Search Documents

By J. Yao

Synthetic indium-bearing zinc ferrite (IBZF) was mechanically activated by a tumbling mill. Subsequently, the behavior and kinetics of leaching zinc and indium from the initial IBZF and activated IBZF in sulfuric acid solution were investigated. The results demonstrate that the tumbling ball milling enhances the leaching reactivity of IBZF and increases the zinc and indium extractions. The reaction temperature and H2SO4 concentration have an important effect on leaching zinc and indium from the initial IBZF and activated IBZF. The kinetic model -ln(1-x) = kt was used to describe the kinetics of leaching zinc and indium from the initial IBZF and activated IBZF and to calculate the kinetic parameters. The tumbling ball milling causes a decrease in activation energy and reaction order, which indicates that the dissolution of IBZF after mechanical activation by a tumbling mill becomes less sensitive to temperature and H2SO4 concentration. For the zinc leaching, the activation energy decreases from 73.9 to 70.3 kJ/mol and the reaction order decreases from 0.72 to 0.70. For indium leaching, the activation energy decreases from 78.1 to 73.5 kJ/mol and the reaction order decreases from 0.82 to 0.69. It is found that the dissolution rates of the initial IBZF and activated IBZF are controlled by surface chemical reactions.

By Alexander N. Winchell

"Location. The iron formation of the Gogebic district (including the Penokee) comprises a narrow belt south of the western part of Lake Superior extending from Michigan for a considerable distance into Wisconsin. From the mouth of the Montreal River (about twenty-five miles west of Ashland) the iron formation is only about twelve miles distant, but it is about twenty miles south of Ashland. Beginning on the east at a point about eight miles east of Wakefield, Mich., it extends through Sunday Lake, Bessemer and Ironwood in Michigan, and thence south of west through Hurley, Wis., and only a short distance south of Mellen, Wis. About eight miles westward from Mellen, the formation disappears, but six miles further west it is known to reappear and continue thence for about eight miles. The total length of the range is about sixty miles, two-thirds of this distance being in the State of Wisconsin; but nearly all of the ore has been obtained from the eastern third of the district, which is in the State of Michigan.History. This iron range was first noted in 1848 by Dr. Randall, assistant geologist to Dr. David Dale Owen, while following the fourth principal meridian northward. It was also reported by Charles Whittlesey in an account of a survey made in 1849, and published by Owen in 1852. Analyses of the iron ore are given with the following account. ""The bed of magnetic iron ore south of Lac des Anglais is of extraordinary thickness,25 to 60 feet. * * In the wild and deep ravines where the Bad River breaks through the range, there is a cliff of slaty ore * * * probably three hundred feet thick. * * * I estimate more than one-half of this face to be ore, and in places the beds are from ten to twelve feet in thickness, with very little inter-mixture of quartz * * * . The position of the best exposures of ore which I saw is such as to require from eighteen to twenty eight miles of transportation to reach the lake."""

Jan 1, 1920

By T. Eisele, T. Weldum, G. Rajala, H. Walqu, S. Kawatra

The throughput and energy efficiency of comminution circuits is reduced when ore characteristics change, causing bottlenecks in some portions of the circuit while leaving excess capacity in others. The results of analysis of three years of daily grinding circuit operating data for a full-scale plant has shown that the location of bottle-necks in the circuit will change depending on the nature of the feed ore. The circuit tends to be limited by the primary mill capacity when the ore hardness increases, and limited by the secondary mill capacity with softer ores. Two methods are considered for shifting load in the circuit to relieve bottlenecks by taking advantage of excess capacity in other parts of the circuit: 1) Addition of supplemental comminution operations, such as high-pressure grinding rolls, to process specific size fractions that are not handled efficiently by the primary mill; and 2) Use of screens to shift grinding load upstream from the secondary mills back to the primary mills. In the case of the plant examined in this study, it was found that 58% of the time there was sufficient excess capacity in the primary mills that the throughput of the circuit could have been increased by using screens to shift the grinding load from the secondary mills to the primary mills.

