Search Documents

By M. P. Sinha, Rashmi, Tata Steel Ltd., S. Jaiswal, A. Bhatnagar, P. Garg

Reagents form a crucial part of flotation process as they have the power to control the surface properties of minerals. In case of coal flotation, collectors and frothers are two the important and commonly used reagents. But other types of reagents such as promoters, modifiers and depressants are also finding wide applications because of their availability to improve selectivity and increase recovery. Because of these chemicals the scenario has changed from flotation being used mostly for high rank hydrophobic coals to low rank, oxidised or inferior coals. In the present work a case is presented in which two non-ionic surfactants (abbreviated as A & B) known to act as emulsifier have been studied to get a better flotation performance for a preparation plant with varying feed characteristics. Laboratory flotation tests were conducted with both the surfactants using three feed samples of the plant by varying the dosage at three different levels. The highest combustible matter recovery (more than 80%) was obtained for surfactant B at a dosage of 75 ppm (parts per million) for feed 1. But at the same dosage the recovery decreased by 21% and 10.2 % respectively for feed 2 and 3. On the other hand surfactant A showed a consistent result for all the three feed samples giving the highest value at a dosage of 50 ppm followed by a decrease beyond this point. The tests were complemented by contact angle measurements that demonstrated the same trend. In addition to this the reproducibility of data for Surfactant A was checked and found that the standard deviations for both separation efficiency and combustible matter recovery were within the permissible range (less than 5%). Finally an in-plant process optimization study to evaluate the performance and applicability of surfactant A at industrial level was done and has been discussed.

Jan 1, 2016

By R. Pandher, R. Schonewille, N. Tripathi

Advanced process modelling tools were developed for the Xstrata Nickel Sudbury Smelter flow sheet using a combination of FACTSAGE TM METSIMTM and ARENATM software. These tools were employed to assist with plant optimisation, emission control, process design, troubleshooting and case studies as applied to the smelter. The development of process modelling tools in pyro-metallurgy commonly involves an in-depth study of the therm o-chemical properties (heat capacity, liquidus temperature, etc.) of all high temperature phases, usually matte and slag. The preferred tool for this purpose is FACTSAGETM, a well-known then-no-chemical modelling software package. The results from the therm o-chemical model have been subsequently incorporated into the development of a process model for the entire smelting facility using the METSIMTM platform. Such process model also includes a number of special features adopted to effectively simulate the complex industrial phases; matte and slag systems. To ensure the robustness and reliability of the process model, several process parameters and assumptions were verified with laboratory and plant data. Consequently, the model includes the ability to examine the impact of a number of plant parameters on potential plant performance. Primary variables are the percent sulphur elimination during concentrate roasting, furnace matte grade and degree of matte metallization. Key output parameters include operating temperature, furnace coke requirement, sulphuric acid production, CO2 emissions and overall energy requirements. Selected outputs from the process model (for example blowing time, slag production rate and matte converting rate) can then be utilised in the development of a discrete event model (DEM) for the smelter using ARENATM software. Incorporation of reliable and validated mass balance parameters enhances the robustness and reliability of the DEM model. The integration of then-no-chemical, process and DEM modeling tools has a distinct advantage in planning plant trials and examining process design initiatives at Xstrata Nickel's Sudbury Smelter. The present paper discusses the development of these three models and the synergies existing among them. Typical results from all three models are also presented.

Jan 1, 2011

By R. Pandher, R. Schonewille

Advanced process modelling tools were developed for the Xstrata Nickel Sudbury Smelter flow sheet using a combination of FACTSAGETH METSIMTM and ARENATM software. These tools were employed to assist with plant optimisation, emission control, process design, troubleshooting and case studies as applied to the smelter. The development of process modelling tools in pyro-metallurgy commonly involves an in-depth study of the therm o-chemical properties (heat capacity, liquidus temperature, etc.) of all high temperature phases, usually matte and slag. The preferred tool for this purpose is FACTSAGETM, a well-known then-no-chemical modelling software package. The results from the therm o-chemical model have been subsequently incorporated into the development of a process model for the entire smelting facility using the METSIMTM platform. Such process model also includes a number of special features adopted to effectively simulate the complex industrial phases; matte and slag systems. To ensure the robustness and reliability of the process model, several process parameters and assumptions were verified with laboratory and plant data. Consequently, the model includes the ability to examine the impact of a number of plant parameters on potential plant performance. Primary variables are the percent sulphur elimination during concentrate roasting, furnace matte grade and degree of matte metallization. Key output parameters include operating temperature, furnace coke requirement, sulphuric acid production, CO2 emissions and overall energy requirements. Selected outputs from the process model (for example blowing time, slag production rate and matte converting rate) can then be utilised in the development of a discrete event model (DEM) for the smelter using ARENATM software. Incorporation of reliable and validated mass balance parameters enhances the robustness and reliability of the DEM model. The integration of then-no-chemical, process and DEM modeling tools has a distinct advantage in planning plant trials and examining process design initiatives at Xstrata Nickel's Sudbury Smelter. The present paper discusses the development of these three models and the synergies existing among them. Typical results from all three models are also presented.

