Search Documents

By M. Nete

Qualitative analysis of three mineral samples from northern Mozambique was accomplished by X-ray diffraction (XRD), microscopy, scintillation, and magnetic determination. Chemical characterization was accomplished using ICP-OES and pressed powder XRF after dissolution using microwave-assisted digestion and flux fusion. The samples consisted of a mixture of minerals with tantalite as the major component, with minor amounts of microlite, quartz, mica, accessory garnet, and tourmaline. Incomplete dissolution was obtained using microwave digestion, with maximum recoveries of 90.25 ± 0.06% and 88.90 ± 0.04% for Nb2O5 and Ta2O5 respectively. Complete dissolution using Li2B4O7 resulted in recoveries of 98.5 ±0.2% Nb2O5 and 100.4 ± 0.3% Ta2O5.

Jan 1, 2012

By R. M. Tyler

The importance of the heavy mineral sands to southern African economic well-being cannot be over emphasized. The value of the worldwide titanium dioxide industry is estimated at $7 billion. Six out of eight of the world?s proposed new project areas are in southern and eastern Africa, and a seventh at Tamil Nadu in India is partly owned by a South African company, Kumba Resources. The HMS industry can be highly lucrative; its return on capital is the best in the mining industry. There are, however, problems, particularly in the smelting technology. The often considerable affect of mining on environmentally sensitive coastal dunes is also becoming increasingly important. The medium-to long-term demand for Ti-dioxide pigment and ceramic grade zircon appears very healthy. If, as predicted, the costs of producing titanium metal can be reduced, then there is huge potential for the metal. In summary, the HMS industry is an attractive one to be in, with good returns and generally manageable risks. New business opportunities and their economic implications for Africa are discussed.

Jan 1, 2004

By Steven Rupprecht

"The South African Code for the Reporting of Exploration Results, Mineral Resources and Mineral Reserves (the SAMREC Code or the Code) sets out minimum standards, recommendations and guidelines for public reporting of exploration results, mineral resources and mineral reserves in South Africa. The Code is being rewritten and updated, with work on-going since 2013. The revision is being undertaken to ensure the Code remains relevant, incorporates developments within the international mining industry and keeps abreast of technology. Some of the changes and their reasons are discussed. INTRODUCTION The South African Code for the Reporting of Exploration Results, Mineral Resources and Mineral Reserves (the SAMREC Code or the Code) sets out minimum standards, recommendations and guidelines for public reporting of Exploration Results, Mineral Resources and Mineral Rreserves in South Africa. The Code is being rewritten and updated, with work on-going since 2013. The proposed 2016 update is not significantly different from previous versions of the Code. The revision is being undertaken to ensure the Code remains relevant, incorporates developments within the international mining industry and keeps abreast of technology. The Code sets out minimum standards, recommendations and guidelines for public reporting. However, reporting may be unclear, ambiguous, selective, obscure, distorted or unintelligible. Part of the purpose of the update is to provide better guidelines to Competent Persons (CPs) to ensure that the material information is being reported and that reporting is transparent in accordance with the guidelines of the Code. The information provided should, as required by the Code, be adequate to inform the “reader/investor or potential investor to make a reasonable and balanced assessment of the significance of the information”. The Committee for Mineral Reserves International Reporting Standards (CRIRSCO) highlights the importance of the reporting Codes, as provided from a quote from their website (www.crirsco.com): The mining industry is a vital contributor to national and global economies. It is a truly international business that depends on the trust and confidence of investors and other stakeholders for its financial and operational well-being. Unlike many other industries, it is based on depleting mineral assets, the knowledge of which is important prior to the commencement of extraction. It is therefore essential that the industry communicates the risks associated with investment effectively and transparently in order to earn the level of trust necessary to underpin its activities."

