Search Documents

By L. Aspola

The platinum group metals (PGMs), i.e. ruthenium, rhodium, palladium, osmium, iridium, and platinum, are typically found in association with each other, but also importantly, with nickel and copper. Typically Platinum and palladium have the greatest economic significance and other PGMs are typically produced as co-products of these. With recent increases in the value of nickel, cobalt, and the PGMs and an ever increasing focus on the sustainable use of metals, interest in the processing of secondary feed material sources and metal-bearing residues has also substantially increased. As an efficient, proven, and environmentally sound technology Outotec Ausmelt Top Submerged Lance (TSL) Technology has gained commercial acceptance in the nickel industry, including PGM recovery. For the processing of nickel- and PGM-bearing feed materials, five Outotec Ausmelt Furnaces at four sites are in operation. Two of these furnaces are located in Rustenburg, South Africa, which is by far the biggest mine-based producer of platinum. Outotec also has hydrometallurgical processes available for further treatment and extraction of PGMs, which are also briefly discussed.

Jan 1, 2012

By Serhan Guner, Sundar Chiluwal

Although significant research has been conducted on helical piles, there is a lack of research and official design provisions on the helical pile-to-pile cap connections. These connections may govern the global response of pile cap systems—especially in tall and light structures where uplift forces dominate the response. To improve future industry practice, the objective of this study is to understand the influence of pile-to-pile cap connections on the global behavior of concrete pile caps and develop recommendations for their efficient design. For this purpose, 108 high- fidelity nonlinear finite element simulations are conducted to quantify the influences of bracket type, embedment depths, longitudinal reinforcement percentages, shear span-to-depth ratios, and loading conditions on the load, deformation, cracking, and failure behavior of concrete pile caps. The results indicate that the design configurations involving single bracket type may result in premature concrete cracking around the anchorage zones with a 25% reduction in the uplift load capacity. The investigations also confirm that high reinforcement percentages and low shear span-to-depth ratios result in higher load resistance in all bracket types, and that the load capacities of the double and studded bracket types are similar. For applications requiring the use of lower embedment depths, it is recommended that either double or studded bracket types be used. The research findings have applicability to both helical and micro piles given that both include similar connection types. INTRODUCTION Tall and light structures may experience significant uplift forces at their foundations due to large overturning moments from wind or other lateral loads. Helical piles (e.g., Fig. 1a) can provide a cost- effective, practical, and resilient solution for resisting significant uplift forces. In order to realize the full potential of helical piles, it is imperative that the helical pile-to-pile cap connections (see Figs. 1b, c and d for examples) are properly designed to resist the applied forces. However, there is a lack of research and associated knowledge on the influence of pile-to-pile cap connection detailing on the global response of pile caps.

Jan 1, 2019

By P. K. Kankar, Manoj Khandelwal, Ramesht Shukla

One of the utmost severe mining catastrophes in underground hard rock mines is rock burst phenomena. It can lead to damage to mine openings and equipment as well as trigger accidents or even threat to life as well. Due to this, a number of researchers are forced to study some easy-to-use alternative methods to predict the rock burst occurrence. Nevertheless, due to the extremely multifaceted relation between mechanical, geological and geometric factors of the mines, the conventional prediction methods are not able to produce accurate results. With the expansion of machine learning methods, a revolution in the rock burst occurrence has become imaginable. In present study, three machine learning methods, namely XGBoost, decision tree and support vector machine, are utilized to predict the occurrence of rock burst in various underground projects. A total of 134 rock burst events were gathered together from various published literatures comprising maximum tangential stress (MTS), elastic energy index (EEI), uniaxial compressive strength and uniaxial tensile stress (UTS) that have been used to develop various machine learning models. The performance of machine learning methods is evaluated based on the accuracy, sensitivity and specificity of the rock burst prediction.