Jan 1, 2011

RISKGATE is an interactive online risk management system developed by the Minerals Industry Safety and Health Centre (MISHC; University of Queensland) to maximise health and safety performance in the mining industry through implementing continual improvement in management of major unwanted events. A comprehensive knowledge base has been captured for risk management of tyres, collisions, fires, isolation, strata underground, ground control open cut, explosions, explosives, manual tasks and slips/trips/falls. This will be extended to outburst, coal burst and bumps, interface displays and controls, tailings dams and inrush in 2013. RISKGATE presents substantive and leading edge controls to assist industry stakeholders in the design, management and reporting of organisational and regulatory compliance requirements. From a broader industry perspective, RISKGATE provides an environment for knowledge capture and knowledge exchange to drive innovation and best practice in the identification, assessment and management of risk. By capturing operational knowledge from industry experts, RISKGATE provides a cumulative corporate memory at a time of high personnel turnover in the coal industry. All major Australian mining companies are contributing topic experts to this broad industry initiative. In the first 24 months, nine coal mining companies nominated over 80 different experts who contributed over 400 days of individual time to the knowledge capture workshops. RISKGATE is the largest single Australian Coal Association Research Program (ACARP) occupational health and safety (OHS) initiative to date.

Aug 1, 2013

By C. Hunt, J. Vivas, D. M. Allen

"In this paper we present the results of a combined remote sensing - numerical modelling study to improve our understanding of groundwater seepage in open pit and natural rock slopes, and to investigate the interaction between groundwater, in-situ stress conditions and brittle fracture. The location of seepage is a key element in the characterisation of rock slope stability in open pits and natural slopes; definition of the fracture network geometry and its inter-relationship with seepage, however, remains a challenge in engineering practice. Current techniques used to characterize groundwater seepage locations require direct access to the rock slope for a detailed and exhaustive investigation, and can be severely limited due to access and the limited coverage of the survey methods both with respect to the height and length of the sampled rock slope. Herein, we describe remote sensing methods, which allow the collection of structural and seepage information over a wide spatial area. Remote sensing techniques are shown to have the potential to provide additional insights into the orientation and general connectivity of fractures that may play a major role in the general stability of a rock slope. A preliminary methodology for the use of remote sensing in the characterization of seepage zones in a natural slope through the application of ground-based LiDAR and Infrared Thermography (IRT) is presented.One of the most challenging problems in rock slope engineering is the realistic simulation of fluid flow through fractured and deforming rock slopes. Deformation and the development of brittle fractures can cause significant local changes in the permeability and strength of the rock mass, potentially affecting rock slope performance. Understanding fluid flow through fractured and deforming rock slopes, especially in highly-stressed environments is of great importance for slope stability analysis. Changes in the interconnectivity of fractures during slope failure are an important, yet are relatively unstudied in rock slope engineering. We present the results of hydromechanical UDEC-Trigon rock slope models focussing on the potential role of brittle fracture in rock slope failure, its interaction with groundwater pressures, and the influence of the assumed in-situ stress ratio. Damage intensity parameters are used to characterize damage within different regions of a biplanar slope failure."

Jan 1, 2015

By A. I. Collins, D. Danninger

Newmont has installed advanced control and online optimization on the dry grinding circuit and both roasters on its Carlin Trend refractory ore treatment plant. Subsequently, Newmont has developed a master plan for institutionalizing advanced control and optimization on all its processing units. Newmont is committed to a program of applying advanced control and optimization from crushing of ore to pouring of gold. This paper describes Newmont's experience with implementing advanced control and optimization and the benefits received. In addition, identification of advanced control and optimization opportunities and their economic justification is discussed.