Jan 1, 2015

By J. Utt

Recent developments in circulating fluidized bed (CFB) once-through supercritical technology (OTSC) have enabled this technology to be offered as a utility-scale alternative competing head-to-head with pulverized fuel (PF) OTSC offerings. One clear example is the CFB supercritical unit at the Lagisza Power Plant in Poland, owned by PoludniowyKoncernEnergetyczny SA (PKE). This unit has now been in commercial operation for three full years, exhibiting very good performance, and has validated Foster Wheeler's performance model at this utility scale as well as for units in the 600 MWe and 800 MWe size ranges offering net efficiency of -43 per cent (LHV basis). This operating unit has also proven the use of the world's first FW/BENSONTM vertical-tube OTSC low mass flux technology. Since the Lagisza original international tender specified OTSC PF technology, it is important to note that the alternative selection of CFB OTSC technology over conventional PF technology is of historic significance, not only for the validation of the CFB supercritical platform as a viable alternative to conventional PF technology, but it also positions the CFB OTSC with fuel flexibility for offering of sizes up to and including 800 MWe units. This paper explores the differences between CFB OTSC technology and standard PF OTSC in utility power generation. Selection criteria, fuel burning range in both technologies, and other selection drivers are discussed. Economic analysis of both technologies, based on existing cases, is also provided. Also discussed are the technical advantages and uses of each technology. Foster Wheeler has recently been awarded a contract for four units of CFB OTSC technology, which utilizes a 2 on 1 configuration of two 550 MWe CFB OTSC boilers on two single 1000 MWe turbines. Essentially this provides a fuel-flexible low-emissions alternative to a 2 x 1000 MWe solid fuel power block.

Jan 1, 2012

Although not fully accepted in South Africa, the Theory of Sampling originally proposed by Pierre Gy is fast becoming the cornerstone of sampling practice throughout the world. The growing acceptance of Gy?s Theory of Sampling in South Africa can be attributed to a number of factors, chief amongst them being the development of a tradable mineral asset market, the promulgation of the Mineral and Petroleum Resources Development Act (MPRDA), the growing number of commercial and academic courses that are offered on sampling, and the regulation of the industry through internationally acceptable guidelines and rules for reporting and trading in mineral assets. The size of the South African minerals industry and the dependence of our economy on mineral production have also meant that correct sampling is of key importance to mineral trade. ISO standards have been the principal guides for producers of mineral bulk commodities who produce to customers? specifications, whereas Gy?s insights have been most readily accepted by precious and base metals producers whose product is sold into metal markets. Understanding of small-scale variability is essential in the precious and base-metal industries, but detailed studies of the effects of heterogeneity have not been as productive in the bulk commodities. Sampling practices at different stages of mineral development from exploration, face sampling and grade control, ore processing and handling, metallurgical sub-sampling, point of sale sampling, and sampling in the laboratory are considered in the gold, platinum, ferrous metal, and coal industries. A summary of the impact of poor sampling in these industries is presented. Generally it appears that poor sampling practice is most likely to erode mineral asset value at the early stages of mineral development. The benefits of good sampling are considered, especially with regard to the financial implications of bias and error on large and consistent consignments of bulk commodities. Keywords theory of sampling, sampling errors, ISO standards.

Jan 1, 2014

By J. J. Stacey

Large mining companies generally follow the distributed ownership corporate model, with a board of directors responsible for decisions that affect both shareholder value and stakeholders of the company. The board is simultaneously responsible for setting the culture and values of the corporation, which drive performance and priorities. Companies listed on the Johannesburg Securities Exchange (JSE) commit to sustainable development in various ways, either by virtue of implementing the King Code of Governance 2009 (King III) and/or through their own public reporting on social and environmental matters. Many mining companies make public statements regarding their support for environmental stewardship, ethical behaviour, and fair treatment of communities. It is a local, regional, and indeed, global phenomenon that companies fail to deliver on these statements. Research was carried out through the Institute of Directors of Southern Africa in 2007, and followed up in 2012, regarding directors? understanding of sustainable development issues, the relative priorities, what is needed for ?radical change? to effect sustainable development, and what enables or constrains the latter. Pertinent findings of both surveys are presented in this paper, and it is suggested that ?on-the-ground? performance may be indicative of the nature of leadership and decisions in the topmost ranks of the company. The results indicated that environmental concerns fall consistently below social issues. Financial capital ranked most important, and while environmental issues are recognized as being of strategic concern for the long-term, they ranked as being the lowest importance of all ?Five Capitals? (Financial, Manufactured, Social, Human, and Natural). Social capital ranked second lowest, with black economic empowerment being the only high-priority social issue. There is also evidence that certain companies within the mining sector fail to recognize their absolute dependence on natural resources. Much is made in academic and popular literature of the need for a new type of leadership for the radical shift to sustainable development: at company level this implies therefore a new type of director. The research found that only 14 per cent of directors felt that board decisions are consistent with their personal values; while intentions are strong to behave ethically and serve sustainable development, actions to give effect to these intentions lag significantly. Respondents indicated that the top impediments to courageous leadership for sustainable development related to personal issues of maintaining the image of being a director, fear of appearing weak, fear of being a lone voice, and bowing to board-colleague peer pressure.