Mar 17, 2021

By I Gray

While in some cases it is vitally important to measure the virgin stress in rock prior to tunnelling there are a number of cases where such stress measurement is really not required. The concept of having to have a number to put in a numerical model representing a continuum is often quite inappropriate. This paper discusses the cases when virgin stress measurement is required and the most appropriate means to measure it. The techniques discussed cover all of the overcoring systems, hydrofracture and borehole breakout measurements. It also covers the use of rock surface stress measurement, which is easy to perform and should be used to check for changing stress conditions. Unfortunately this technique has been ignored in recent times because engineers are addicted to numerical models requiring far field stresses rather than measuring the real values that exist where they matter – at the tunnel wall. A case is presented in which the measurement of surface stress in a tunnel boring machine tunnel enabled the far field stress to be determined in what must be the world’s largest overcore.CITATION:Gray, I, 2014. The measurement of rock stress for tunnelling, in Proceedings 15th Australasian Tunnelling Conference 2014 , pp 153–160 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Sep 17, 2014

By H Barrie, L Ameglio

"Current methods for acid rock drainage (ARD) mitigation revolve around insulation of the waste from oxygen in the atmosphere with water and dry (soils and liners) covers. Both are of limited effectiveness and present significant drawbacks. Furthermore, while they reduce oxygen influx, they reduce but do not prevent ARD.The inert gas mixture (IGM) technology has been designed to overcome the deficiencies of gas transfer into tailings and waste rock materials across the surface and instead block oxidation by controlling the transfer of oxygen from the atmosphere. Inert gas is injected into the mine waste at low flow rates and the positive but low flow purges the oxygen that tends to diffuse into the sulfide waste across the surface. The positive flow of IGM compensates for inward oxygen diffusion and creates a pressure drop across the cover layer preventing inward diffusion of oxygen and, hence, acid generation.The IGM processes are evaluated in this paper: theoretically, with a review of the theory of ARD and the current state of knowledge about the performance of notably dry covers as atmospheric oxygen barriers via models to predict the performance of the IGM technique practically, with a bench-scaled experiment.The results demonstrate the effectiveness of IGM to prevent oxidation of sulfide minerals and ARD in waste rock. The laboratory experiments further identified key variables to be used to monitor performance and control efficiency of IGM, which is now awaiting field-scale application.The IGM technology is a patented process by Terraquest Technologies Ltd (Canada) in USA, Canada and Australia. The IGM technology can be cost-effectively manufactured on-site with commercially available compressors and gas filtration membranes. It is predicted that low-cost covers can then be utilised with IGM, which in turn lowers the overall capital costs of ARD mitigation.CITATION:Ameglio, L and Barrie, H, 2015. Acid rock drainage prevention using inert gas mixture technology, in Proceedings Tailings and Mine Waste Management for the 21st Century, pp 85–96 (The Australasian Institute of Mining and Metallurgy: Melbourne)."

Jul 27, 2015

By E Amini, N Beaton

Application of ore pre-treatment technologies such as Grade Engineering and in-pit ore treatment require comprehensive analysis to show proven value is being generated by switching away from the current practices. Considering most mining operations last several decades; the pre-treatment technologies must be investigated over life of mine (LOM) to truly show value was generated or lost. Therefore, a value chain flowsheet model has been usually developed to run those analysis/simulations over LOM using a set of defined ore characteristics and the operating equipment capacities. The Integrated Extraction Simulator (IES) has been used for this type of LOM simulations as it has the necessary flexibility to accommodate the whole mining value chain flowsheet in one platform and provide essential computation power to run mass simulations for the given scenarios using the resource block model. This allows to forecast the value chain response over the LOM for all given scenarios in a feasible amount of time. The traditionally developed flowsheets for the analysis/simulations are usually limited to the mineral processing plant and were developed based on a single audit conducted on the plant in a very short time, e.g. several hours. The process plant is being operated with a certain ore at a certain condition during the short audits which would not reveal the true capacity of the operation equipment when ore characteristics changes. Therefore, it would be impossible to define the true capacity of the equipment and identify the constraints along the value chain. In this study, a methodology was developed to identify the constraints along the value chain and implement those in the IES flowsheet using process information. Also, the equipment models were fine-tuned to represent the true capacity of the equipment in the flowsheet using the same data set. The methodology was tested over five process plant in two operations and showed the performance forecast was significantly improved over the traditional methods being used so far. Then those high-fidelity flowsheets were used to evaluate the ore pre-treatment technologies proposed by the relevant resource company experts.