Jan 1, 2005

By Thom Mambe, Angelo Fernando

The objective of this work was to identify key parameters related to the reduction of the sulphur content in ammonium diuranate (ADU), and the identification of solid products being formed under various operating conditions. A laboratory-scale, continuous ADU precipitation circuit was assembled using a glass reactor vessel immersed in a temperature-controlled water bath. A pre-prepared uranium feed solution was fed continuously using a peristaltic pump. Anhydrous ammonia mixed with compressed air was metered and pumped using a double-headed peristaltic pump. An over-head mixer was used to keep the solids in suspension. The precipitated ADU slurry was removed from the reactor at the same flow rate as the feed solution. The slurry was sampled to generate ADU solids for chemical and physical analysis after four hours of continuous operation to ensure steady-state conditions. The following variables were studied to determine their influence on the sulphur content in the resulting ADU. (a) sulphate:uranium molar ratio in the reactor (1:1, 1:5, 1:10 and 1:20) (b) precipitation pH ( pH 6, 6.5, 6.9, 7.5 and 9) and (c) precipitation temperature (40C and 60C) A total of 18 tests covering the above parameters were conducted, and the results indicated that the sulphur content in the resulting solids was highly dependent on the reaction pH. Out of the pH levels tested, tests completed at pH 6.9 and 7.5 generated solids containing the lowest sulphur contents, between 0.5 to 1.5% S uranium basis (UB). The sulphur content in solids was found to vary linearly with the sulphate:uranium ratio in the feed solution, showing a statistically valid correlation. The impact of temperature was minor between 40°C and 60°C, with a slight advantage at higher temperatures. Tests at a sulphate:uranium ratio of 1:1 produced solids with the lowest sulphur contents. The influence of pH or temperature on sulphur content in solids was almost non-existent at this sulphate:uranium ratio. Precipitation pH is the major factor influencing ADU sulphur content, followed by the uranium to sulphate ratio.

Jan 1, 2020

By R. W. Boyle

"PREFACEThis paper comprises two parts. The first part discusses lithochemical surveys based on major, minor and trace elements in whole-rock samples from primary enveloping halos. The second part continues with discussions of surveys based on leakage halos, zonation of elements, stable and radioactive isotopic distributions, geothermometric, geobarometric, thermolumin escent, pedochemical, hydrochemical, atmochemical and biogeochemical methods. A selected bibliography follows the second part of the paper.ABSTRACTAreal (two-dimensional) geochemical surveys, based mainly on secondary dispersion halos and trains in media such as surface waters, soils and drainage sediments, have proven particularly successful in recent years and have greatly assisted in the discovery of a large number of varied mineral deposits. Three-dimensional surveys based on primary halos and leakage halos have received relatively little attention by research organizations and the mineral industry. Such surveys offer many novel approaches for discovering blind mineral deposits deeply buried below overburden or within their host bedrocks. This paper reviews detailed methods for the discovery of blind mineral deposits based on sampling rocks (lithochemical methods), soils and other types of overburden (pedochemical methods), waters (hydrochemical methods), gases (atmochemical methods) and biological materials (biogeochemical methods) . All are applicable to the discovery of blind deposits; the efficacy of each or in combination depends essentially on the type of mineralization, the extent of the development of the primary and secondary dispersion halos, and the finances available for sampling techniques employing overburden drilling, bedrock drilling, and sophisticated water, gas and vegetation collection."

