Search Documents

By G. McIsaac

This paper discusses the formulation of a linear program with the objective of maximizing the cash flow of the long-term plan of an underground operation. A mine is split into eleven independent ore zones, each characterized by its reserves, required development, and mining method. Based on practical considerations, maximum rates are set for each zone?s development and production capacity. For the total mine plan, a minimum daily production rate is set along with a minimum feed grade to the mill. A maximum daily development capacity is set based on the equipment and labour available. Proven and probable ore is mined preferentially over possible ore in the early years of the plan. Also, the fixed cost of operating the mine is taken into consideration whenever mining activities take place. Once the linear program is solved, it can be critically reviewed, and unrealistic results can be identified. Some resource allocations constraints can then be changed in certain sectors, and new results can be analyzed. ?Whatif? scenarios can also be examined, such as determining the impact of hiring a contractor to develop a given sector, or dropping an ore zone completely from the plan.

Sep 1, 2005

By Luke Evans

What are Best Practices? What are Bad Practices? Who Uses Best Practices? Why Bother? Acceptability Guidelines Published Best Practices Published Bad Practices Additional Best Practices Some Best Practices for Low Grade Deposits Interpretation Guidelines Big Bear Gold Deposit Example Conclusions Questions

May 1, 2013

By Jason Mewis

Jason Mewis is the founder and president of ENGCOMP, a consulting engineering firm in Saskatoon Saskatchewan focusing on the mining sector. He has over 17 years of experience working on projects for the heavy industrial market in Saskatchewan, as both a Senior Structural Engineer and Project Manager. Jason is an entrepreneur with experience in many areas including estimating, planning, facilitation, and Risk Analysis. He has presented at many seminars and conferences across North America on the subject of Project Risk and and has become recognized for his expertise in this area of project management.

May 1, 2013

By Jack Parker

"The original design of mine openings and pillars at White Pine was based upon traditional assumptions; i.e., that the principal load is due to the weight of overburden, and that horizontal stresses are equal to one-. third of the vertical.Beginning January, 1964, borehole deformation measurements confirmed suspicions that the principal stress is horizontal and several times greater than the vertical.Vertical stress appears to be directly related to depth, but lateral stresses vary considerably in direction and magnitude. The direction of major principal stress lies parallel to the bedding and is apparently parallel to tension joints. Lateral stresses appear to originate from an igneous intrusive.Mining induces stresses which add to the primitive stresses, in some areas precipitating failure. In this paper, rock failures are illustrated, methods of measurement and typical results are given, and measures to counter or utilize the lateral stresses are described."

Jan 1, 1966

By J R. Barber, B. E. Rogers, R. Fukuzawa

Mercury is sometimes present in non-ferrous ores including gold, zinc, copper and lead. The selection and use of appropriate mercury abatement technology is critical for preventing its release to the environment during ore processing and to protect workers from mercury exposure. This paper presents a general decision-making process for mercury abatement technology selection for major emission sources in the gold industry. Mercury abatement can be accomplished through making process changes and/or by adding a process step to an existing system (i.e. "end-of-pipe control"). Advantages and limitations of the technologies along with waste / by-product management options are discussed through case studies of current operating mercury abatement systems.

