Search Documents

By J. Kretschmann, H. Mischo, O. Drusche, S. Krause

Mining is and will be inevitable in spite of the increasing implementation of circular strategies. The energy transition and population growth will boost the demand for critical raw materials. It is the opportunity for miners to reinvent their business models. Generating attractive new business cases with sustainability-oriented innovations meets investor and stakeholder expectations from the beginning. This contribution guides managers on their pathway towards sustainable success. Contextual sustainable business model patterns will be presented that can be adjusted individually, regardless of actor’s size. It will also point out, how sustainable value creation can be measured on business model level.

Feb 1, 2023

By B. J. Arnold

Froth removal rate is perhaps the most important parameter in scaling up from laboratory tests to plant froth flotation circuits. The WEMCO froth factor, the percentage of the froth height over the lip of the cell, accounts for the difference between laboratory and plant froth removal rates. As laboratory tests are generally run with a high pulp level, laboratory results are directly comparable to plant results only when a high froth factor (high pulp level) is run in the plant. When a plant does not have a high pulp level, laboratory results must be modified by using a ratio between laboratory and plant kinetics to predict plant results. This ratio, which accounts for the froth factor is important in testing any new reagents or reagent dosage levels. CQ Inc. researchers have used predictive scale-up equations using this froth factor coupled with equations for kinetics and the residence time distribution in the froth flotation cells. These equations show that modifying the froth factor in the plant can have a large impact on flotation cell yield and reagent use. EQUATIONS DESCRIBING FROTH FLOTATION Scaling up from laboratory- to plant-scale coal froth flotation has been the subject of much research and debate over the years. Determining the proper rate equation to describe flotation kinetics and an equation to describe the residence time distribution in a bank of flotation cells have been particular areas of study. As shown in Tables 1 and 2, various rate equations and residence time distribution equations have been proposed in the literature. These equations consist of one or two parameters giving the rate constant(s) in units of time-1 and a parameter for recovery at infinite time, [Roo]. Dowling, Klimpel, and Aplan (1985) evaluated the rate equations in Table 1 for their ability to fit froth flotation kinetic data and found that all of the equations give reasonably low standard errors. Several of these equations, however, proved to have superior confidence limits for the parameters. Other equations have also been proposed to describe the recovery versus time profile. Most recently, Diao et al. (1992) proposed a first-order form with a sine wave distribution. As indicated in Table 2, investigators have suggested many equations to describe the residence time distribution in continuous froth flotation. To measure the water residence time distribution in a bank of cells, a soluble tracer is often used, with the level of tracer being measured over time at the tailings discharge. This response is then put on a dimensionless basis as described by Austin et al. (1984) in order to compare mixing behavior between cells or under different conditions. Coal flotation cells are generally of the flow-through or hog-trough variety allowing for some degree of backmixing and short-circuiting. While the water residence time distribution (see Figure 1) at the end of one cell in a bank of 1.4 cu ft WEMCO froth flotation cells may approximate perfect mixing conditions (Arnold, 1989), the water residence time distribution at the end of a bank of cells does not. Figure 2 shows the RTD for various flotation cell types compared to that for perfect mixers in series. To better describe the water residence time distribution in coal froth flotation cells, Van Orden (1986) used an equation similar to the perfect-mixers-in-aeries equation. This