Nov 10, 2020

By Eugene F. Poncelet

Glass squares compressed on edge by steel jaws in poor contact with them developed jagged "partial-contact" cracks caused by the formation of local tensile stresses. Compressed by steel jaws in perfect contact, they developed smooth "release cracks" during release of the pressure, caused by the formation of local tensile stresses during the release of the pressure. All these cracks were parallel to the pressure. A Microflash photograph of a disintegrating specimen under sufficient pressure reveals a network of fractures roughly normal to each other, also "release cracks" and a disintegration cloud. The Griffith theory is amended to account for the formation of a first crack. A new theory, based on the theory of thermal agitation and wave propagation, is proposed to account for the progress, velocity and forking of cracks. The network of fractures is shown to have been caused by reflection at a free boundary of pressure pulses emanating from a first crack. Postulating equal distribution of energy in the pulses emitted on either side by a crack in particles, the smaller fragments are shown to continue fracturing preferentially while some of the coarser fragments remain as residual pieces. As comminution of the smaller fragments proceeds, the solid is reduced to a collection of residual particles of smaller and smaller sizes, accounting for the disintegration cloud. Experimental Procedure Apparatus A crushing device was built, consisting of a steel frame and two polished, hardened steel jaws to crush plate-glass samples. This device was placed between the plates of a hydraulic press (Fig. I). For centering the pressure, the upper jaw was guided vertically and provided with a rounded top. Later it was replaced by a steel jaw ground out of a 2½-in. steel ball, the lower surface being ground to an optical flat (Fig. 2). Four rollers were used to eliminate side thrust. Screws and centering blocks served to center the specimen in the crushing device. Hard steel shims placed under the shoulder of the upper jaw limited the "follow through'' of the jaw after failure of the specimen. Stress distribution was first examined with a polariscope mounted on an optical bench, using monochromatic light. It was found convenient to mount polaroid, quarter wave and monochromatic plates directly on the device itself. A Microflash unit* was used as light source for the photographs. A gold-paint pattern was fused on the plate-glass specimens and connected into the spark-

Jan 1, 1947

By Eugene F. Poncelet

Glass squares compressed on edge by steel jaws in poor contact with them developed jagged "partial-contact" cracks caused by the formation of local tensile stresses. Compressed by steel jaws in perfect contact, they developed smooth "release cracks" during release of the pressure, caused by the formation of local tensile stresses during the release of the pressure. All these cracks were parallel to the pressure. A Microflash photograph of a disintegrating specimen under sufficient pressure reveals a network of fractures roughly normal to each other, also "release cracks" and a disintegration cloud. The Griffith theory is amended to account for the formation of a first crack. A new theory, based on the theory of thermal agitation and wave propagation, is proposed to account for the progress, velocity and forking of cracks. The network of fractures is shown to have been caused by reflection at a free boundary of pressure pulses emanating from a first crack. Postulating equal distribution of energy in the pulses emitted on either side by a crack in particles, the smaller fragments are shown to continue fracturing preferentially while some of the coarser fragments remain as residual pieces. As comminution of the smaller fragments proceeds, the solid is reduced to a collection of residual particles of smaller and smaller sizes, accounting for the disintegration cloud. Experimental Procedure Apparatus A crushing device was built, consisting of a steel frame and two polished, hardened steel jaws to crush plate-glass samples. This device was placed between the plates of a hydraulic press (Fig. I). For centering the pressure, the upper jaw was guided vertically and provided with a rounded top. Later it was replaced by a steel jaw ground out of a 2½-in. steel ball, the lower surface being ground to an optical flat (Fig. 2). Four rollers were used to eliminate side thrust. Screws and centering blocks served to center the specimen in the crushing device. Hard steel shims placed under the shoulder of the upper jaw limited the "follow through'' of the jaw after failure of the specimen. Stress distribution was first examined with a polariscope mounted on an optical bench, using monochromatic light. It was found convenient to mount polaroid, quarter wave and monochromatic plates directly on the device itself. A Microflash unit* was used as light source for the photographs. A gold-paint pattern was fused on the plate-glass specimens and connected into the spark-

Jan 1, 1947

By A. Gungor

Thermal treatment can significantly improve industrial comminution operations by reducing energy requirements and increasing mineral liberation. The assistance of thermal pretreatment prior to grinding by microwave heating is scrutinized in this study. For this purpose three copper ore samples with different chemical composition were used. To see the effect of microwave heating on ore; work index, breakage rate and breakage distribution functions of original and heat treated samples were examined. The results obtained in this study did not show significant improvement in the grindability characteristics of the studied ores with microwave heat treatment. The inadequate amount of microwave transparent matrix was the main reason for the absence of cracks and micro-fractures in the microwave heated copper samples.

