logs, although the system allows for the entry of new information as it becomes available. The structure and format (i.e. number and length of variable fields) is user defined. The log for this project comprises an abbreviated section for name, competency, grain size, texture, colour and alteration. Fields are defined for structural information regarding bedding and fault angles and up to seven mineral identification columns and percentages are available. Assay information consisting of the sample number, width and six determinations (Cu%, Pb%, Zn%, Fe%, Au opt., Ag opt.) along with a comments section of 25 characters completes the log. One powerful feature of the data- base is the ability to set up function variables that are mathematical and/or logical combinations of the project variables, enabling the user to develop economic variables such as Net Smelter Return. These functions include current treatment costs, deductions and penalties and can be rapidly recalculated within the database to accommodate changes in metal prices, forecast recoveries and exchange rates (Table 1).

Jan 1, 1987

By M. E. Wadsworth, R. D. Groves, B. W. Madsen

Experimental results obtained from large-scale (7 ton), long-term (500 days) leaching of low-tenor copper sulfide ores were interpreted by a geometric model involving the movement of a reaction zone through the ore fragments. The postulated leaching mechanism involved mixed kinetics and included diffusion of the reacting species through the host rock and a chemical reaction at the surface of the mineral particles. Ore properties which were considered were Ore particle size and shape, grade of ore, and mineral content. Excellent agreement between the mixed kinetic model and the laboratory leaching results was obtained when -1/2-in. ore particles were not included in the leach. Also, a satisfactory prediction of copper recovery from +1/2-in. ore with a different particle size distribution was made; however, it was necessary to modify the shape factor in the model. Prediction of copper recovery from an ore sample which included the -1/2-in. fine fraction and leaching for 460 days was 56.5%, and the recovery from leaching was 47.0%. The presence of fines caused solution channeling and a deficiency of oxygen which inhibited the leaching rate. Also, compaction of fines around coarser particles interfered with circulation around the ore particles, and bulk solution transport probably was a rate controlling factor.

Jan 1, 1976

By Yannis Faitakis, Simon Bergeron

"This paper describes the implementation of model-based predictive control (MPC) on a vacuum heat exchange network at Alcan's Jonquière Complex. The plant produces alumina from bauxite ore using the Bayer Process.The MPC system was commissioned in 2000, by Alcan personnel and an Advanced Control Engineer from Matrikon. The application has achieved an uptime of about 97%.The benefits were:• Energy savings : 0.5% - live steam reduction• Reduced variation: standard deviation of key temperatures reduced by 50%• Plant throughput increase: via the automated use of surge tanks• Decreased supervision time: by plant operations• Faster and easier diagnosis of problems and plant constraintsThe benefits were higher than expected with a payback time around 12 months.The purpose of the vacuum heat exchange network is to recover energy from a hot stream (pregnant liquor) by transferring it to a cold stream (spent liquor). The system is very tightly coupled, and the main constraints of the multivariable control system are: • to control the many tank levels involved in the process (8 tanks)• to maximize the temperature of the heated streams (spent liquor)• to prevent cooled streams (pregnant liquor) from cooling to below a minimum temperatureThe system includes 9 manipulated, 14 controlled and 2 feedforward variables.The advanced control implementation runs under an expert system environment and provides added flexibility to the overall control strategy."

Jan 1, 2003

By P. D. Duever, P. J. Readyhough, P. L. Silveston

"The Plastojrost apparatus, developed by Ritter and Juranek in the 1960s has been modified primarily, to automatic temperature control. A limited investigation of carboniferous and cretaceous coal samples showed that both coals undergo the same changes when coked in the Plastofrost unit. Dilatometer plastic range and total dilatation correlated well with the Piastofrost plastic range for the carboniferous Appalachian coal, but not for the cretaceous western Canadian coal. Work to elucidate the characteristics of some western Canadian coals using the Plastofrost is in progress. IntroductionCoke quality is critically affected by the properties of the plastic phase in the coal-to-coke transformation. Since these properties exist in a coke oven and are not measurable, they are approximated by standard laboratory tests, such as free swelling index, plastometry, and dilatometry. Of particular interest to us is the anomalous behaviour of western Canadian coals of cretaceous origin. By many predictive methods, these coals should not produce coherent cokes. When coked in pilot scale test ovens, in fact, they produce cokes of good stability. They typically make up 20% of Japanese coking coal blends and are also used in South Korea, Mexico, Brazil, and several European countries. This anomalous behaviour has been studied in our laboratory<8.9.1O). Plastic properties or rheological tests examine just the external ramifications of the coal-to-coke transformation. To understand the process, however, microscopic examination of the coal and coke seems to be needed. The Plastofrost unit has been shown to be a convenient means to this end."