Jan 1, 1993

By Karan S. Keith, Haydn H. Murray

The sorptive and impervious nature of clay materials was recognized long before modem technology allowed us to understand the physiochemistry of clays. In Cyprus, around 5000 BC, sorbent clays were used for fulling woolen cloth. One of the earliest recorded uses of clays as an impervious liner was more than 3200 years ago in the construction of a bitumen-sand-gravel-clay liner built along the embankment of the Tigris River at Assur. Today one of the most common applications for clay minerals in the environmental industry is for use as impermeable barriers to fluid transport. Clay minerals are fine in particle size and thus have large surface areas, which, as a result of intermolecular forces and ionic substitutions, make them excellent materials for both ion exchange and ion sorption. Therefore it was logical to consider certain clay materials for applications in the environmental industry, where they can act as physical and chemical traps for potential contaminants. Certain bentonites have historically been used as sealants for fluids (Grim, 1962), and other clays have been used as sorbents for both water and oil (Haden, Jr., 1972). Prior to the 1980s, disposal of the nation's municipal, commercial, chemical, and nuclear wastes was largely indiscriminate and poorly regulated. This created environmental pollution problems, especially with respect to groundwater contamination. Promulgation of federal regulations like the Resource Conservation and Recovery Act (RCRA) and the Comprehensive Environmental Response, Compensation and Liability Act (CERCLA - commonly known as Superfund) have forced a more careful and costly approach to waste disposal. The US Environmental Protection Agency (EPA) is the government body charged with enforcing and ensuring compliance with federal regulations governing toxic waste production, storage, and disposal. Widespread application of geotechnology to the design and construction of waste containment systems and to pollution remediation has become necessary due to such laws and because of increasing use of the subsurface environment for the disposal of hazardous wastes. In the United States nearly 1.8 kg of trash is disposed of per person per day, which amounts to nearly 181 Mt per year. This amount is more than double the waste produced in 1960, and these numbers are expected to continue to increase. Canada produces approximately 1.68 kg of waste per person per day. The EPA estimates that in the United States 64% of our garbage is landfilled, 18% recycled, and 18% incinerated. Even with continued increases in recycling and incineration efforts, the EPA projects that the United States will still need 82% of today's landfill capacity in the year 2000. Data on the exact number of landfills in the United States and Canada is conflicting in part because of changing definitions of what constitutes a landfill and because of inadequate record- keeping by some states and provinces. An estimate of the number of municipal solid waste (MSW) landfills in the United States in 1990 to 1991 ranged from 4 462 to 10 467 with an average estimate of 6 600. In Canada the number of MSW landfills reported varied from 5 493 in a 1982 survey to 1 766 reported in a 1991 National Solid Wastes Management Association (NSWMA) survey (Repa and Sheets, 1992b). A properly operated MSW landfill on average requires 1 hectare of land per year per 25 000 people (Coates, 198 1). In addition to MSW facilities, there are also waste disposal facilities needed for hazardous waste and low level and high level radioactive waste. In 1983, approximately 266 Mt of hazardous waste was generated in the United States, almost 50% of this by industrial production. About 68% of the hazardous waste generated in the United States and Canada and almost 50% generated in Europe are still landfilled (Egger, 1987). In 1985, approximately 51 000 m3 of commercially produced low level radioactive waste was disposed of in radioactive waste disposal sites across the United States (Jungling and Greeves, 1989). By 1987, the US defense program had produced approximately 10 000 t of high level nuclear waste. In addition, there were approximately 14 000 t of spent fuel assemblies from commercial nuclear power plants in temporary storage; the amount is expected to nearly triple by the year 2000. Presently most commercial spent fuel assemblies are stored in water-filled pools at nuclear power plants. Defense waste is stored on federal reservations in surface and underground steel tanks (Anon., 1987). Under federal law nearly all waste disposal and containment sites are required to have a liner or barrier and a cover system design that meets federal- or state-approved hydraulic conductivity requirements for the type of waste that the site will receive. When approved materials, necessary for constructing the waste containment structure, are not available on site, one of the most economical ways to meet these requirements is through the use of clay, soil/clay, or clay/geotextile materials. In a 1991 survey by NSWMA of its members and other select private sector facilities, 81% of the surveyed landfills had some type of liner system. About 75% of these liners were clay materials (either recompacted or natural). The remaining 30% were using flexible membrane liners as part of a liner system. Clay liner materials are most commonly used in the South, Midwest, West-central, and South-central regions of the United States. About 93% of the landfill facilities had closure plans, and of these 91 % were using clay materials in some way for the final cover material (Repa and Sheets, 1992a). Waste barriers and containment structures take several basic forms, and they are used in a variety of different applications: 1) cutoff walls, 2) soil or soil/clay-admix liners, 3) geosynthetic bentonite-clay/geotextile membranes, 4) radioactive waste containment barriers, and 5) synthetic geomembranes. This chapter will discuss the first four of these categories. The field of synthetic geomembranes, although a related topic, is beyond the scope of