Jan 1, 1999

By Howard E. Mitchell

Modern mechanical raise drilling was first introduced into the United states in 1962. This method of safely, economically raising a shaft has steadily increased in popularity since that time. This steady growth has been largely enhanced by the many advantages to be gained by mechanical raise drilling, as well as by the increasing reliability of the equipment required. It is the purpose of this paper to briefly review the state of the art of raise drilling to include: the various applications and advantages; the equipment requirements, including a comparison of the various types and sizes of machines and rock cutting equipment; equipment performance; and future trends.

Jan 1, 1978

By A. F. Baker

Laboratory flotation of synthetic coal-pyrite mixtures and high-pyritic-sulfur coal indicated that ferric chloride (FeCl3), aluminum chloride (AIC13), chromium chloride (CrCl3), and cupric sulfate (CuSO4) are potential pyrite depressants. For each metal salt there appeared to be a range of pH values and optimum concentration for selectively depressing pyrite in coal flotation. Pyrite was depressed in the pH range from 4.5 to 7.0 with ferric chloride, from 5.8 to 7.0 with aluminum chloride, from 6.4 to 7.1 with chromium chloride, and from 6.4 to 7.4 with cupric sulfate. Best separation of pyrite from coal was achieved in flotation solutions containing from 2 X 10-5 to 5 x 10-5 mole/ liter of each salt. In solutions containing excess salt concentrations, coal was also depressed. These Bureau of Mines tests demonstrated on a laboratory scale that, in solutions containing 4 X 10-5 mole/liter ferric chloride and from 1 x 10-4 to 2 x 10-4 mole/liter sodium hydroxide, the pyritic sulfur content of a Lower Freeport bed coal sample could be reduced from about 2.51 percent to less than 0.60 percent with recoveries of over 83 percent. The depression of pyrite appears to be due to the adsorption of positively charged colloidal metal hydroxide at the pyrite-solution interface. This hypothesis is supported by zeta potential determinations which showed that suspended coal and pyrite particles in solutions containing metal salts are positively charged in the pH range where depression occurs.

Jan 1, 1971

By J. K. Whyatt

During the past 2 years, U.S. Bureau of Mines (USBM) researchers reviewed an in situ stress investigation conducted in 1977 at a test site on the 4250 level of the Lucky Friday Mine, Mullan, ID. Although the field measurements of over core strain were found to be useful, significant deficiencies were found in the stress field estimation procedure. The stress field estimate was then updated to incorporate recent progress in statistical capabilities and new understanding of stress concentration factors for doorstopper cells, as well as to correct an error in the original stress solution procedure. Simple models were used to examine the possibility that systematic variations in material properties and stress fields existed at the site. These models succeeded in reducing the sum of squared error somewhat, but fell short of completely describing stress variations throughout the test site. This overcore measurement will be supplemented by two other measurements described in the second and third reports of this series. Other observations of stress field characteristics are discussed in the fourth report The final report will also characterize the natural in situ stress field in the vicinity of the Lucky Friday Mine. This work was undertaken to support USBM research on mining method design and rock-burst control for deep mines.

Jan 1, 2010

By S. Y. Fantaye

This paper discusses the design, construction and performance of a 30 inch thick diaphragm wall recently constructed in Baltimore, Maryland at Inner Harbor East. The project site is an established marina and waterfront property. The geologic profile consists of fill, alluvial silt and clay, Potomac sand and clay, decomposed rock, and weathered rock on top of gneiss bedrock. The approximately 1,300 ft long perimeter diaphragm wall serves as a water cut-off, perimeter basement wall and support of excavation for the high rise project?s five levels of below-ground parking. Due to the fast track construction and the tight schedule, the initial design of the diaphragm wall was based on a limited amount of geotechnical information. Supplemental borings were then performed along the alignment of the wall to verify the design assumptions and wall depths. Additionally, the design criteria required that the wall be keyed into a competent stratum with low permeability. The Engineer of Record for the diaphragm wall, Mueser Rutledge Consulting Engineers (MRCE) completed the design and produced accelerated construction documents allowing the developer to release the drawings early for bid. This paper describes the theoretical soil parameters used to develop the loads, moments, and shears for the design of the diaphragm wall and the temporary tieback anchors. It also discusses the refinement of these soil parameters using the observed wall displacements and tieback pre-stressing forces as a baseline. This paper compares the calibrated parameters to the values used in the design of the wall. It explains how the soil structure interaction model was developed and the soil parameters calibrated using PLAXIS 2D non-linear finite element software. The paper also discusses the use of the proprietary CWS joint by Soletanche/Inquip JV as end stop between consecutive wall panels and the unique top-down construction utilized to support the diaphragm wall at the Southeast double re-entrant corner.