Jan 1, 2014

By M. W. Hildyard

A blast was engineered close to a tunnel in a deep-level mine with the purpose of studying the wave interaction with the tunnel. Numerical modelling of seismic wave propagation was used in both the forward planning and the back-analysis of this experiment. This paper presents the forward analysis. A likely model for a propagating blast was investigated and developed. This model predicts the influence of the velocity of detonation of the blast and the rise-time of the blast on the radiation pattern, and hence on the expected positions of maximum velocity at the tunnel surface. An attempt was made to quantify the parameters of the blast model, using waveform data from two other blast experiments. The blast model caused particle velocities which were an order of magnitude lower than these measured velocities. Large values for blast pressure were required to match the amplitudes of the recorded velocities. An alternative ?effective? source with a similar mechanism but increased diameter was proposed to account for the higher velocities. It was established that distant velocities scale with the square of the borehole diameter. A calibration blast at the experimental site was also modelled. Agreement in the waveforms was not sufficient for conclusive statements to be made about the validity of the source model. However, the modelled waveforms show an encouraging correspondence with some measured waveforms. This gave confidence that the model could be used to examine wave propagation around the tunnel, and allowed a scaled source to be developed for forward analysis of the main experiment.

Jan 1, 2001

By K. Y. Lee, J. M. Park, C. W. Park

Laboratory-scale experiments were carried out to investigate the effect of ladle slag composition on the decreasing behaviour of aluminum content in steel, i.e., reoxidation of molten steel in contact with slag, using a vertical tube furnace. The steel composition and the slag system chosen were for ultra low carbon and CaO-Al2O3-FeO-MnO-MgO based to simulate ULC steel grades. It has been observed that the reoxidation rate shows a good relationship with (T. Fe) content in slag. In case of (T. Fe) content higher than 1.5 weight per cent, the reoxidation rate increases linearly with increasing (T. Fe) content. For a given (T. Fe) content, it decreases with the increasing (CaO)/(Al2O3) ratio of the slag. However, in the case of (T. Fe) content lower than 1.5 weight percent, no particular relationships are observed between the reoxidation rate and (T. Fe) content or the (CaO)/(Al2O3) ratio of the slag. These results were used to propose for an aim a slag composition for the ULC steel grades in a steelmaking plant. Keywords: slag deoxidation, slag killing, slag composition control, IF steel, ULC steel grade

Jan 1, 2004

By C. Lemmer

The resolution with which the different categories of resources are forecast for Witwatersrand gold reefs should ideally tie in with block sizes that are optimal in terms of the variability structures of the reefs. A tool, called the ?variancegram?, is proposed as a basis for block size choice. A variancegram is intrinsic to the particular reef and mine concerned. A further requirement is the ability to attach global confidence limits to weighted average estimates built up from combinations of local kriged estimates. Approximations to derive global kriging variances based on variables derived from local kriging deliver hugely inflated results if ordinary kriging is used, and markedly better, but not accurate, values if simple kriging is used. These approximations improve as the number of samples used in kriging each block is increased. It is shown that the behaviour of the different components of the global kriging variance with increasing number of samples, all differs, but they all link to the variancegram for the particular reef. The variancegram can thus be used to correct the different components to the values they would have had if all samples were used in kriging each block, and so deliver the ?correct? global kriging variance, even though only a limited number of samples were used in kriging each block. The desirability of having very stable solutions implemented in production systems is taken into account in the proposals. It is anticipated that the same variancegram findings will also hold for other densely sampled deposits, but this remains to be investigated.

Jan 1, 2014

By S. H. Badenhorst

Zacharias Lewala discovered a diamond at Grasplatz, near Lüderitz in April 1908. While this stone may not have been the first diamond to be found in this barren, inhospitable area, it most certainly was the stone that set off a chain reaction ultimately leading to the creation of Namdeb as we know it today. [ ]