Jan 1, 1987

By C. Y. Sun

In this paper, the development of research and practice on the separation process of fine complex Cu-S ore at a Copper Mine in south of China, was discussed. Since early 1980s, various processes were used, such as the process of bulk flotation-separation, the process of selection flotation and the process of iso-flotation. Especially when the process of selective flotation was put into operation, further development was achieved in beneficiation results, but the difficulty in separation of chalcopyrite from pyrite has not been solved yet, so causing a poor quality of Cu concentrate. The author has successfully developed a new separation process characterized by the use of collector thionocarbamate and fast flotation. The selectivity of fine chalcopyrite and pyrite was greatly improved. This process supplied a new route for the treatment of fine complex Copper sulphide ores.

Jan 1, 2014

By G. Sahasrabudh, G. DeIulii, K. P. Galvin

A novel agglomeration process has been developed, utilising a concentrated water in oil emulsion as a binder. The binder is typically 95 wt% water, the remainder being an organic phase containing kerosene and the emulsifier— sorbitan mono oleate (SMO) in equal mass portions. The binder can selectively agglomerate hydrophobic particles in less than 7 seconds. The agglomeration has proven to be highly effective in recovering any particle of virtually any size, provided the particle is hydrophobic. The approach has previously been applied to coal, magnetite and silica. In the paper, sodium isobutyl xanthate (SIBX) and its much longer chain version, potassium tetradecyl xanthate (C14X), were used separately as collectors to selectively agglomerate chalcopyrite from the ore. Xanthates attach to metal sulphides, thus hydrophobising sulphide mineral particles. Chalcopyrite is a sulphide mineral; however, it hydrolyses quickly in the presence of water. Ethylenediaminetetraacetic acid (EDTA) was used to remove the oxide layer formed at the surface of the mineral particles after wet grinding, leaving behind a freshly generated metal sulphide surface. After the EDTA treatment, the ore particles were quickly conditioned at an optimised pH as well as collector concentration and agglomerated. The agglomeration achieved a copper recovery of 75% within 7 seconds, consistent with the recovery achieved by flotation over much longer time scales. Keywords: Chalcopyrite, agglomeration, froth flotation, interfacial chemistry, X-ray fluorescence, selective hydrophobisation

Jan 1, 2020

By Z. Xingrong, L. Jiongtian, Z. Yangge, C. Yijun, L. Liang, H. Yangjia, H. Long

In this present research, a polymer poly (acrylamide-allyl thiourea) (PAM-ATU) with an average molecular weight 1.66 × 104 g/mol was used as a depressant in the flotation separation of molybdenite from chalcopyrite and galena. The micro-flotation experimental results showed that PAM-ATU exhibited a strong influence on the flotation performances of chalcopyrite and galena, but had little effect on that of molybdenite within a pH range of 9.0~11.0 at a dosage of lower than 6.0 mg/L. The flotation results of bulk flotation tests illustrated that PAM-ATU showed a stronger depression power than sodium thioglycollate (TGA) and sodium sulfide. The comparison results demonstrated that PAM-ATU and TGA had an advantage in reducing galena and chalcopyrite impurities in the molybdenum concentrate, respectively. With the combination of PAM-ATU and TGA, a good flotation separation result was achieved through one roughing operation where the recovery of molybdenum in the froth product was over 75.0%, while the recovery of copper and lead were less than 22.0% and 15.0%, individually. It was also found that a molybdenum concentrate with Mo grade of 46.32%, Cu grade of 0.16%, Pb grade of 0.06%, was obtained by a closed circuit flotation flowsheet of one roughing-three scavenging-five cleaning with a Mo recovery of 77.67%. Keywords: PAM-ATU, molybdenite, depressant, chalcopyrite, galena