Jan 1, 2008

By P. N. Anyalebechi

The effects of Si (0.19-0.52 wI. %), Fe (0.20 wI. %-0.44 wI. %), and Cr (0-0.10 wI.%) on the response of ingots of aluminum alloy 3004 solidified at different rates (O.IO°C/s to 30°C/s), to homogenization heat treatment have been investigated. Response to homogenization was determined by quantifying the following microstructural parameters: (i) amount of Mn retained in solid solution; (ii) types, relative amounts, sizes, size distributions, shapes, and chemical compositions of second phases; (iii) extent of the alpha-transformation (i.e., AI6(Fe,Mn) --7 (XAI12( Fe,MnhSi) reaction; (iv) dispersoid formation; and (v) alleviation of dendritic microsegregation. Of the three alloying elements investigated, Si had the greatest effect on the response of aluminum alloy 3004 to heat treatment. Increase in Si content, fineness of cast microstructure, and homogenization temperature facilitated alpha-transformation reaction, dissolution of Mg2Si, loss of Mn supersaturation, and formation of AldFe,Mn)3Si and Mg2Si dispersoids.

Jan 1, 2008

By Frédéric Flament, Mary Goldman, Gille Barbery

"Pilot plant testwork on autogenous and semi-autogenous grinding was conducted at the Centre des Recherches Minérales du Québec (CRM) to investigate the effect of ball addition, solids concentration in the feed, specific energy consumption and feed size distribution.A mathematical model that incorporates breakage and surface mechanisms of loss of mass was developed to predict load size distribution of autogenous and semi-autogeneous mills. An empirical model was used to predict product size distribution of autogenous and semi-autogeneous mills.IntroductionAutogenous (AG) and semi-autogenous (SAG) grinding circuits offer potential capital and operating cost savings when compared with conventional grinding. Empirical relationships have been developed to calculate specific power consumption for SAG mills using pilot plant test results (Mular et al., 1989; Barrett and Allan, 1986). Empirical models can be obtained from plant practice data but are applicable only to the particular ore and set of operating conditions used in the model. Grinding mechanisms have to be taken into account in order to develop a model that can be applied to different ores and mills.In AG and SAG, the ore itself acts as grinding media, with the finer sizes being crushed by the coarser ones. Surface mechanisms of loss of mass, in which the inner core of the particle remains intact, also take place. In SAG milling, the effect of balls also has to be taken into account. The role of each one of these mechanisms changes with particle size and the model has to account for such changes."

Jan 1, 1991

By S. G. Malghan

Steel consumption forms a significant part of the operating cost of a minerals processing plant. An estimated one-quarter billion kilograms of steel in the United States and over one-half billion kilograms in the world are consumed each year in wet grinding alone. The depletion of ore grades and the necessity of finer grinding of abrasive ores are expected to increase the steel consumption even higher. Through continued research and development by the mining industry and the suppliers of steel products several remedies to reduce steel consumption have evolved. This paper critically reviews the mechanisms involved in steel wear, and currently available methods, materials and operating conditions for reducing steel consumption. Particular attention is given to liner design modifications and substitution of steel liners with rubber liners.

Jan 1, 1981

By G Fakes

This paper describes the history of civil engineering tunnelling and developments in soft ground techniques during the 19th and 20th centuries, culminating in the design and use of high speed tunnelling machines for all types of ground conditions. Some of the problems encountered are outlined, and descriptions are given of some contemporary developments in rock tunnelling techniques, with comparisons of methods used in the mining industry.