Jan 1, 2015

By G. Scamans, J. Lazaro-Nebreda, Z. Fan, J. B. Patel

"Melt quality is crucial for both continuous and shape casting of light alloys and in order to improve it, removal of excessive inclusions, accumulated impurities and unwanted gases becomes necessary. To address this problem, a new multi-purpose high shear liquid metal treatment technology, based on intensive melt shearing by a rotor/stator unit, has been developed. The device ensures achieving uniform temperature and chemical composition and an efficient dispersion of gas and solid phases in the melt with minimal disturbance of melt surface, and can be used, among other things, for de-gassing and de-ironing of Al-alloy melts. Efficient degassing is achieved by dispersing each argon bubble into many small ones, which significantly increases the overall surface area of the bubbles and the time they remain in the melt capturing hydrogen before reaching the surface. This allows using much lower inert gas flow and mixing times than conventional rotary degassing. When applied to Al-alloy scrap, with accumulated iron, the high shear melt treatment speeds up the nucleation and growth of primary Fe-rich intermetallic particles, allowing a faster removal rate. Therefore it reduces processing times, increasing recycling productivity, and facilitates full metal circulation of secondary Al-alloys.INTRODUCTION Aluminium and magnesium alloys are widely considered in various industrial sectors for the purpose of weight saving (Polmear, 1995). In particular, they are used in the automotive sector, where components are mostly manufactured by high pressure die casting (HPDC) because of its simplicity and relatively low cost, but efficiency and mechanical properties are quite low compared with other processes. To date, the main focus to overcome this problem has been for improving the HPDC machinery (Dahle, 2001) but little attention has been paid to the opportunities of controlling the nature and quality of the molten metal supplied to the die. In this way, liquid metal treatment, which involves a variety of physical and/or chemical processing methods to improve melt quality prior casting, is crucial for both continuous and shape casting products."

Jan 1, 2018

By M. Drozd, M. Somppi, Z. F. Yamak

"Activated carbon is used to recover gold in solution from various gold circuit designs including carbon-in-pulp (CIP), carbon-in-column (CIC), and carbon-in-leach (CIL) processes; however, the characteristics and durability of the activated carbon can vary. Handling this medium poses challenges when fine carbon losses occur in the CIP, CIC, and CIL circuit discharge. Additionally, adsorbed gold on carbon fines can be lost to the discharge. The studies in this paper aim to standardize laboratory methods for testing activated carbon qualities specific to the gold industry. The investigation focused on size distribution, platelet content, wet attrition resistance, gold kinetics, and gold loading. The results demonstrate that a 96 h attrition test and a gold loading test form a basis for ranking carbon samples based on expected economic performance of the plant. ¦ KEYWORDS Activated carbon, Carbon attrition, Carbon ball-pan hardness, Carbon evaluation, Carbon fines loss, Carbon turnover RÉSUMÉ Le charbon actif sert à récupérer l’or en solution dans diverses conceptions de circuits de broyage d’or, notamment les procédés de charbon en pulpe (CIP, de l’anglais carbon-in-pulp), de charbon en colonnes (CIC, de l’anglais carbon-in-column) et de lixiviation au carbone, ou charbon au lixiviat (CIL, de l’anglais carbon-in-leach). Toutefois, les caractéristiques et la durabilité du charbon actif varient. La manipulation de ce milieu pose des problèmes dès lors que se produisent des pertes de carbone à grain fin dans la décharge des circuits CIP, CIC et CIL. Par ailleurs, l’or adsorbé sur les fines de charbon peut se perdre durant la décharge. Les études réalisées dans le cadre de cet article visent à normaliser les méthodes de laboratoire pour tester les qualités du charbon actif spécifiques au secteur de l’exploitation aurifère. L’étude portait sur la composition granulométrique, la teneur des plaquettes, la résistance au frottement humide, la cinétique de l’or et l’adsorption de l’or, à savoir la capacité du charbon actif à adsorber l’or sur une période spécifique. Les résultats ont montré qu’un test de frottement de 96 heures et un test d’adsorption de l’or constituaient le fondement d’un classement des échantillons de carbone du point de vue économique en vue de trouver l’échantillon générant le meilleur rendement des usines. ¦ MOTS CLÉS charbon actif, dureté de la batée des billes de carbone, évaluation du carbone, perte de fines de charbon, renouvellement du carbone, résistance au carbone INTRODUCTION The use of activated carbon for gold extraction is among the most common processes in modern gold-milling plants. Although gold losses are inevitable in the milling process, operators often focus on reducing gold loss to tailings. One focus area that is important in gold recovery is optimizing the use of activated carbon and examining potential changes to realize improved economics for carbon-in-pulp (CIP), carbon-in-column (CIC), and carbon-in-leach (CIL) processes. Granular activated carbon (Figure 1) is widely used for gold extraction from slurries containing gold-cyanide complexes. This activated carbon—produced from charred coconut shells—is well suited because of its adsorption kinetics, gold-loading capacity, attrition resistance, and ability to be reactivated. The extraction process using activated carbon technology involves adsorbing gold from an aqueous leachate; the gold is dissolved using a cyanide salt and consequently a dicyanoaurate(I) ion (Au(CN)2–) is formed, as noted in equation 1."