Jan 1, 2004

By Mark S. Stagg, Michael J. Nutting, Douglas R. Barlet

Single-row bench blasts at a reduced scale in the field were used to study the effects of delay time on fragmentation. Nine shots with a spacing of 21 in. and a burden of 15 in. (S/B = 1.4) and nine shots with a spacing of 30 in. and a burden of 15 in. (S/B = 2.0) were completely screened to determine particle size distributions. Delay intervals between holes ranged from 0 to 45 msec, corresponding to effective delay intervals ranging from 0 to 36 msec per foot of burden. Each shot was instrumented with strain and pressure gages to measure in situ blast dynamics and to evaluate interactions between blastholes. Finer overall fragmentation was produced for shots with delay intervals between 1 and 17 msec per foot of burden. Only for shots within this optimum delay range was it observed that the strains induced by stress waves constructively interacted with strains induced by gas pres- sure from an earlier detonated hole. Coarsest fragmentation resulted at delay intervals less than 1 msec/ft, where stress waves from each hole were observed to interact destructively, and at delay intervals greater than 24 msec/ft, where no inter- action between holes was observed, indicating a condition that can be considered as the firing of single hole shots independently.

Jan 1, 1986

By M. Tirmarche, J. Brenot, J. W. Piechowski

INTRODUCTION Statistical analysis of results obtained by workers survey is very often limited to the determination of the collective dose and mean annual values. This information is not sufficent by itself for radiation protection purposes since the cumulative exposure of a worker during his working life is not taken into account. A more complete analysis would be obtained by considering both: [ ] These two notions are different. The former depends on the evolution of the protection with time and is related to annual limits (ICRP, 1977 and ICRP,1977 a). The second depends mainly on the works performed during the period of survey, the interference of protection evolution is weakened. So it is useful to foresee how the individual average annual doses would be when considering survey periods of 5, 10 years, or more. STUDIED POPULATION In order to analyse the two previously mentionned notions it is necessary to follow up a well defined population of workers which is continuously exposed to the radiological hazards on a long enough period. The selected population is that of the workers of the French uranium mines always present during the 1971-1979 period (9 years). The number of individuals is 937. The radiological hazards are: 1. exposure to radon and short-lived daughters 2. exposure to long-lived ore dust 3. external irradiation. The workers of the uranium mines are exposed to these radiological risks in different ways according to the work performed. It is therefore necessary to subdivide the population into more homogeneous groups with respect to the radiological risks. Three groups are considered: - Group 1 : 227 individuals whose radon and dust exposures are always nil - Group 2 : 449 individuals whose radon and (or) dust exposures are different from 0 every year - Group 3 : 261 individuals represent a mixed category. Some features of these groups are given in table 1. [ ] The 937 followed individuals represent approximatively the half of the whole population of the French uranium mines. Taking into account the procedure adopted to select the studied population, it appears clearly that this population is not representative of the whole miner population. DOSIMETRIC RECORDS Three risk indicators are considered .Individual annual exposures to radon and its shortlived daughters. They are determined by ambient monitoring based on radon concentration measurements and recording of the times elapsed in the various workplaces. For each worker the annual exposure is represented by the mean annual air concentration and is expressed as a fraction of the maximum annual concentration (MAC). Given the administrative arrangements and the effective state of equilibrium between radon and its daughters, the maximum annual concentration is practically equivalent to 1 Working Level = 2.08 x 10- 5 jm-3. (UNSCEAR, 1977). . Individual annual exposures to long-lived ore dust which are also expressed in fraction of the maximum annual concentration. . Individual annual doses of external irradiation (X and [y] rays) expressed in mSv (= 10-3 Sievert). Irradiation monitoring is based on use of individual film badges.