Jan 1, 2020

By Peter Zhang

The immediate roof at the test site consisted of laminated clay shale that is likely to be susceptible to weathering. With seasonal dry-wet cycles in the mine ventilation air, the roof in the intake air would deteriorate and lose integrity over time. In many instances, the weathering-prone roof would gradually scale off piece by piece between bolts and over the entry corners. In some long-term entries, roof weathering could develop up to 2-3 ft deep with columns of bolted roof hanging under the upper intact roof. If wire mesh is used, the roof bags down and reduces the entry height, and the weight of the bagged roof tends to pull the bolts out and poses a safety hazard. An effective measure is always sought to support this type of weak roof that is sensitive to moisture degradation. One solution is to seal the roof shortly after entry development to prevent moisture penetration so that roof integrity can be maintained for the long term. In this application, a polymer-based sealing material was tested in the Rockspring mine for weathering protection and reinforcement. The material is a one-component polymer powder for spray application onto the roof and rib. The sealant material is mixed with water in the spraying nozzle to apply to the roof and rib as a paste. The product sets within 5-10 minutes and forms a membrane as thin as about 1/8-in. The membrane will quickly develop tensile and bonding strength within minutes and completely cure within a few days. Through sealing, bonding and lateral reinforcement, the integrity of the weak roof can be maintained. The material has been tried in a long-term evaluation over several months in a set of main entries in the Rockspring mine owned and operated by Rockspring Development, Inc., affiliate of Foundation Coal. Preliminary evaluation has shown that the material is an effective and economical solution to ensuring the long-term stability of weathering-prone roof.

Jan 1, 2009

By D Wiseman, D Weedon, T Akroyd

Models of the size specific processes of particle collision, attachment and detachment have been combined in a form suitable for analysis of a full-scale flotation plant. The model is based on a first principal approach to particle collection by bubbles. Parameters considered include particle size, bubble size, particle contact angle, energy dissipation, entrainment and froth recovery. Using data structures configured within an Excel spreadsheet using æLimn: The Flowsheet ProcessorÆ, the model has been incorporated into a description of the individual cells within a simple roughing and cleaning flotation circuit. Simulated feeds of varying particle size distribution, feed tonnage and varying mineral hydrophobicity representing different mineral phases are presented to the model to produce predicted final concentrate grades and recoveries. Practical techniques for measuring many of the parameters included in the model are now becoming readily available. However, relating the terms included for contact angle, energy dissipation, entrainment (ie water recovery) and froth recovery to measurable parameters in operating plants remain as continuing challenges for research.

Jan 1, 2004

By B. R. Adams

This paper discusses the use of a reacting CFD model to determine chalcopyrite kinetics in a drop tube furnace and to predict the reaction of a chalcopyrite concentrate in the reaction shaft of an industrial smelter. Reacting CFD codes can be applied to determine the consequences of the interaction of gas with particles and therefore improve the ability to derive kinetic parameters that take into account the temperature and oxidation histories that different particles will see. This paper describes the application of such a model to derive improved kinetic parameters for the pyrolysis and oxidation of chalcopyrite. The reaction shaft model includes the effects of turbulent fluid mechanics, entrained flow mixing, turbulent particle dispersion, heterogeneous particle reactions, radiative and convective heat transfer, and&apos; surface and bath deposition rates. Particle reaction and composition characteristics are predicted as a function of particle trajectory and deposition and are used to aid in evaluating shaft performance.

Jan 1, 1999

By A Horeth, E Duggan, B Versiani, A Gooch

Iron ore screening plant productivity is strongly affected by the performance of the shuttle operation, with the control system robustness and associated strategies having a direct impact on utilisation and throughput rates. A number of aspects intrinsic to the methods and technologies traditionally applied on the motion control of shuttles are often observed to decrease the efficiency of the ore distribution process across multiple screening modules, which ultimately impacts the plant’s production. Those include the number of operational scenarios that are unmanaged by conventional discrete control logics which may intermittently lead to undesired events, such as ‘high level alarms’ and ‘low level interlocks’ in the screening bins. This outcome usually demands excessive manual intervention by Control Room Operators. A Reinforcement Learning (RL) Artificial Intelligence (AI) solution was deployed to control the motion operation of the scalping and products screening shuttles in an iron ore handling plant. The RL algorithm was trained in advance to handle multiple operational scenarios and generate optimised outputs. In nominal operation, the shuttles are moved in a modulating and highly predictive fashion over the whole range of safely reachable bins. When the need occurs, the AI control agent deviates from that modulation behaviour, limiting its travel range or moving to other groups or clusters of available bins by employing ‘bin jumping’ or ‘ore break’ strategies. The RL platform capabilities include the ability to learn in real time to account for and auto-adapt to new operational patterns and process changes whilst applying improved strategies for shuttle dwelling times, target setting and speed control. The deployed Reinforcement Learning AI solution has better managed the existing challenges associated with the shuttle operation and control, and resulted in a more efficient ore delivery process, superior operations stability and increased production rates.

Nov 8, 2021