Jan 1, 2019

By Alessandro Navarra, Amir Arash Rafiei, Kristian Waters

Mathematical programming (MP) is fundamental to systems engineering and operations management, but has not been widely adopted by mineral processors. The current paper presents an interdisciplinary approach, using MP to incorporate basic notions of mass balancing into an optimization framework. The optimization is typically related to economic profitability. Two- and three-product formulas are extended to give formal descriptions of alternate modes of operation, and a global representation of the concentrator. At this global level, the framework may assist plant managers to make tactical decisions based on incoming feed data. The MP framework is demonstrated through sample calculations describing a lead-zinc concentrator.

Jan 1, 2015

By G. Shore

"The development of a DC resistivity instrumentation system for data acquisition in extreme topography yielded an unexpected bonus in the quality and information content of the type of raw data gathered. Not only were traditionally too-rough areas being investigated systematically for the first time, the unique high-density, multi-directional raw data were providing the opportunity to greatly expand the range of conductive and resistive signatures that could be accurately recognized. Using a true 3-D inversion algorithm, 3-D earth resistivity imagery began to show distinctive anomaly patterns for hydrothermal mineralization at the seldom-visited resistive end of the spectrum, successfully imaging epithermal gold systems in conditions that had defied conventional ground and airborne geophysics for years.The traditional mode of designing a resistivity survey to test a specific target concept was shown to be an unnecessary limitation on exploration thinking when a single sweep acquisition of dense, multi-directional 3-D raw data could effectively provide the all-targets “shotgun” coverage that would take 10 or 12 conventional surveys to duplicate. Exploration planning has the opportunity to move from single steps of trial and error geophysical testing to a mode of mapping everything at the outset, and then sorting out what might be of exploration interest in the results. Survey parameterization becomes the non-technical process of marking an outline on a map and declaring “survey here.” Operational and interpretational reasons for not testing where the exploration interest lies have all but disappeared. With any-terrain capabilities and unprecedented raw data density and overlap to support interpretation of complex conditions, that outline can be drawn through extreme rough terrain, including lakes, cliffs, industrial facilities, and grounded installations to test for generally unknown sub-surface features of potential interest.As more is learned about the larger-scale signatures of ore settings and the district-scale geo-electric imprint of hydrothermal processes on lithology, it appears that the only thing certain about ore signatures is their uncertainty. This is not bad news. New opportunities abound as the advanced 3-D technology allows a first effective look at the previously unexplorable rough terrain or complicated settings, and a second look at areas where older results may now be known to be non-definitive."

Jan 1, 2002

By Dave Ecobichon, Éric Côté, Simon Nickson, Michel Leclerc

The feasibility assessment of a mineral deposit requires the construction of a geo mechanical model of rock mass properties, for the purpose of preliminary mine design. The Noranda Technology Centre has been involved in the development of geomechanical models for several mineral deposits at the feasibility stage. Each model is constructed through intact rock testing, geotechnical core logging, geotechnical mapping and rock mass classification. Rock mechanics principles are used to develop conceptual stope designs and propose possible mining sequences. This methodology is applicable to both new mining projects and to the development of new areas within anoperating mine. This paper explains how the geomechanical model is developed and reviews its application to several case studies.