Jan 1, 2003

By P. F. Ledwaba

South Africa?s Northern Cape Province houses a variety of gemstone minerals. Diamonds remain the primary gemstone in the province, but there are also tiger?s eye, sugilite, rose quartz, jasper, amethyst, amazonite, tourmaline, and topaz deposits. This paper gives an overview of the tiger?s eye (TE) mining sector in the Northern Cape, with the objective of assessing the potential of TE to contribute to job creation and poverty alleviation in the province. Mining of TE started in 1803, but gained momentum during the 1960s. Mining is conducted on a small-scale basis, and in most cases without any legal framework. The majority of the residents are not benefiting from TE. It has been reported that an estimated R700 000 to R800 000 worth of TE is leaving the country every week through illegal trading. The potential for TE mining to contribute to socioeconomic development is hindered by illegal operations, lack of technical skills and knowledge, exploitation by dealers, lack of market information, community rivalries, and absence of value addition. A multi-stakeholder approach is needed to formalize the sector and reap the benefits. Mintek is spearheading a TE project in the Prieska area of the Northern Cape, which aims to establish a gemmological centre where TE can be beneficiated into saleable and high-value products..

Jan 1, 2014

By P. Share

As part of JAGUARS (JApanese-German Underground Acoustic emission Research in South Africa), high-resolution tiltmeters were installed during 2007 at Mponeng gold mine, South Africa, in an attempt to study and understand the risks posed by in-mine seismic events and to subsequently use the knowledge gained to mitigate these risks. In December 2007, an M2.2 event occurred at Mponeng and was detected by the tiltmeters installed. The tilt expected for the M2.2 event is modelled through the use of numerical and analytical tools and utilizing, as input parameters, characteristics of the event such as the rupture area, the amount of slip, elastic properties of the rock, the state of stress before the event occurred, and frictional parameters of the rupture. The calculated tilt values are correlated and compared with the recorded data, and are found to be of the same order. Certain input parameters, such as the initial rupture point, are further constrained by varying them, during modelling, until the calculated and observed tilt values are approximately equal. From modelling, it can be concluded that the rupture point had to be closer to the location of the tiltmeters. In addition, possible locations acting as source areas for the observed aftertilt are determined. Further modelling still needs to be done to quantify the effects of the tunnel close to the tiltmeters and plastic deformation of the rock mass. Keywords high-resolution tilt data, Mponeng gold mine, M 2.2 seismic event, analytical modeling, numerical modeling, rupture plane.

Oct 1, 2013

By R. G. B. Pickering

The purpose of this report is to chronicle the interaction between Anglo Platinum and Sandvik Tamrock during the development and application of Long Hole drilling as a viable, practical method of narrow reef mining. Appendix I identifies the people from Anglo Platinum and Sandvik Tamrock who participated in the project. Experiments in long hole drilling and blasting were conducted in a number of mines during the late 1950s and early 1960s. Some of the advantages envisaged were: ? The face would require no daily examination ? No barring down or immediate face support required ? No sockets to clean ? Only a few well drilled holes rather than large numbers of short blast holes ? Avoidance of poorly-marked holes, inaccurate drilling and improper charging ? Better supervision and better face advance. Holes 50 mm in diameter were drilled parallel to the stope face by pneumatic rock drills from advance strike gullies. Holes were charged with amon gelignite cartridges. All holes were traced with detonating cord to ensure propagation of the explosive along the entire hole length. Rock was blasted onto a rolling scatter pile. A number of reasons have been cited for the failure of this mining method. The two most important were: ? problems in controlling hole deviation which was exacerbated by the fractured ground ? difficulty of charging up holes in fractured ground, due to relative movement of rocks along fracture surfaces over the hole length. The idea of drilling a few long holes parallel to the face rather than a large number of short blast holes perpendicular to the face was an attractive option to mechanize narrow reef mining. This particular project started in 1998. Initial trial The Union Section of Anglo Platinum purchased a Solo H 606 RA long hole drilling rig from Sandvik Tamrock in 1997. In order to evaluate long hole drilling as a basis for face mechanization. The Solo series of drill rig was selected because it combines high drilling performance and good drilling accuracy with high versatility with a compact size. The mining method proposed by a consultant to Anglo Platinum was based on the current trackless mining method practised by Union Section. The trackless mining method is better described as a trackless development and transport method. Reef strike drives were installed approximately 40 metres apart on dip using trackless equipment. Reef was mined using hand-held pneumatic rock drills and scraper winches and the excavation was supported by sticks and crush pillars at the top of the panel. Rock scraped from the panel was loaded by Lad Haul Dumps (LHDs), trammed to an inclined haulage, and loaded into trucks for transport to the surface. It was intended that the Solo would drill from the bottom advance strike drive to the top advance strike drive. Blasted rock would still have to be scraped from the panel. It is also difficult to understand how well the support would have survived. In the initial motivation it was stated that drilling accuracy was of the utmost importance. Previous work by Sandvik Tamrock had shown that, in other parts of the world, 20 metre long holes could be drilled with a deviation not exceeding 100 mm provided that: ? holes were collared with low thrust pressure ? operating percussion pressure was kept low -often not ? exceeding 10 MPa ? drilling was to be with tubes, not rods. In early 1998 initial trials with the Solo H 606 RA were inconclusive; attempts were made to drill 40 metre long holes from one advance strike drive to another. Difficulty was experienced in measuring hole straightness and accuracy, there was no survey of the holes, and no blasting was done to expose hole barrels. A long hole mining system was then defined, with a corresponding face mining system, which led to the design of a test programme for the Solo drilling rig. Earlier experience at Telfer The proposed face mining system was based on a mining method that is practised at Telfer Gold Mine in Western Australia. Telfer has a series of domed reefs that dip at angles of less than 20° to more than 40°. The ore bodies are all thin, with widths from 0.6 metres to 0.8 metres. Figure 1 shows the face layout. Experience at Telfer indicated the following: ? The spacing and size of the dip pillars should be based on the rock mechanics of Union Section ? The length of the face is determined by the ability to correctly position the blast holes. Positioning of blast holes is a function of hole straightness and hole direction. Blast holes in Telfer were drilled 64 mm diameter with retrac bits from Sandvik. Extension steel was T38 by 1.8 metre lengths. A guide rod was fitted immediately behind the bit. To achieve greater accuracy' operators drill at relatively low percussion and thrust pressures. Results show that very good accuracy is achieved in 20 metre stopes and only fair accuracy in longer stopes. In 34 metre long stopes 10-15% of the holes are not sufficiently accurately positioned and have to be re-drilled. All blast holes are