Jan 1, 1981

By Michael L. Cramer, Kunsoo Kim

INTRODUCTION The Columbia River Basalt beneath the Hanford Site is one of several rock types being considered by the U.S. Department of Energy for the first commercial high-level nuclear waste repository. The Basalt Waste Isolation Project (BWIP) has been carrying out a comprehensive investigation to assess the feasibility of constructing a repository in basalt and to develop technology required for repository engineering and site characterization. Development of a nuclear waste repository requires a number of design considerations not usually necessary in the planning of conventional underground structures. Principal among these is the thermally induced stress field resulting from waste emplacement. This thermal stress, super- imposed on the preexisting in situ- and excavation- induced stress fields, needs to be evaluated in the design process due to the sensitivity of waste emplacement density to this factor. The nature of this thermally induced stress field is dependent on several factors, including the temperature distribution around the repository openings, the thermal properties of the rock mass, and the deformability of the rock mass. Re1iable performance of the repository structure is largely determined by the ability to under- stand the above rock mass characteristics and predict the underground opening response to waste emplacement based on this understanding. To gain an understanding of this rock mass behavior in a repository condition, the BWIP has conducted a large-scale heated block test in basalt at the Near-Surf ace Test Facility (NSTF) on the Hanford Site. This paper presents the results from the elevated temperature phase of the block test in which the thermal, mechanical, and thermo- mechanical response of the closely jointed basalt rock mass was evaluated as a function of stress and loading orientation at varying temperatures up to 200 oC. SITE GEOLOGY The NSTF test areas, including the block test site, are located in the entablature zone of the Pomona basalt f low. The Pomona flow entablature is structurally typical of most flows in the Columbia River Basalt Group in that it is characterized by long, undulating, polygonal columns up to 2.4 m in length. Columns vary from 15 to 30 cm in diameter, averaging about 20 cm. Over their length, columns can vary in dip by 150 to 200. The column structures are dissected by subhorizontal cross joints that are generally discontinuous and seldom cross columnar joint boundaries. Spacing of the cross joints also averages about 20 cm. A representative joint pattern taken from mapping of the tunnel wall near the block test site is shown in figure 1. Application of the Rock Mass Rating System (Bieniawski, l979) to the block test site yielded an overall value of 62. The corresponding value from application of the Q-system (Barton et al., 1974) is 8.7 (Cramer and Kim,1985). The NSTF is approximately 50 m below the surf ace. However, this depth is sti1l above the water table and conditions are generally dry. Interstitial water exists as the result of percolation of meteoric water from the surface.

Jan 1, 1986

By Heinz Hofmann

In order to drive tunnels with low overburden in town regions, it is necessary to give a quantitative prognosis for the deformation of the underground as well as of the tunnel-lining and of its stability. These problems and their solutions are illustrated by the works for a two-track-railway-tunnel in Bochum (Ruhrdistrict). In this case the most important problems consisted in passing below buildings in a vertical distance to the fundamentals of only 3.20 m and a railway earthdam with rapid train traffic in a distance of 4 m under the dam. The results of measurements from other comparable tunnels and finite-element-computations were used to prognose the settlements on the surface and in the ground, their distributions and their development in time. To obtain these values at the site, a special system of tunnel driving was executed. The tunnel was excavated with the newest type of tunneling-machine from DEMAG, cutting an inclined front, immediately followed by a special temporary support of shotcrete and inclined steel-ribs. By using theoretical and empirical methods to assess the behavior of the tunnel and underground and by executing an optimal driving system in technical as well as economical aspects, the expected behavior, expressed in quantitative values, was almost equivalent to those received by control measurements.