Jan 1, 1996

By Daniel Hodouin, Bruno Girard, Claude Bazin

"Mineral processing and smelting operations have the reputation of being complex processes. Process operators of these unit operations have acquired from experience and basic training a superficial understanding of the relationships between process output and manipulated variables. This understanding, which is a very valuable know-how for process control and optimization as well as for operator training, is however difficult to extract. This paper describes a time series based approach that can be used to extract this know-how, which can subsequently be used for operator training. The method is illustrated with an example taken from a smelting process.IntroductionThe emergence of Total Quality Management and Statistical Process Control philosophies in the North American industry (Brown, 1991 ), as well as the increasing popularity of the IS0- 900x standard accreditation, provide a strong incentive to review the training procedures for plant operators. The importance of employees training is acknowledged in two of the 14 points of Deming's road to quality (Deming, 1982) as well as by control engineering consultants (Hendricks et al., 1994). However guidelines for the operation of industrial processes are often transferred from an operator to another one with allowance for flexible interpretation of many operating rules. Such transfer is typical of complex industrial processes to which particulate processes of ore dressing and extractive metallurgy belong.Capturing the knowledge of experienced operators is often a difficult task. Cognition engineers are able to extract this knowledge and transfer it into rules that ultimately are implemented into expert systems. Cognition engineers are however not always available and plant supervisors may prefer to keep training as an internal liability. Another difficulty through interviewing of the process operator is its lack of willingness to transfer what he has learned by hard working to other people, whose ultimate goals are to develop computerized applications that may replace him in a near future, or reduce its importance in the production line.This paper shows that time series can be advantageously used to capture the average process understanding of plant operators and to help reviewing training procedures. The method s illustrated with data collected during normal operation of electric arc furnaces and it is shown how the approach has led to a fruitful discussion between process engineers and operators that ultimately provided savings of more that $200,000 per year per furnace. There are 9 furnaces in the studied foundry. Such savings are not the results of application of advanced control strategies nor of ""miracle press button software», but of an improve awareness by engineers and operators of the importance of properly defining the operating procedures to control a process."

Jan 1, 1996

By B. Dubreuil, J. -M. Lalancette, D. Lemieux, C. Chouinard

The extraction of gold and silver has been examined via the use of mixed halogens: chlorine and bromine. The reaction was performed in a brine of sodium chloride with a small amount of sodium bromide at 40-50°C, in an atmosphere of chlorine, at atmospheric pressure. The process takes advantage of the very high reactivity of bromine towards gold, while the higher oxidation capacity of chlorine leads to gold chloride as end product with recycling of bromine. Fast reaction rates are observed along with gold recovery at the level of 95%. Chlorine-consuming common elements are removed prior to chlorination.

Jan 1, 2011

By K. G. Chipman

THE discussion on Dr. Chipman's paper, Mapping by the Topographical Survey, centred mainly on the urgent need for aerial photographic reconnaissance of unmapped portions of the country as an initial phase in the provision of suitable topographical maps for use as a guide in geological mapping, prospecting, exploration, and development work in the areas concerned. This, ir was pointed out, would necessitate an expansion of the topographical work of the Bureau of Topography to keep pace with the photographid work and to meet the demand for maps for use by geologists of the Bureau of Geology and others. A resolution requesting Institute Council to make representations to the proper authorities, urging that aerial photography and topographic map-ping be greatly expanded to meet the pressing needs of the situation, was moved by DR. W. F. JAMES and seconded by MR. L. G. SMITH, and carried unanimously. The full text of the resolution appeared on page 221 of the May BULLETIN. Others who spoke in support of the resolution were MESSRS. E. S. MOORE, L. B. ROCHESTER, F. D. SHEPHERD, and W. J. WRIGHT. Dr. Hansen then presented his paper, Mapping by the Geological Survey, discussion on which is summarized below. MR. FRANK EBBUTT(l) spoke of the very valuable service the Geological Survey bas rendered the mineral industry, and indeed the whole nation, in its century of existence, despite the very inadequate financial support it has received. Now, more than ever before, a strong and expanded Geological Survey is essential if the Industry is to continue to prosper. Our known ore reserves are being depleted very rapidly, and we are finding no new orebodies to take their place. As an aid in prospecting and exploration, and also in the development of orebodies when they have been found, we must have geological maps of the vast sections of the country which at present are unmapped or inadequately mapped. The Institute should make every endeavour to have the Geological Survey strengthened so that such maps may be made available as soon as possible.