Jan 1, 2006

By P G. Gaylard

The minerals Industry carries out over US$ 2 billion of R&D annually. The R&D is performed at three levels ? in house research facilities; on a one-on-one basis with a research provider; as syndicated (collaborative) research. The key to a successful R&D programme is project selection. It is suggested that at this time of ?bounty? for the minerals industry the research portfolio should contain some ?blue sky? but potentially long term significantly beneficial projects. This said, the projects should align the technical concepts with the business imperatives. AMIRA has found the ?Road Map? concept has assisted in defining areas for R&D. Examples of this methodology using a Copper Road Map developed by AMIRA for the copper producers operating in South America is presented. The benefits derived from syndicated research are also discussed.

Jan 1, 2007

With the passing of Dr William Bleloch on 20th June 1991, we lost not only a colleague and a good friend but a man who made a very significant contribution to the industrial development of South Africa. He can rightly be called the father of the ekctrochemical and electro-metallurgical industries in this country, since his ideas and dedicated work were directly responsible for the development of these large industries as we know them today.

Jan 1, 1991

An efficient and portable mathematical model has been developed for simulating heat transfer in six-in-line slag-resistance-heating electric furnaces for smelting sulphide ores to produce base metals and platinum group metals. This model is a steady-state one relating furnace conditions and performances to various control and input parameters. Some transient effects occurring in electric furnaces are neglected for computation efficiency. This article describes the model development and modelling results. The present model is capable of predicting: (i) temperatures at various locations in a six-in-line furnace, such as slag bath, matte bath, solid charge, freeboard space, freeze lining, cooling water, and air gaps between solid material components, etc. (ii) freeze lining thickness, (iii)smelting rate, and (iv) heat loss rate, etc. A typical feature of the model is that it is easily portable to different application platforms and sufficiently efficient with execution times less than a few seconds. Therefore, it is possible to apply the model for online prediction and control of heat transfer and freeze lining thickness in industrial electric furnaces.