Jan 1, 1994

By Herbert H. Einstein

Swelling ground is one of the most feared problem in tunneling. Although usually not causing sudden failures during construction, the very large and long- lasting deformations give contractors and designers the impression of facing an inevitable phenomenon against which there is no technical defense. Examples are concrete liners of up to 1 m that have been cracked or crushed, tunnel inverts that are heaving at a rate of 0.1 m per week during construction and continue to heave during decades. This author (Einstein (1979). Einstein et al. (1972) (1975)) and others (Huller (19761, Grob (1972) (1976). Golta (1967), tienke (19761, Krause (1976). Lee et al. (1976)) have described many cases of swelling tunnels that illustrate the significance and magnitude of the problem. This paper intends to show that tunnels in swelling rock can be built to perform satisfactorily. . For this purpose the tunnel builder needs to know how to determine the existence and specific behavior of swelling ground, how to select appropriate structural characteristics of the tunnel and predict its performance under the specific ground conditions, and finally how to construct the tunnel such that the anticipated performance will be achieved. This concept of a design-construction procedure is the same as that for tunneling in general; it is mentioned here for NO reasons: 1. To point out that design and construction cannot and should not be separated. The way in which the tunnel will be built has to be an integral part of the design. 2. This paper will be structured corresponding to the procedural concept.

Jan 1, 1981

By R. N. Torbin

The degree to which automation, remote control and robotics have been used in the mining industry is far less compared to general industry. Reasons include the adverse working environment, the difficult work operations and the generally lower incentive for capital investments in new mining equipment. The author and his employer has been involved in the development of advanced mining and underground construction equipment. This equipment is designed to he remotely operated, to improve both productivity and worker safety. Many of these machines employ advanced electronics and robotic features. The types of machines that will be discussed include remote shaft lining and placement, remote excavation and mining operations, and automated conveyor systems.

Jan 1, 1991

By Howard S. Avery

The wear resistance of an alloy in heavy mill liner sections may be quite different from that of the same alloy in small balls and other light sections. This fact was established by means of a well standardized laboratory test that was validated for ball mill service. The salient properties of mill liner alloys, including austenitic manganese steels, medium alloy hardenable steels, and abrasion resistance cast irons are described, critically compared, and quantitatively ranked in terms of mill liner service. Where they are tough enough the high-chromium and other martensitic cast irons (e.g. Ni-Hard) promise maximum economy, but their section sensitivity is a limitation. To overcome this an extensive research program studied over a hundred experimental alloys, ranking them for heavy section abrasion resistance by means of the laboratory test. Computer analysis of the results provided a parabolic formula in terms of composition, the apexes of the parabolas providing optimum wear resistance. From the extensive family of parabolas one apex was selected for intensive study in experimental liners, pilot plant production, development of quality controls, and finally full scale production. This is a 19% chromium iron critically alloyed with nickel and molybdenum and hardened by heat treatment. It has excellent resistance to various types of abrasion, but its chief advantage is in heavy mill liner sections, where it has about 25% better abrasion resistance than the best of the quality controlled chilled Ni-Hards. It is also tougher. After production problems were solved, a number of ball and rod mill tests were arranged on a guaranteed performance basis. Two of these, representing about 138,000 pounds of mill liner castings, have recently confirmed in service the superiority that was predicted by the laboratory tests.