Jan 1, 1974

By M. Cambridge

The design process must anticipate all construction and operational conditions and make an allowance for all eventualities. Prevention of failure during operational life, and in some cases post abandonment, requires that a risk assessment is undertaken during the design stage which defines not only the factors of safety to be used but the criteria for the structure during its operating life. These operating criteria should clearly identify the at-risk structures and the inspection and monitoring routines required to ensure compliance with the design objectives. On a mine site, it is the role of mine management to ensure that the operating rules be applied throughout the mine life and that a system of inspection and monitoring procedures be instigated. Where the risks are significant, high-level auditing by an independent engineer will also be required. Within the inspection and auditing framework, the designer needs to incorporate instrumentation to enable the operators and management to confirm that the facility meets the requirements of safety, stability and environmental compliance whilst continuing to meet its operational function. This paper reviews the use of inspection and monitoring to prevent failure and provides, in summary, examples of where monitoring has been undertaken and prevented failure. A tabulated review of legislation, regulations and technical guidance generated as a result of previous failures or untoward events is provided.

Jan 1, 2005

By Fredrick J. McGarry, Fred Moavenzadeh

In order to reduce the resistance of hard rocks to excavation by the available cutters, and in order to reduce the wear of the cutters and thus to increase the efficiency of the conventional tunneling machines and to improve their rate of advance in hard rocks, surface active agents have been studied in the controlled laboratory environments and have been experimented with in the field of actual tunneling operations. The laboratory study utilized a notched beam test to measure the effectiveness of 30 different surface active agents on the amount of surface energy required to cause stable fracture in the rock specimens. In addition to the type of rocks, geometry of the specimens, type of surface active agents, the effects of environmental factors such as temperature and degree of saturations were studied. The principle mechanism responsible for weakening of rocks in the presence of surface active agents was believed to be that of stress activated corrosion. The application of surfactants in the laboratory resulted in a reduction of up to 50% in rock strength, while limited experiments in the field have shown a 10 to 20% increase in the rate of advance. Substantial efforts both in the laboratory and in the field are needed to fully understand the mechanism involved and to utilize the effectiveness of surfactants as weakening agents for hard rocks.

Jan 1, 1997

By A. Iordanidis

Ieropigi Fe-Ni deposit is located in northern Greece, in Kastoria?s prefecture near Albania?s borders. A laterite profile exposed by opencast mining and placed near the basement of the mine was chosen for sampling. Ten samples, representing ?pure? Fe-Ni ore, ophiolite-bedrock, ?red? ophiolite and ?enriched? ophiolite were collected, based on macroscopic observation. The samples were analysed by X-Ray powder Diffraction (XRD) for the determination of the mineral phases. Iron and Nickel concentrations of the same samples were also determined by X-Ray Fluorescence (XRF) methodology. The Fe-Ni ore samples consists predomi-nantly of hematite, goethite, chrysotile (serpentine), saponite and quartz. ?Red? ophiolite samples consists mainly of chrysotile, hematite, quartz, calcite and small amounts of paragonite and goethite. Bedrock and ?enriched ophiolite samples are comprised of chrysotile, hematite, calcite, quartz, nontronite, paragonite and sepiolite. The Fe content ranges between 13% and 49% in Fe-Ni ore samples and between 7% and 10% in ophiolite samples. The Ni content ranges between 1% and 3% in Fe-Ni ore samples and between 0.3% and 4.5% in ophiolite samples (the higher content is observed in the so-called ?enriched? ophiolite). A correlation between Fe-Ni concentrations and mineral assemblages is being attempted in order to evaluate the nickel-rich mineral phases. It is shown that the nickeliferous minerals are serpentine, hematite, goethite and saponite.

Jan 1, 2006

By Lyman H. Hart

Michiquillay is a major copper deposit situated about 70 km (44 miles) northeast of Cajamarca in northern Peru (Fig. 1). It occurs largely in quartz monzonite in a shatter zone produced by the intersection of northwest- and northeast-trending fault systems. The geology is described by Hollister and Sirvas (1974). The ore deposit was drilled in 1963-1964 by Asarco, with results indicating a resource of 575 Mt (634 million st) averaging 0.72% Cu-0.03% Mo. A decision to bring the project to production has been deferred because world capacity to produce copper has been more than adequate to serve world requirements since 1964. The chain of events leading to the discovery of Michiquillay was set in motion by a decision by the management responsible for the Chilete mine to undertake an investigation of the surrounding area for a possible supplementary source of mill feed to bolster its decreasing lead-zinc reserves. Chilete, located 100 km (62 miles) inland from Pacasmayo, is the most northerly, important mining project in Peru, and little was known of the mineral potential north and east of that point. Relatively unimpressive reports of Colonial silver mining had attracted little investigation, and little was known of the regional geology. One study by Benavides (1956) had been made of the Cretaceous in the region of Cajamarca.