Jan 1, 1944

By Larry J. Hulbert, Louis J. Cabri, John L. Campbell, Gilles Laflamme, Chris G. Ryan, Soey H. Sie, Rolando Lastra

"Abstract- The mineralogy of samples from the Wellgreen Cu-Ni-Pt-Pd deposit, Yukon Territory, is complex and the metals of interest are contained within fme-grained minerais which vary in size and mineral associations from sample to sample.The major Ni carriers are pentlandite and violarite; the major Cu carrier is chalcopyrite. About 700Jo of the major Ni carriers and 78% of the major Cu qmier will be liberated at a grind of 80% -'}f,.7 J.Uil (-500 mesh) for typical low-grade gabbro (West zone, sample No. 1). For typical ""clinopyroxenite"" (West zone, sample No. 2) and typical ""clinopyroxene-rich peridotite"" (East zone, sample No. 3) the liberation would be 71""lo and 65% for Ni, 74% and 60% for Cu, respectively. Fine grinding, however, cannot liberate the Ni contained in silicates and pyrrhotite, estimated to total24.5% for typicallow-grade gabbro (West zone, sample No. 1), 15% for typical ""clinopyroxenite"" (West zone, sample No. 2), and 12% for typical ""clinopyroxene-rich peridotite"" (East zone, sample No. 3).The principal Pt carrier (sperrylite) varies considerably in grain-size between samples and fme grinding, required to liberate the major Ni and Cu carriers, may be counterproductive in samples where sperrylite is relatively coarse-grained, as sorne would then be unrecovered (e.g. sample No. 1). In addition, sperrylite in samples No. 2 and No. 3 is closely associated with magnetite.Palladium occurs in diverse ways. It occurs in the form of several Pd minerais (e.g. sudburyite, testibiopalladite, merenskyite), as a replacement for Ni in several tellurides, antimonides and tel­ lurantimonides (e.g. melonite), or as very low concentrations in pentlandite (up to 34 ppm), chal­ copyrite (up to 9 ppm), and pyrrhotite (up to 5.6 ppm). Palladium minerais and Pd-bearing minerals are strongly associated with chalcopyrite in sample No. 1 and will thus report to a Cu concentrate. However, 22% of the total Pd is in solid solution in pentlandite in this sample and will thus report to the Ni concentrate. In samples No. 2 and No. 3, less Pd occurs in solid solution in pentlandite, and the PGM are distributed mainly among pentlandite, chalcopyrite, and magnetite in sample No. 2, and among pentlandite, pyrrhotite, chalcopyrite, and magnetite in sample No. 3. Therefore, the feasibility of magnetic concentration should be investigated for samples like No. 2 and No. 3, in order to improve Pt and Pd recoveries.Rhodium was found to occur as rare hollingworthite, in sparse cobaltite (up to 2.7%) and, as low concentrations in chalcopyrite (to 10 ppm) and pentlandite (to 12 ppm) by proton microprobe. Iridium occurs in hollingworthite (up to 7.8%) and very rarely as the native metal or in an undefmed Re-Ir-Os-Ru alloy. Ruthenium was determined by proton microprobe analyses to occur in pentlandite (up to 13 ppm) and chalcopyrite (up to 10 ppm), in addition to rare laurite and Re alloy."