Jan 1, 1982

By T. Banerjee, Y. Dumais, R. Lemire

"Agnico Eagle LaRonde produces over 300,000 ounces of gold annually, along with silver, zinc and copper by-products. The LaRonde plant consists of a conventional SAG and Ball mill circuit, followed by flotation. Load management is imperative for maintaining stability in grind size and ensuring consistency of the flotation feed. In flotation, the mass pull profiles of the cells and reagent dosage rates need to be managed to maximize recovery, while maintaining product quality. This is achieved by controlling the air, level and reagent addition set-points across the circuit. However, disturbances such as feed size distribution, ore hardness and mineralogy make it challenging to maintain efficient control using traditional control techniques. LaRonde therefore worked to develop a progressive control strategy, targeting stability and optimization. This started with a year-long instrumentation upgrade campaign, focusing on optimizing the performance and reliability of plant-wide instrumentation. The process was then optimized, leveraging AwaRE advanced controllers running in closed loop with PARC modules embedded within the PLC. The installation of 8 PFC cameras in flotation changed the flotation operation from a level and air control based strategy, to one that controls mass pull. This new control philosophy has allowed the plant to effectively manage mill load, improving grind size and providing increased throughput with better flotation performance.INTRODUCTION Background Since 2015, Agnico Eagle LaRonde has been conducting an optimization campaign to maximise the performance of plant assets leveraging instrumentation and control. The grinding and copper flotation circuits were identified as the areas having the greatest opportunities and value proposition for advanced control. A disciplined approach was followed to design and implement a progressive control solution to achieve operational excellence."

Jan 1, 2018

By C. S. Parsons

Introductory The steady advance in the price of lead and zinc in the metal markets of the world during the past few years has resulted in an active search for new deposits of these metals, and to the re-examination of known properties that are now idle. There are a number of deposits of these metals in eastern Canada, the best known of which are the Tetreault mine at Notre Dame des Anges, Que.; the Reader mine on Calumet island, Que.; the Stirling property in Cape Breton, and the Federal mine in Gaspe district, Que. . The author has run concentration tests on the ores from the first three mentioned properties. The ores from Notre Dame des Anges and Calumet island presented problems in concentration which were not successfully solved until the past year. Although considerable experimental work has been done by the Ore Testing Laboratories on the ore from the Stirling mine we have not been successful in developing a satisfactory method of treatment. Character and Types of Ores Metallurgically the lead-zinc ores of eastern Canada may be classified into three distinct types as follows: First: - Ore in which the sulphide minerals are fairly coarsely crystalline and can be freed by crushing to 40 or 60 mesh. The zinc blend has a high iron content (not marmatite) and gives a dark brown colour on grinding. The ores of Notre Dame des Anges and Calumet island, Que., may be cited as examples.

Jan 1, 1925

By C. O. Swanson

THE purpose of this paper is to discuss the use of the dip needle in the field of mining geology. The subject matter is new only to the extent that it reflects the writer' s added experience in recent years. The treatment is brief, and those interested in a more extended analysis are referred to the articles listed at the end of the paper. In the discovery and tracing of mineral deposits, the clip needle, like most other geophysical instruments, can be used in two ways. One method may be termed direct; the other, indirect. The direct method is that of finding or following certain ore-bodies by virtue of their magnetic properties. Explicitly, or by inference, it is the method which many mining men and practising geophysicists have in mind when they consider whether or not a certain property should be studied by geophysical means. In the ideal case, this method demands little or no knowledge of geological conditions. The indirect method uses the clip needle as an aid to purely geological studies in working out the structure of a mineralized area. Although not as simple as the direct method, it enjoys a much wider field of usefulness. The writer's experience in the past fifteen years, which is believed to be typical, is that the indirect method finds application four or five times as often as the direct method. The indirect method is of particular value because it furnishes information concerning the relationship between geological structure and ore occurrence. Most mining engineers will agree that structural control is an important factor in the localization of ore deposits, and therefore that the subject under consideration is of some interest in mining geology. Specifically, then, the present purpose is to outline the principles and practice of using the clip needle in working? out geological structure.

Jan 1, 1941

By P Hulteberg

In this article, an integrated catalytic reactor design approach to produce PEMFC-quality hydrogen is evaluated. The design is based on a high level of heat exchanging perfonned internally, highly active noble metal catalysts and a woven wire mesh catalyst support system. It consists of an integrated steam refonner and a catalytic burner section, a water gas shift section and a preferential carbon monoxide oxidation section. The reactor design is described and experiments perfonned to evaluate the two reactors built are reported. The two reactors have a maximum output of 3 Nm and 10 Nm3 hydrogen/h respectively. In the larger unit, a total of six fuels, three gaseous (methane, natural gas and liquid propane gas) and three liquid (naphtha, ethanol and methanol), were evaluated. The larger unit was also used to investigate how variations of parameters such as steam-to-carbon ratio in the steam reforming part of the system and the relation between oxygen and carbon monoxide in the preferential carbon monoxide oxidation affect the reactor perfonnance. The smaller unit was used, after implementing automatic control, to run a system life time test for 480 h. The refonner system perfonned according to specifications on almost every aspect.