Jan 1, 1991

By C. L. Schaidle

Computers and communications technology have revolutionized many industries. In earthmoving, this revolution has just begun. The major changes are still to come but are just around the corner. This article presents a vision of this upcoming revolution in the earthmoving industry. This vision includes basic communication, machine monitoring and diagnostics, job and business management, planning and operations, and machine control. For the industry to realize maximum benefits from the information age, it must have a vision of the future that is shared by technologists, machinery producers and miners. History of the information age During the 1970s, the information age revolutionized retailing, starting with the grocery business. By using bar codes and laser scanners, individual items were connected to the information world of computers and communications. Once this connection was made, the data on what an item was and where and when it moved became valuable information for inventory control and marketing. Not only was the information valuable, it was easy and inexpensive to collect and manipulate. The effort required to provide the consumer with the right product at the right time was reduced. Costs went down and customer satisfaction went up. But there were casualties of this revolution. Many inventory counters lost their jobs. There were also equipment casualties. Cash registers that could not interface with the digital world were discarded long before the end of their useful life.

Jan 1, 1994

By Jack Parker

Pillar design ought to be a matter of matching pillar strength against pillar loading conditions to come up with dimensions that will cause the pillar to behave as required - to stand forever, or for a limited time, or to fail in a controlled manner. But something has gone wrong in recent years, seriously wrong. Pillar design practices have departed from reality. The laboratory and computer have taken the place of the mind. If the mind is put to work and such pillar design practices are examined critically, it becomes obvious that they deal with conditions that do not exist in a real mine. Time, effort and money are wasted, and the end results are incorrect and misleading. These questionable pillar design practices have been passed on from generation to generation over the past 30 or 40 years. The hope here is that we can return to reality as soon as the questionable practices are generally recognized. We could then go back into the mines and define how pillars really work, and mining safety, efficiencies and costs should improve. Let us take a close look at current textbook design practices and compare them to real-life conditions. Common sense might then prevail. (At this time, we will consider only mine pillar design, but similar scrutiny should be applied to design criteria for width of openings, floor conditions and roof support systems.)

Jan 1, 1993

By Vincent Castronova, Jane Y. C. Ma, Joseph K. H. Ma, Chang-geng Mo, Carl J. Malanga

A group of bisbenzylisoquinoline (BBIQ) alkaloids has been shown to exhibit various degrees of effectiveness in preventing silicainduced fibrosis in animal models (1, 2). The capability of these drugs to interact with phosphatidycholine vesicles and rat alveolar macrophages was studied using fluorometric and equilibrium dialysis methods. The lipid binding affinity of these alkaloids was found to depend upon several structural factors including hydrophobic substitutions, the chiral configurations, and the double oxygen bridge-restricted conformation of the benzylisoquinoline moieties. Tetrandrine, which is highly effective in preventing fibrosis, showed strong binding to both lipid vesicles and alveolar macrophages. In contrast, curine and tubocurine, which have little or no effect on silicosis, exhibited only weak binding to lipid vesicles and almost no binding to cells. The moderate binding affinity of fangchinoline to vesicles and cells corresponded to a moderate effectiveness of the compound as an antifibrogenic agent. Methoxyadiantifoline, an alkaloid of unknown antifibrogenic potential, also exhibited high binding affinities for lipid and cells. In summary, this study indicates that alveolar macrophages exhibit large binding capacities for certain members of the BBIQ alkaloids. A positive correlation was observed between binding affinity to alveolar macrophages and the reported antifibrotic potency of these compounds, suggesting that the ability of these drugs to interact with alveolar macrophages may be a key step in inhibition of the progression of silica-induced pulmonary disease.