Jan 1, 1993

By T. R. Kostecki, J. Hirschi

"Analytical engineering design is based upon a trial-and-error iterative and deterministic process. Within this process, the engineer obtains some estimated values, plugs them into de facto closed-form equations, and receives output, which is expected to be a single number that serves as the basis for the design. This number is typically a factor of safety, or some other equivalent strength-stress ratio. This provides the engineer with a quick and relatively quantitative design methodology. This can create problems however, because in mining applications, the actual in situ system is highly variable, complex and often chaotic, which can lead to potentially incorrect conclusions that result in unsafe designs. A more appropriate and reliable approach is a probabilistic analysis for engineering design. This process is used widely in civil and other engineering disciplines, but is often overlooked for applications in coal mine design. Additionally, it seems the amount of past studies using this approach, especially in coal mine design, are rather limited. Considering how the probabilistic approach can account for uncertainty in parametric values and how unpredictable a mining design can be, the authors believe that this approach has potential in mining. This paper will focus on a summary of past studies related to probabilistic analysis in coal mine pillar design and provide recommendations for future work that could improve design reliability. INTRODUCTIONFeasibility and fundamental design procedures for excavations often start with an assessment of mine openings, pillar design, and the surrounding structure by using a deterministic approach with closedform solutions (ex. Bienawski, Obert-Duvall). These are reliable methods, but designers know that the inputs in these closed-form solutions are estimates at best and do not incorporate any degree of uncertainty. This is therefore not an optimal design procedure.A more appropriate design should incorporate some uncertainty because of the chaotic and variable nature of geology and mining conditions in general. In an attempt to account for uncertainty, a probabilistic approach can be adopted, which accounts for deviations in variables and provides a more optimum design solution. By applying a probabilistic approach, a designer can look at multiple failure scenarios at the same time instead of using the same deterministic design over and over."

Jan 1, 2015

By David Ulrich Furrer

Materials and manufacturing processing technologies have played an important role in the development and capabilities of nearly all manmade products. Market drivers for fuel efficient, low cost, environmentally friendly, more capable transportation has given rise to significant materials challenges and subsequent developments. The rate of improvement and implementation of new commercial products can be limited by the rate of new materials and process technology development and qualification. As many materials are being designed for unique requirements and even location-specific unique properties, the implementation of tools and methods to enable rapid characterization and validation of material definitions and design systems must also be accelerated. A review of new integrated material, manufacturing process and component designs, engineering processes and emerging tools and methods will be discussed. The linkage of computational materials science and engineering into component design will also be discussed along with the challenges and opportunities for further growth and application of integrated computational materials engineering (ICME).

Jan 1, 2015

By G. Danha, D. Glasser, D. Hildebrandt, N. Hlabangana

"In this work, we apply a method known as the attainable region (AR) technique in order to optimize the milling and leaching process of a low grade gold ore. No work on the application of the AR optimization technique has been done yet on a joint milling and leaching process. The advantage of the AR approach lies in its ability to simplify the optimization problem, as searching over a defined set for the maximum of an objective function is a fairly standard procedure. The objective function we selected was that of optimizing a linear function of the value of the recovered material minus the cost of both milling and leaching. Using the three variables (milling time, leaching time and recovery), we constructed a 3D plot and used it to obtain all the possible recoveries from the different milling and leaching times. The optimum for our chosen objective was then found by overlaying a contour plot of the objective function on the 3D plot. Our results show that the optimum was obtained at 90 % recovery with a profit value of 600, milling time of 30 minutes and leaching time of 1750 minutes.INTRODUCTIONAfrican countries are facing a general increase in demand for electrical power. In South Africa, the demand is now surpassing the supply from ESKOM, the main electric power provider of the country. This has resulted in load shedding strategies being implemented in many local communities in an attempt by ESKOM to cope with the high demand. A potentially better approach is to propose use of energy efficient technologies in the country’s largest power consuming industry; the mineral processing sector.In mineral processing, milling is the highest consumer of energy. The objective of carrying out milling is not merely to reduce the particle size of the feed material, but also to liberate the constituent minerals that make up an ore. This is done so that valuable minerals can be separated from the gangue, in downstream processes, such as leaching. However, the milling process is generally performed relatively poorly, and at a considerable expense in terms of electrical power utilization. Tomanec and Milovanovic (1994) estimated that milling accounts for more than 50 % of the total power used in the concentration process, but this can rise to as high as 70 % for hard or finely dispersed and inter-grown ores. Apart from being relatively wasteful of energy, the milling process is also inefficient with regard to mineral liberation because of the indiscriminate nature of the grinding force."