Jan 1, 1969

By Duk-Wom Park

INTRODUCTION he most important information required for the engineering design of structures in rock is its inherent rock properties. Without providing adequate parameters on the rock property, high quality of the engineering design cannot be expected. Many laboratory rock testing methods have been developed, but they are mostly for testing intact homogeneous rock without appreciable discontinuities. In reality, the intact rock properties do not represent the behavior of rock masses because rock masses include irregularities such as joints, cracks, interlayers, variations in mineral compositjon, etc. In general the type and intensity of the rock defects may be much more important than the type of rock which will be encountered (Terzaghi, 1977; Bieniawski, 1984). In-situ rock testing can provide better information on rock parameters, but it takes a tremendous amount of effort, time and cost. In recent years, numerical modeling techniques are gaining popularity as a design tool for structures in rock bodies due to the enormous progress in computer technology (Wang, 1985; Kripakov and Melvin, 1983; Park and Ash, 1985), but it is quite dangerous to use numerical techniques without having proper input data. Bieniawski (1984) states as follows: Some design methods such as numerical techniques have outpaced our ability to provide the input data necessary for the application of these methods. In order to provide an adequate amount of input data, an extensive testing program may be required, thus major efforts in time and expenses are commonly spent, In this paper, a new rock testing method, which utilizes the principles of laser holographic interferometry, is introduced. This method requires only a few minutes to measure the modulus of elasticity and quality of a rock core sample, which does not have to be cut or ground. The Poisson's ratio can also be measured in a few minutes from a sliced rock specimen. HOLOGRAPHIC INTERFEROMETRY Principles of Holographic Interferometry In 1948, Gabor invented holography, which is the technique of reviving three dimensional images using a monochromatic light source. A full demonstration was not made then because a clean monochromatic light source was not available. Later, in the early 19601s, the laser was adopted as a monochromatic and coherent light source (Leith and Upatnieks, 1963). Since then, applications have been made in many different fields ranging from crime prevention to three dimensional television. In this technique, an image is usually recorded by means of constructing a hologram which is a record of the interference pattern formed on a high resolution photographic system as shown in Figure I. A beam splitter divides the laser beam into two beams: (1) an object beam which is expanded while passing through a beam expander and illuminates the object, and (2) a reference beam which is expanded and illuminates the holographic plate. Subsequent to an exposure, the holographic plate is developed and illuminated by the expanded reference beam. The original scene of the object is revived in a three-dimensional image. The entire instrument should be placed on a vibration free environment. The wave fields reflected by the object in two slightly different configurations can be superimposed to form an interference fringe pattern. Thus, it is possible to measure the displacement that took place between the two configurations. This method is called holographic interferometrv. There are two methods of achieving this phenomena. The first is the double exposure method, wherein two exposures of the object are made prior to developing the film. The first exposure represents the original condition and the second, the disturbed condition. The second method is the real time holographic method. In

Jan 1, 1986

By T. W. Camm

"Search Amazon.com using the keyword “leadership,” and you are rewarded with a list of over 144,000 titles. There is no shortage of books and articles, many of them well written, with excellent ideas. So why is there still a constant cry for effective leadership in organizations? Often, the person in charge has personal blinders that prevent them from seeing or understanding how to implement the newest leadership idea or method. Either deliberately or subliminally, there is a disconnect from learning about leadership, and actually modelling and implementing what has been learned. Some current research in leadership theory can provide insight and tools to address this issue. INTRODUCTION Let’s start with a discussion of terms. Pseudo is defined by Webster (2015) as “Being apparently rather actually as stated.” Synonyms include affected, assumed, contrived, fake, false, phony, pretended, and artificial. Antonyms for pseudo provide additional clarity, especially when we consider terms normally associated with effective leadership: genuine, natural, unaffected, unforced. A similar type from my world of academia is a pseudointellectual, “A person exhibiting intellectual pretentions that have no basis in sound scholarship” (Dictionary.com, 2015). This is a person pretending to have an interest in intellectual activities for status; a fraudulent intellectuality. The thought of applying the term pseudo to leadership first occurred to me while reading Power and Innocence by Rollo May (1972). In discussing the dynamics of power, and how the innocent are often victims of the exercise of power, especially the more aggressive and destructive types of power, he saw two types of innocence. The first he called authentic innocence, which he described as a childlike clarity that carried into adulthood. The second type he called pseudoinnocence, a type of naïveté that makes a virtue of powerlessness, weakness, and helplessness; being more childish than childlike (p. 48-50). Just as there is the potential for pseudoinnocence and pseudointellectuals, it is also possible to experience pseudo leadership. (I hope the gentle reader is not troubled by my decision to use pseudo leadership as two separate words, in contrast to the use of compound words for pseudoinnocence and pseudointellectual cited above. Since there is no official definition of pseudo leadership, I have exercised my author’s prerogative to treat it as two words, both because I think in this particular application this makes the most sense, and also I think it is easier to read). As we will see, there are numerous theories applied to leadership. What is so troubling to many of us in organizations, there are a myriad of “leaders” in companies espousing pet theories, but very few instances where the actions of these individuals match their words. This is a particularly important problem when we look at the effect this has on the day to day operations of a mine, and where effective leadership (or its absence) can have a profound impact on both productivity and safety."