Jan 1, 1991

By Louis M. Schlesinger, Joel F. Hutwelker

A pilot plant was constructed to attrition scrub and deslime North Carolina phosphate ore. A 15.2 cm (6 in.) diameter pilot-scale Linatex Hydrosizer was used for desliming. Four tests were run at a throughput of 2.9 t/hr/m2 (0.30 ton/hr/f t2) and a teeter bed density of 1.12-1.13 g/cm3. The teeter water flow rate was varied to create an upflow velocity ranging from 0.25-0.47 cm/s. (This latter value is equal to the terminal settling velocity of a 200 mesh (74 micron) particle having a density of 2.7 g/cm3.) The recovery of +200 mesh oversize to the sizer underflow was 95%-96%, regardless of the teeter flow rate. The recovery of -200 mesh undersize to the sizer overflow increased from about 90% at 0.25 cm/sec to 100% at 0.47 cm/sec. A metallurgical balance that was conducted for one of the tests showed that less than 2% of the phosphatic value was lost to the overflow. A fifth test was run using an upflow velocity of 0.47 cm/sec; however, the teeter bed density was maintained at 1.05 cm/sec. This change had the effect of reducing the cut point (d50), while maintaining complete slime removal.

Jan 1, 1993

By S. A. Kupferman

Cement raw materials used in southern California and Arizona cement plants occur in a wide variety of geologic environments. Limestone, the primary material making up 75-80% of the raw materials mix used in cement manufacture, is mined solely from Paleozoic Era deposits. During this time, 570-245 million years ago, vastly more limestone was formed in California and Arizona than at any other time (continent was in southern latitudes). The general geology of each limestone deposit associated with the eight southern California and two Arizona cement plants will be reviewed. Many of these deposits are geologically complex, having been subjected to Mesozoic deformation, metamorphism and instrusive events, along with Cenozoic uplift and deformation. All of the cement plants in this region require the use of secondary additives in order to meet the chemical requirements for cement. The kinds of additives vary greatly, depending upon the chemistry of the limestone deposit, cement specifications, and the cement manufacturing process. In some cases, clay deposits with suitable chemistry on or near the limestone sources are utilized. Where alkali limitations or need for alumina-rich materials exist, then specialty clay deposits are used. Iron-rich sources in the region include iron ore deposits and slag or mill scale from local steel mills. The geologic setting and location of these additives, along with gypsum and silica-rich deposits presently in use, will be mentioned.

Jan 1, 1987

By Daryl R. Tweeton

The Bureau of Mines has developed PC-based tomographic computer programs for geophysical applications, and is making them available to potential users as part of a Bureau plan to facilitate the transfer of technology to industry. These programs were initially developed for predicting and monitoring the flow pattern of leach solution during in situ mining. Seismic crosshole tomography has located fractured zones of low seismic velocity and saturated zones of high seismic velocit above the natural water table. The Bureau has applied tomography also to assessing high-wall blast damage, and examining the integrity of mine-related geological structures. The tomographic programs include the straight ray pro ram BOMTOM (Bureau of Mines Tomography? and the program BOMCRATR (Bureau of Mines Curved Ray Tomographic Reconstruction) that can be run with either curved or straight rays. Both programs provide optional mathematical constraints to help counteract the nonuniqueness of reconstructions that occurs with crosshole data. They are available from the Bureau free of charge. With typical data sets, they can be run on a 386 microcomputer with 640 Kb RAM. BOMTOM was described in the Bureau Report of Investigations 9159, so this report emphasizes BOMCRATR. Curved and straight ray reconstructions using seismic field data collected while simulating injection of leach solution were compared. The curved and straight ray tomograms were quite similar even with velocity variations of more than 50 percent.