Jan 1, 2011

By F. T. Wang, Y. Yuan, S. H. Tu, X. G. Zhang

Hydraulic support instability (HSI) is one of the most common causes of disasters in underground coal mining, posing a threat to the safety of mine workers and normal operation of the equipment. It is prone to occur in fully-mechanized mining faces with a large mining height (FMMLMH), especially when the dip angle of the coal seam is large. The key to controlling HSI is to deduce its mechanism and employ effective control techniques. This paper focuses on the analysis of HSI types, the key parameters and techniques to control HSI in FMMLMH, the establishment of a model of HSI in FMMLMH, and a multi-parameter sensitivity mechanical model of different HSI forms in the no. 7219 longwall face in Xutuan Coal Mine, Huaibei Mining Group, by using sensitivity analysis . The results show that HSI mainly presents in three forms: hydraulic support gliding (HSG), hydraulic support tilting (HST), and hydraulic support tail twisting (HSTT). The occurrence of the above three forms depends mainly on support anti-instability capability. In the no. 7219 longwall face, HSG and HST are the main two forms of HSI. The dip angle of the working face and the friction coefficient between floor and hydraulic support are the sensitive parameters for HSG, while HST is strongly dependent on the dip angle of the working face and the friction coefficient between roof and hydraulic support. By the applications of measures such as the oblique layout of the working face, cutting the floor into a step pattern, moving the support under pressure, and raising the setting load, the support stability was controlled effectively.

Jan 1, 2015

By Luthie Els, Chris Coetzee, Roedolf Koekemoer, Casper Els, Olof Vorster, Frank Fereday

Three electric arc furnaces (EAFs) were constructed by Union Carbide at Tubatse Ferrochrome (TFC) in1975. This plant was expanded to four furnaces at the East Plant in 1985 and the gas cleaning system was enlarged without taking flow dynamics into consideration. In order to improve competitiveness, TFC has increased furnace capacity, partially substituted coal for coke, and changed the charge mix. These changes increase the heat and dust load into the air pollution control (APC) system. TFC embarked on a project to improve the APC capacity by: ?Quantifying the required off-gas capacity ?Optimising the aerodynamic flow through the system by means of Computational Fluid Dynamics (CFD) ?Converting the old Brandt Koyo top entry baghouses to bottom entry ?Upgrading the cleaning system ?Installing Gore-tex® ePTFE membrane filter bags The estimation of required off-gas capacity is traditionally done by rule-of-thumb methods such as specific extraction. A more elegant evaluation method was developed for simulating the effects of changes in charge mix on the furnace reaction gas volume and the APC system. Reaction gases are calculated by modelling the reactions taking place in the furnace and off-gas volumes are compared to both fan and baghouse capacity. This enables identification of bottle-necks in the off-gas system. CFD techniques were utilized to evaluate geometric imbalances and propose solutions to ensure optimal use of off-gas capacity. The system modifications were completed over several months and the modified equipment went into service in May2006.

Jan 1, 2007

By A. McLean, M. Iwase, K. Wakimoto, K. Katogi, Y. Kikuchi

One of the greatest obstacles to the application of physical chemistry principles to the elucidation of slag-metal reactions is a lack of knowledge of activities of the reacting species. To a large extent, oxygen potential of the slag phase governs iron and steelmaking practice. Without oxygen control by means of appropriate sensors, the behaviour of the other elements cannot be managed. In this paper, measurements of the FeO activity with various types of electrochemical FeO sensors will be described together with examples of their applications for improved strategies toward better practice for ladle metallurgy and sulphur and manganese distributions between slag and metal phases during steelmaking. Measurements of FeO activity have also been made in order to improve dephosphorization reactions. This type of work has led to significant reduction in volume of slag generated within the steelmaking vessel, which in turn has important implications for refractory wear, metal yield, alloy recovery and improved productivity. Finally, an on-line sensor is described that permits the oxygen potential to be determined for both the metal phase and the slag phase during steelmaking in the BOF.

Jan 1, 2004

By K. Coleman, C. Visser, N. Selby, X. W. Pan, T. Letwaba

A prototype weightometer soft sensor is under development at DebTech, De Beers in partnership with Venetia Mine. Its main function consists of an online data source; data validation; dynamic weightometer models; online detection of weightometer failure; alarming and replacing a weightometer when it fails. 31 dynamic models have been developed. All models with R2>= 85%are accepted as good models and they generally have satisfactory results with a daily error of less than 8% within 90% confidence level. By using those models, a set of soft sensors can be set up to cover all weightometers in the entire processes starting from primary crushing to the stock piles of dense medium separation (DMS). Some practical issues and concerns on model development are discussed, including data collection, data analysis, data pre-processing, variable selection, time delay, model training and model validation. When deploying the weightometer soft sensor system, the potential benefits could include: reduction in plant downtime; reduction in unplanned maintenance and even replacement of non-critical weightometers, particularly where nuclear weightometers are used. Keywords: soft sensor, weightometer, dynamic modelling, neural network modelling

Jan 1, 2003