Jan 1, 2016

By R. J. Zagrocki

Rekindled public concern about lead toxicity has intensified the awareness of lead-bearing materials in contact with the environment and the human population. Lead's common association with other metals makes the increasingly restrictive lead standards a primary concern for a large number of metal and metal-related industries. In commercial scale demonstration tests with lead-bearing wastes, the HRD Flame Reactor process has a demonstrated ability to separate and recover over 95% of the lead as a recyclable oxide, while rendering the remainder of the material as vitrified, non-leaching, and nonhazardous co-product. Recent Flame Reactor process tests and preliminary cost data for secondary lead soda slag are highlighted.

Jan 1, 1992

By D. A. King

Monitoring devices have been installed at several mine sites to assess the potential for the development of acid rock drainage (ARD) in waste rock, spent ore, and tailings material. In addition, instrumentation has been designed for monitoring changes in water quality in pit lakes. These devices monitor temperature, oxygen, water movement, and water chemistry in the waste rock pile. This information can be used to evaluate the current and future ARD risk, the effectiveness of caps and covers, and provide information to aid in the design of other reclamation strategies that limit ARD development. The instrumentation includes nested thermistors and oxygen sensing ports, neutron access tubes, heat dissipation units (HDUs), and lysimeters. Many of the monitoring devices can be linked to a datalogger and telemetry systems, enabling monitoring data to be downloaded from the office.

Jan 1, 1997

By J. M. Dean

The MANpower SUPport Program (MANSUPP), a part of the Mine Management Support System (MMSS), allows an underground coal mine site user to determine manpower requirements for production support work by considering ongoing operations and current mine conditions. The objective of MANSUPP is to plan a detailed manpower structure for a mine by incorporating a production schedule, condition-modified work standards and management preferences to predict required manhours for various activities, for example, belt moves. The program is written in Quattro Pro version 3.0 (Lotus 1-2-3 compatible) utilizing the command language, macro and formula capabilities.

Jan 1, 1992

By P. C. Miller

The paper reports laboratory results from the bioleaching of chalcopyrite ores and concentrates using a mixed moderate thermophile culture. Copper dissolution in excess of 95% was achieved in all cases with the majority of tests indicating copper dissolution in excess of 98%. Aspects of downstream processing are discussed in relation to iron rejection by precipitation. The laboratory results indicate the need for careful control of the pH value to prevent co-precipitation of the copper with iron. A comparison of capital and operating costs is given indicating potential differences between the use of toll smelting and bacterial leaching on site. A discussion is given highlighting the sensitive cost areas of these two process routes.