Jan 1, 2005

By Rodrigo Serna-Guerrero, Ted Nuorivaara

"As the demand of raw materials increases in a global scale due to demographic and economical changes, it is necessary to find technological solutions that improve current production processes in an efficient and sustainable manner. The main objective of this preliminary study is to evaluate the properties of foam produced with a cellulose-based molecule as an alterative to commercial frothing agents in flotation operations. There are two potential advantages identified with the use of cellulose-based additives for the proposed application: i) these chemicals can be produced from sustainable sources (i.e., non-edible crops), and ii) cellulose can be functionalized with chemical groups, in order to rationally modify its behaviour. In previous studies dealing with foam stabilization, it has been proposed that some functionalized cellulose species present a self-assembly mechanism at gas-liquid interfaces. Unlike typical surfactant-based frothers, cellulose molecules form networks around the bubbles that act as protective layers to the mechanical stress and to the occurrence of ionic species. This work studied the properties of foams produced in the presence of a partially hydroxylated cellulose molecule (i.e., hydroxypropyl methyl cellulose; HPMC) under various concentrations in a laboratory-scale column. A comparison with a commercial surfactant frother (i.e., Nasfroth 240) is presented, and a rationalization of the behaviour of HPMC in terms of foam properties (i.e., surface tension, bubble size, foam depth) is proposed. INTRODUCTIONThe search for green alternatives to chemicals produced from non-renewable sources has become a topic of major relevance in industry (Choi et al., 2015). As a chemical consuming industry, the mining sector faces now the challenge of adopting new processing technologies and strategies in order to achieve the future demands of sustainability. Evidently, a move towards green chemistry can only take place by identifying substitute molecules offering potential advantages in terms of recovery and grade."

Jan 1, 2016

By B. Voyzelle

The deterioration of highway and bridge infrastructure has received considerable attention in recent years because of the billions of dollars needed for repair and rehabilitation. A major factor contributing to this problem is the corrosion of steel reinforcement. Reinforcing steels with good corrosion resistance and relatively low price are therefore essential for maintenance, cost reduction and service-life extension of the steel-reinforced structures. This article describes the development of low-alloying steels that offers good corrosion resistance with reasonably low cost. Topics include steel microstructures, mechanical properties, corrosion performance, industrial applications and economic aspects.

Jan 1, 2005

By Terry A. Lee

The grade of mineral resource estimates underpin all major economic considerations of mining prospects. While there are reporting requirements to satisfy when stating a mineral resource estimate, an examination of NI 43-101 style reports suggests that authors believe that there are a number of ways of addressing these requirements. As a reader of these reports, some processes are helpful and others are not so in getting a real appreciation of how realistic is the mineral resource estimate. Critical to the quality of the mineral resource estimate is how realistic is the economic mineral concentration component. In many cases, this parameter is called grade. The industry experiences increased difficulty in achieving realistic grade estimates as the sample grade variability increases. The level of transparency in a mineral resource estimate is very much a function of the competence of the author. An objective presentation of the critical data, such as grade sampling variability can assist a reader in determining if the geology of the mineral concentration is adequately understood and how realistic is mineral resource grade estimate. This paper has considered 75 NI 43-101 style reports released in the past 2 years. The review demonstrates features that were helpful in determining how realistic the mineral resource grade estimate is. If there is insufficient transparency in a report, there is a high probability that the mineral resource grade estimate will not be realistic.

Apr 1, 2005

By J. -P. Lempainen, K. Haavanlammi, K. Valkama

The Outotec nickel matte chloride leaching process is especially suitable for small capacity standalone nickel refineries. The matte is leached with hydrochloric acid and oxygen in multiple steps in atmospheric OKTOP® reactors. Oxygen feed prevents reductive conditions and hydrogen formation in the leaching process. A sufficiently high acid concentration keeps the iron dissolved in the solution. The iron is removed after leaching by precipitation and the pregnant leach solution continues to metals recovery. The metals recovery from the chloride solution is done by solvent extractions, where ammonia is used for pH control. Copper and nickel are stripped from the organic phases into sulphate solutions. Metals can be recovered from the purified solution by conventional electrowinning. Other alternatives are precipitation or hydrogen reduction processes. The ammonium chloride in the solvent extraction raffinate is regenerated into ammonia. The calcium chloride solution from ammonia regeneration is routed to acid regeneration where sulphuric acid is added and ultra-pure gypsum is precipitated. The hydrochloric acid solution from this step is recycled back to leaching. The process yields good overall metal recoveries and expensive