Jan 1, 1992

By Carl D. Broadbent

The goal of slope mechanics is to achieve efficient and safe pit slopes. In the case of ultimate mine slopes the objective is to minimize the stripping ratio by steepening the slopes to their maximum stable limit. On the other hand, the inclination of working slopes is normally established by operating space requirements and the function of slope mechanics should be the continuous surveillance of slope conditions to assure long-term stability and to provide timely warning of impending slides so that corrective measures may be incorporated in production planning. In 1956 Kennecott instituted a program to provide the resources necessary to develop and maintain a slope mechanics capability, including technical knowledge, engineers, and instruments. Although the goal has not been fully achieved, very significant progress has been made. Other papers (3, 4, 7, 8, & 12) have described the purposes and detailed results of the most recent study carried out at the Kimbley Pit. The study has provided valuable experience and has given the industry a three dimensional picture of the stability variables that influence a mine slope. The study has better defined those areas which require further investigation and has indicated those design and analytical techniques which can be used with confidence. In effect, the study has paved the way for the next step; that of refining the design process and applying it for profit.

Jan 1, 1968

By G. Brooker

The back-filling operation at Olympic Dam mine supports a 9.8 million tonne per year (Mtla) underground Longhole Open Stoping operation. It involves a combination of Cemented Aggregate Fill (CAF), Surface Rockfill and Underground Rockfill. The fill material is introduced from the surface via bore-holes into the top of the stopes. Due to the deleterious effects of the dust and the thickness of the atmosphere in the stope it is not possible to view the filling process. The need for a means to measure the filling process is driven by the stopes, which have both rock and CAF fill placed in them. The large volumes and high relative cost of CAF makes the minimization of CAF fill a business priority. Incorrect filling can lead to the sterilization of ore reserve. The use of millimeter wave radar to view through the haze to give real-time data on the relative positions of the different fill masses is being developed as an in-stope fill control device. This paper reviews the salient points of the technology and its potential use in the back-filling operation at Olympic Dam. The broader potential of this technology for use in the mining industry by 'seeing' through water and dust laden environments such as that found in stopes and ore passes is discussed.

Jan 1, 2001

By Qiong Wu, Li Guo, Qinghua Gu

BP neural network is a multilayer feedforward network trained by error back-propagation algorithm, which is one of the most widely used neural network models. However, BP neural network has exposed more and more shortcomings and deficiencies with the expansion of the application scope. In the prediction of ore quantity, BP neural network has the characteristics of slow convergence and easy to fall into local minimum point. In order to obtain the global optimal solution, and to improve the defects of BP neural network, this paper proposes combination optimization algorithm of genetic algorithm (GA) and BP neural network to improve the speed and accuracy of forecasting the main design flow chart and the analysis of the sort distinguish algorithm are offered, and then some problem in the design and debugging of the algorithm are discussed. On this basis, the GA-BP neural network model is constructed and applied to optimize the initial weights and threshold value of BP neural network. This model choices the floating point coding method to encode the connection weights and thresholds, and divides subjects into several populations. Through the introduction of selection, mutation, crossover, initial weight and other operators, making operational synergies between the various groups. This study selects 30 geological units, 8 quantitative variables (Pb, Zn, Cu, Mo, Si, Ni, Co, V) and 12 qualitative variables to carry out empirical analysis. Then the simulation of the algorithm is carried out in MATLAB and the parameters are analysed. By normalizing the input samples, 22 groups of observation data are used as the training data for prediction, and the latter 8 groups of observation data are used as the test data to be verified. The results show that when the ore quantity characteristics are not very significant, the model will produce prediction bias. But the improvement of the algorithm increases the efficiency of the function approach capacity of BP neural network and conquer the BP neural network system’s instability. It provides an auxiliary guide for ore prediction, which have higher reference value.

Jan 1, 2017