Jan 1, 1997

By Edoardo Misano, Magali Schivre, Martino Semeraro, Alain Bochon

"INTRODUCTIONThe Laboratoire Souterrain de Modane (LSM) is an underground research laboratory located in the Western Alps on the French – Italian border. It is located in the middle of the 13km long Fréjus highway tunnel that links Modane (France) to Bardonecchia (Italy) in correspondence of the highest overburden of 1800m. The LSM current activity is mainly based on the investigations about the dark matter and requires very sensitive instrumentation which shall be protected from cosmic rays. To comply with the new legislation about safety in tunnels, the highway tunnels owners agreed in 2007 the construction of a parallel safety tunnel, at an average distance of 50m from the existing tunnel and the laboratory owner CNRS (Centre National de recherche scientifique) also decided to extend the existing laboratory with the construction of a new 17000 m3 cavern allowing the installation of wider and more powerful instruments that could increase chances of success of research. Preliminary studies stated that, considering both geological and functional issues, the most appropriate location for the new laboratory shall be in between the two tunnels and parallel to the existing laboratory (see Figure 1).Hence, the CNRS asked Systra, already in charge of the works supervision of the safety tunnel, to realize the design of the new cavern. The laboratory extension includes the construction of a 40 m long cavern with 300 m² cross section and two smaller tunnels connecting the safety tunnel and the existing laboratory to the new cavern. The minimum distance between the front face of the new LSM and the highway tunnel is about 25 m.The French and the Italian authorities, responsible of the exploitation of the Fréjus highway tunnel, asked the CNRS to develop an accurate study indicating that the construction of the laboratory extension would not have any negative impact on the existing structures of the highway tunnel. The main results of the developed study by Systra are summarized hereafter. The construction works should begin prior to the end of the construction works of the Fréjus safety tunnel in 2018."

Jan 1, 2016

By Charles F. Herbert

It is difficult to decide who most deserves credit for discovery of a mining district that has produced over 19.8 Mg (7,000,000 fine oz) of gold and still contains a large reserve of low grade ore. Fred Culver may have made the first discovery. In the summer of 1867, the ship's log of the Hudson's Bay Co.'s steamer Otter stated that Fred Culver had been found nearly lifeless from wounds, without provisions, but with some 2.3 kg (80 oz) of rough, placer gold in the bottom of his canoe; he told of finding a rich creek and an attack by Indians that caused the death of his two companions. The place of rescue was about 24 km (15 miles) southeast from the present site of Juneau. Possibly John Muir, the hero of modem environmentalists, should be given important credit. In 1879 he reported finding mineralization along what later became to be known as the Juneau Gold Belt, and it was his report that induced the cantankerous German mining engineer, George Pilz, to grub- stake some of the prospectors who had drifted into Sitka from the declining gold placers of Cassiar, British Columbia. Pilz also offered rewards to Indians who might report gold discoveries.

Jan 1, 1990

By Caroline L. Herzenberg

Recently, neutron-induced gamma spectrometry has been introduced and gained provisional acceptance as a method for continuous automated analysis of coal. Instrumentation based on this technique has been under development for and in testing at coal preparation plants and a coal gasifier. To provide a deeper understanding of, and extended capability in applying this method, theoretical calculations have been performed of the absorption of thermal neutrons by various coal compositions, and the thermal neutron capture prompt gamma emission spectra produced by these compositions have been calculated. Selected results of these computer calculations and the computer generated spectra are presented.

Jan 1, 1984

By K. H. McDaniel

Eight Mine Weather Stations (MWS) installed at the Waste Isolation Pilot Plant (WIPP) to monitor the underground ventilation system me helping to simulate real-time ventilation scenarios. Seasonal weather extremes can result in variations of Natural Ventilation Pressure (NVP) which can significantly affect the ventilation system. The eight MWS(s) (which previously collected and stored temperature, barometric pressure and relative humidity data for subsequent NVP calculations) were upgraded to provide continuous real-time data to the site wide Central Monitoring System. This data can now be utilized by the ventilation engineer to create real-time ventilation simulations and trends which assist in the prediction and mitigation of NVP and psychrometric related events.