Jan 5, 1996

By J. -C. Parisien-La Salle, D. Blais, G. P. Demopoulos

"In this work the dissolution, precipitation and adsorption behaviour of gold in acidic iron sulphate media containing small amounts of chloride ions is investigated via batch laboratory tests and OLI thermodynamic calculations over the temperature range 25-250°C. Chloride ions may be present in industrial pressure oxidation/pressure leaching (POX/PL) operations due to elevated water salinity levels or due to intentional addition as is the case of the CESL Copper Pressure Leaching Process. Thermodynamic calculations show that AuCl4 - may form depending on the free Cl-/Au ratio and temperature, its stability increasing with increasing Cl- concentration and decreasing with increasing temperature. On the other hand during POX/PL, we know that massive precipitation of iron in various hydrolysis products and compounds occurs. Hence of interest is the question how AuCl4 - that may form during leaching interacts with the iron precipitation products. Gold(III) chloride was found to co-precipitate with iron(III) (oxy) hydroxides and arsenates forming in the autoclave but not with natrojarosite. Upon further investigation it was discovered that activated carbon added in the autoclave (as done in CIL) acts as preferential sink for the mobilized gold, hence opening the potential for the development of a direct cyanide-free gold recovery process in acidic sulphate leach reactors.INTRODUCTIONThe recovery of gold from ores and concentrates has evolved over the years with processing routes becoming increasingly complex. Given the high value of gold, any reduction in losses during recovery operations constitutes an important economic benefit. One such loss that needs to be addressed is the potential solubilisation of gold during pressure oxidation (POX) of refractory sulphidic gold feedstocks. For example, Agnico Eagle Mines reported a loss in gold recovery due to the formation of a gold chloride complex (Garofalo, 2009). Gold is known to form complexes in solution, such as gold(III) chloride, if an excess of chloride ions are present. The fate of such soluble gold chloride during pressure oxidation and the subsequent neutralization of the acidic discharge slurry will determine if gold is lost or recovered in the downstream cyanidation operation. Thus it is of interest to know under what conditions gold chloride may stay soluble or co-precipitate along primary (formed in the autoclave) or secondary (formed during neutralization) precipitates. Furthermore, the mode of co-precipitated gold, such as adsorption on iron compounds, like goethite and hematite (Machesky et al., 1990; Nachayev, 1984); substitution in one of the precipitated phases; or finally chemical reduction to metallic state would have consequences in terms of gold recovery."

Jan 1, 2012

By P. J. K Leeuw, E. Uludag, M. J. Nong

Efficient drill-and-blast performance is pivotal to the profitability of deep-level mines, where high operating costs and declining grades demand maximum throughput. This study investigated persistent fragmentation problems at a longhole stoping operation that suffered from simultaneous oversized rock blockages in 300 mm × 300 mm grizzlies and excessive fines. Fifty-one muckpile images from five blasted stopes were analysed with Split-Desktop. Although the overall F₈₀ was 287 mm, nominally suitable for the grizzly aperture, the Rosin- Rammler uniformity index of 0.80 revealed a highly inconsistent size distribution containing both boulders and fines. Existing drill patterns were then evaluated with the Aegis underground software, which showed negligible crack overlap at the toe owing to inadequate spacing and burden. Insufficient energy overlap between blastholes, particularly at the toe, was found to contribute to the coarse fragmentation, while excessive fines were attributed to the crushing and severely fractured zones of blastholes. Optimisation simulations demonstrated that an 89 mm blasthole diameter on a 2 m × 2 m pattern would overlap, aligning with Aegis requirements of balancing powder factor, break angle, and energy distribution. Implementing this revised pattern is expected to reduce secondary breaking, prevent ore-pass blockages, lower fines, and ultimately improve overall production efficiency. The methodology illustrates how integrated image analysis and modelling software can effectively diagnose and address fragmentation issues, particularly in underground longhole mining, where challenging blasting conditions require a balance between achieving optimal fragmentation and maintaining stope stability. This approach offers a cost-effective pathway to improving comminution performance in deeplevel mines.

Oct 2, 2025

By Weiran Zuo, Jinsheng Guo, Chun Yu, Jinyu Zhan, Bao Guo

High-precision prediction of the target minerals’ content in the feed and concentrate products is vitally important for the efficient beneficiation of mineral resources. Visible and near-infrared (NIR) spectroscopy provides a cost-effective way for online measurement of the mineral content in an industrial process. In this investigation, simulated specimens consisting of five different types of minerals that are present in the copper oxide ore, malachite, quartz, calcite, hematite, and chlorite were prepared with a variety of malachite contents, and the mixed specimens were analyzed by a visible-NIR spectrometer in an integral sphere mode. The reflectance spectrum is used as input and the malachite content as output to build the prediction model. The obtained data was modeled by support vector machines (SVM), and a Grey Wolf Optimization (GWO) is proposed with the goal of improving the prediction accuracy. The GWO algorithm has been applied to adaptively search for the best combination of featured values. After cyclic comparison, the optimal penalty factors c and g can be quickly and accurately selected. The experimental results show that the SVM model established by the GWO algorithm has a better fitting effect and smaller prediction error, compared with other models.

Aug 3, 2023