Jan 1, 2014

By M. N. Vu, G. Armand, A. Noiret, D. Seyedi

Extensive experimental observations have been performed around drifts and shafts at the Meuse Haute-Marne (North-Eastern France) Underground Research Laboratory (URL) to assess the extent and pattern of the induced fracture networks in Callovo-Oxfordian claystone. Two types of fractures can be distinguished, namely; shear (mode II) and extension unloading (mode I) fractures. The main goal of the present paper is to provide insights on the extent of damaged zone around the drifts by numerical simulations. A two-step approach is considered. First, 2D elastic calculations are performed for simulating the excavation of a drift following the direction of the major in situ horizontal stress. The evolution of the stress and strain tensors around drifts are calculated by applying an elliptic convergence function on the drift wall. Two failure criteria are then verified for the induced damage. An equivalent strain and an elastic energy based damage criteria are considered. The obtained results show that the employed simplified approach provides rather good results for the extent of the damaged zone for drifts parallel to major horizontal stress. However, post-processing elastic results leads to overestimating the vertical extent of the damage zone and underestimating its lateral extent, where the in situ stress state is quasi-isotropic in a section of drift. Nevertheless, the obtained results show that an energy based damage criterion could be considered as a potential way for reproducing the extent of the damaged zones if a convergence function can be defined for the drift.

Jan 1, 2015

By 13409, B. Arjang

"Ground stresses are one of the significant factors in the context of rockbursts and underground instability at some hard rock mines in Canada. Overcoring strain relief measurements using triaxial strain cells were performed at several mine sites to provide stress data for stability evaluations and mine design. Pre-mining stress determinations carried out at depths between 60 m to 1890 m resulted in the following average stress gradients:Maximum horizontal stress, s Hmax = 8.18 + 0.0422 MPa/m depthMinimum horizontal stress, s Hmin = 3.64 + 0.0276 MPa/m depthAverage horizontal stress, s Ha = 5.91 + 0.0349 MPa/m depthVertical stress, a; = 0.0266 ± 0.008 MPa/m depthThe maximum and minimum horizontal compressive stresses, with an average ratio of J.75 ± 0.45, prevail in east-west and northerly directions, respectively. Horizontal compressive stresses in excess of vertical overburden load were determined, indicating large variations in ratios to a depth of about 1000 m with decreasing trend toward depth.From present data, a particular regional zoning for the magnitude and direction of horizontal stress fields cannot be outlined.A common feature at mines with near vertical orebodies is that the maximum horizontal stress acts perpendicular to strike while the minimum horizontal stress is aligned on-strike. The vertical stress components approach the gravitational overburden load."

Jan 1, 1991

By Adrian Dance

Barrick Gold Corporation?s wholly owned subsidiary, Barrick (Granny Smith) Pty. Ltd. operates the Granny Smith operation in Western Australia and has been operating a SAG/ball mill (SABC) circuit for a number of years processing a blend of secondary crushed, open pit and underground material. With closure of the open pit in 2008, Granny Smith was considering different options to process underground only material at a lower throughput. Using the existing equipment available on site, a number of circuits were evaluated by Metso Process Technology & Innovation (PTI) through modelling and simulation. A single stage SAG mill circuit was selected based on power requirements and operating efficiency. In April 2008, Granny Smith started operating their SAG mill in single stage, processing primary crushed feed and using their existing pebble crusher for scats. Smaller cyclones were used to control the final grind size. PTI surveyed the new single stage circuit in July 2008 under normal and high rock load conditions to show the effect on final grind size. Granny Smith continued to operate this circuit until the end of 2009, when alternate feed sources resulted in a return to the complete SABC circuit at a higher tonnage once again. The variable speed, single stage SAG mill proved itself to be very efficient and flexible option to process the hard underground Wallaby Deeps material. This paper will review circuit performance following the conversion to single stage in comparison with the full SABC circuit.

Jan 1, 2012

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