Jan 1, 1997

By James M. Neilson

In contrast to the rather generalized title of this paper, the subject of discussion will deal with two specific cases of sliding in glacial lake clays in northern Michigan. The slides in question occurred on lands owned by the Upper Peninsula Power Company a short distance south of Lake Superior. Slope failure problems have plagued the Company's operations for many years. It is hoped that this paper may provide some insight into the causes and possible mechanisms of clay slides in general as well as providing details of the two slides specifically cited. LOCATION The Upper Peninsula Power Company's Victoria facilities lie in section 29, T. 50 N., R. 39 W., in Ontonagon County, Michigan. The power site is situated on the West Branch of the Ontonagon River, approximately five miles southwest of the town of Rockland and only one mile to the southeast of the old mining location of Victoria. The area lies wholly within the Precambrian Shield and it is geologically similar to adjacent parts of Wisconsin, Minn¬esota, and Ontario. (Figure 1). THE VICTORIA HYDROELECTRIC SITE The Victoria hydroelectric power facilities were constructed in 1930 on a portion of the West Branch where meanders are conspicuously developed. The generating station is actually sited on an abandoned meander at a place where valley widening has occurred. Water is conducted from Victoria Dam through a 6000-foot, 10-foot diameter wood stove pipe - through a surge tank - and into the powerhouse where

Jan 1, 1971

By M. J. Virnig, Marc Le Vier, Rong Yu Wan, Jan D. Miller

Although most gold production from primary resources is achieved by cyanidation, many research activities have focused on alternative processes using noncyanide lixiviants. Such research has been prompted by environmental concerns and problems associated with the processing of refractory gold ores. Numerous research projects have been conducted to search for a noncyanide hydrometallurgical process for gold extraction. In this paper, a technical review of gold recovery from noncyanide solutions is presented, and new results are reported. Gold recovery techniques that are considered include activated carbon adsorption, cementation, solvent extraction, ion exchange, and reduction-precipitation.

Jan 1, 1993

By R. Hogg, S. Chander

An overview of various physical and interfacial characterization methods for use in mineral processing is presented in this paper. The availability of a large number of characterization techniques and commercial instruments requires that the reliability and limitation of each be carefully examined. The choice of a method may depend upon the manner in which the information is to be used, for example, in development of new processes, plant and process design, plant control, problem solving, etc.

Jan 1, 1987

By Xiaodong Yu, Jinxia Zhang, Fusheng Niu, Hongmei Zhang

The three-dimensional (3D) fractal dimension is an important parameter to analyze the 3D structure and flocculation effect of the hematite flocs. In this work, the 3D fractal dimension of hematite flocs was predicted by establishing the particle swarm optimization (PSO) algorithm optimized with the back propagation (BP) neural network (PSO-BP). The proposed model considers four factors, namely, the flocculant quantity, flocculation time, stirring speed, and temperature, as the input parameters. We normalize the input data during the preprocessing stage. The BP neural network was optimized by using the PSO algorithm for realizing the 3D fractal dimension. The prediction accuracy and effectiveness of the model were evaluated. The experimental results showed that the predictions performed by the proposed PSO-BP neural network were better than that of BP neural network. In addition, the root mean square error (RMSE), mean squared relative error (MSRE), mean absolute error (MAE), and mean absolute relative error (MARE) of the proposed model are lower, compared with the errors of the BP network. Similarly, the measurement coefficient R2 of the proposed model is higher, compared with the BP network. The maximum absolute error of the model is 0.0772, the maximum relative error is 0.02405, and the regression coefficient r is 0.98592. These results showed that the proposed model has a good performance and high prediction accuracy. When verifying the practicability and the effectiveness of the proposed PSO-BP model, the prediction results of 10 groups of hematite flocculation experimental data under different conditions showed that the prediction value of the proposed PSO-BP model was closer to the ground truth as compared to the BP model. Therefore, the proposed model is suitable for predicting the 3D fractal dimension of hematite flocculation.

Oct 3, 2022