Jan 1, 2016

By J. A. Bos, Z. Dai, P. Quinn, A. Lee

A flowsheet is being developed for processing Thompson ultramafic low-grade nickel ores. This type of ore is characterized by high content of serpentinized magnesium silicates which generate slimes during comminution. Flotation of this type of ore proves difficult due to problems associated with the slimes. The flowsheet, based on extensive laboratory and miniplant flotation tests, consists of two key components: desliming by hydrocyclones prior to flotation and using soda ash for pH control in flotation. Satisfactory metallurgical results have consistently been achieved in laboratory full circuit simulation and miniplant tests.

Jan 1, 2009

By K. Niboucha, M. Zergoug, S. Brick-Chaouch, H. Houali

"Corrosion is an electrochemical process that causes degradation and deterioration of various steel works (pipelines, drains, cisterns) whether they are buried, immersed, or not. Worldwide, we estimate that corrosion destroys a quarter of the annual production of steel at a rate of 5 tons per second. Several techniques are used to protect against corrosion. Among these processes, Cathodic protection, which is regarded as the best technique to assure the protection of steel, works against this phenomenon. This is achieved either by a system of reactive anodes, or by the technique of impressed current protection or other techniques. Before any installation of cathodic protection, a preliminary study and pre-evaluation of certain parameters must be made. In the present work, we will present an example of dimensioning of an installation of cathodic protection by imposed current of a pipeline using solar energy, where we discuss the following items:•,Evolutions of the potential as well as current along the pipeline and their respective expressions deduced from the laws known as telegraphists;•,Range of protection (potential of protection along the pipeline);•,Influence of the isolation resistance of the coating on the range of protection;•,Effect of the aging of the coating and its influence on the current of protection."

Jan 1, 2005

By Saravarim A.

"The aim in flotation systems is to control the grade and recovery of the product., numerical method or sampling and analysis is used to achieve this goal. Weakness of measuring instruments, lack of process knowledge, deficiencies in management and data processing in the control of flotation cells, has led the academic and industrial units to use numerical models. In this regard, using the results obtained from the experimental design carried out in the DTP line of Gol Gohar industrial and mining complex, a regression model was developed based on the analysis of variance. To present this model, after variance analysis of factors and interactions, all models were investigated and the final model was created based on 75% of the data. The model was tested with the remaining 25% of the data to confirm the goodness of the fit. The coefficient of determination for training and testing resulted 71% and 74%. Moreover, a feed forward back propagation neural network with different layers and neuron was used to model the results. The best network structure was 6-27-1 and the coefficient of determination for this network was 99% and 98% for training and test respectively. Given the complexity of neural network models and the relatively low accuracy of the regression model, neural network has a priority assuming we have access to a specialist. In addition, the results obtained from neural network can be used as expert systems to control and optimize the flotation cells.INTRODUCTIONThe primary objective of the flotation cell control is to maintain the product's (in this case tailing) recovery and grade, which represents the index for process efficiency and product quality. Continuous estimation of these indexes requires a significant amount of maintenance, and calibration of the continuous analyzers, in order to maintain an acceptable level of accuracy. Weakness of measurement equipment, lack of process knowledge and the general deficiency in management and processing of data in the flotation cell control, will lead industrial units to use smart methods such as mathematical models, regression and neural networks in near future (Panahi, Abdollahzadeh, & Sam, 2010). Artificial neural networks are inspired by the very complex structure of human brain in which millions of nerve cells solve problems or store information by communicating with each other. The duty of neural networks learning. It should be noted that learning in artificial neural networks is limited, and what is practically considered is the computing power of the network. A neural network consists of units called neurons, and has the capacity to handle a set of input data to produce a desired output data set. Each of these input and output data sets can be assumed as a vector (Soltanmohammadi & Noparast, 2006)."

Jan 1, 2016