Search Documents

By Luis Bergh, Juan Olivares, Juan Durán, Juan Yianatos

"In the last decades heuristic approaches based on logic rules have been one favored tool to control flotation units and plants on top of distributed control systems. Fundamental knowledge on flotation processes, expressed in reliable models, and programs to assure the quality of the information, contained in available measurements, are still insufficient to develop robust predictive control strategies. In this paper the idea of using simplified models, when a known disturbance is coming into the process, is explored. As a first approach, the set points of the froth depth and air flow rate controllers are estimated based on this model. Since these target predictions are only approximate, a conventional feedback expert system, based on logic rules, continues the task of calculating the set points of the distributed controllers, in order to decrease the gap between actual and desirable metallurgical targets. This combined actions permitted a fast response of the process to change the operation of the circuit when a measured disturbance is detected. This control system was implemented and experimentally evaluated in a pilot rougher flotation circuit, with three pneumatic cells, with froth depth and air flow rate controllers in each cell. Since the circuit is operated for the air-water system, a phenomenological model, with parameters estimated from industrial data, is fed on-line with real operating variables and virtual values, characterizing the feed (grades, solids%, particle size distribution), in order to predict the metallurgical targets (recovery and concentrate grade). Several cases are discussed showing the benefits of this control system and the possible improvements to work on. INTRODUCTIONToday, large size mechanical flotation cells of 100 to 300 m3 are used in rougher operation in different flotation industrial plants, in Chile and worldwide. However, in spite of the advances in fundamental research and the notable growing in equipment size, with more complete instrumentation systems, there are still a lack of reliable data for industrial flotation modeling and simulation to advance in better control systems design."

Jan 1, 2016

By W. I. van Drunick

AngloGold Ashanti Metallurgy has recently been involved in the successful installation of intelligent, expert control systems on the milling circuits. The benefits of expert systems in these highly interactive circuits are obvious. Existing comminution tools, which have been developed and can now be incorporated into online control, include improved modelling techniques and understanding, robust and competent simulation methods, sampling techniques, and optimization methods. The minerals processing industry has seen the benefits of predominantly research-based centres turning their focus to commercial applications of comminution circuit modelling packages, which in turn can be utilized within expert control systems. Expert mill control has been successfully installed at AngloGold Ashanti South Africa?s Mponeng and Kopanang Gold Plants. Expert control has also been installed at AngloGold Ashanti?s West African Sadiola Mine operation. The implementation of the systems has produced a number of positive outcomes. Plant surveys usually result in the identification of sub-optimal areas, and the control philosophy is then configured to ensure that these areas (along with the entire circuit) are optimized and maintained. Outdated PID loops have generally resulted in oscillatory control with large amplitude deviations from set-point. Significant improvements in SAR Metallurgy operations have been realized, firstly through achieving stability in the load and cyclone control, and secondly through online identification and implementation of optimized set-point conditions for mill mass, discharge density, etc. Soft sensors are configured to accurately estimate parameters that were historically not measurable, viz. breakage rate, mill load condition, onlinekWh/t-75 microns, mill discharge density and cyclone overflow grading. The Mponeng Plant system achieved circuit stability and improved overall grind. The Plant was able to pay back the control installation after 3 months, based on the additional recovery obtained from the finer grind. The system at Kopanang Plant has also resulted in enhanced plant recovery as a result of a finer grind and a lower residue gold content. The payback for this system was calculated at approximately 6 months. AngloGold Ashanti Metallurgy is currently expanding the installation of expert control to its other operations. It is also evaluating the performance of various suppliers of advanced control systems.

Jan 1, 2005

By O. A. Bascur

An expert process operation advisor system can increase plant efficiency, even in a complex, natural environment where not all elements of the production process can be measured directly. This system, develaped and tested in a Frasch process sulphur mine, monitors key process variables with extended qualitative pattern processing and provides on-line advise to the operators of sulphur production wells. An expert system framework captured common sense rules generated from years of experience obtained in the operation of a sulphur well monitoring system. The existent operating procedures were enhanced by process analysis using a qualitative modeling approach which provided extended rules for on-line classification of a well operational status and an expert alarm system. The methodology for development and implementation of the process advisor is presented in this report. Qualitative process modeling, statistical process control and heuristics were incorpomted into the process advisor resulting in a more efficient use of heat, an extended well operational life and the resulting cost savings. The computer process advisor also allowed the operator to concentrate his attention on the weak wells or wells in trouble. A process management system improved plant operation by automatically storing well data into a commercial database for operator validation. The enhanced relational data base system provided valuable and timely information for planning and decision making. The synergistic combination of on-line process analysis tools, expert control systems and database technology is shown to facilitate the flow of materials, energy and infomation in a modern plant.

Jan 1, 1990

It's my privilege to welcome you here this afternoon on behalf of the Sydney Branch of the Australasian Institute of Mining and Metallurgy. This is a most important gathering. You, the delegates, represent what is probably the most influential cross-section of the mining investment community and the oil and gas investment community to meet as a group during the last decade. We are meeting today and tomorrow to review two related issues that will have far reaching implications for the effective use of capital by the natural resources sector in the 1990's. The views which will emerge over the two days will almost certainly 'influence the availability of new capital to the companies in the natural resources sector for the decade ahead. The two principal themes of Minval '89 are, first, the valuation of mining and oil and gas assets and securities, and second, disclosure and reporting requirements in the capital markets which relate specifically to natural resources companies. This conference is taking place at a time when further consolidation and rationalisation appears likely within the mining industry. Although the base metals and other sectors of the industry are enjoying profitability not seen since the early 1980's there is caution about committing to new capacity which could bring about a return to the low prices and excess supply which prevailed during much of the last ten years. These conditions have led to increased takeover activity within the mining industry overseas, and similar trends are as evident here in Australia as they are elsewhere. Recognition of value is therefore likely to become more important than ever before in preserving shareholder wealth, for creating new wealth and for preserving the financial health of the resources industry in the future. Valuation therefore is the first theme of the conference.

Jan 1, 1989

Only West Australian entrepreneurs and Queensland politicians generate more debate than the uses and abuses of expert reports. Indeed, there are 300 of us here today to debate the very issues that, in general, have been unsatisfactorily managed by regulators and industry to date. I imagine Trevor Sykes will have some interesting stories to tell tomorrow morning in describing "Mines that weren't quite".

Jan 1, 1989

By D. Sinha

Mine investment decisions involve both objective and subjective criteria. The subjective criteria are efficiently decided by interaction with company experts. The Expert System approach can incorporate knowledge of experts and interact with conventional computer programs on mine evaluation. Artificial intelligence offers considerable opportunities in comprehensive mine investment decision making process.

Jan 1, 1998

Cerro Matoso (CMSA) in Colombia produces ferro-nickel from lateritic ore containing approximately three per cent nickel, using a Rotary Kiln - Electric Furnace (RKEF) process, in which the calcining of the mineral within the kiln is second in importance only to the smelting of the mineral in the Electrical Furnace. This article discusses the installation and operation of an expert system to control the kiln of CMSA. In the CMSA process, the run-of-mine ore is crushed, homogenised, dried and mixed with a small percentage of coal, before being heated up to 900¦C in a rotary kiln in the presence of natural gas mixed with air in a controlled proportion (air/gas constant ratio). The final objectives of the process are the following: ò To reduce to a minimum the iron and nickel oxides in order to convert these into metallic ions; this is a reduction process to facilitate largely the smelting in the electric furnace. Since the reduction is made with natural gas, it is much more economical than the subsequent complete reduction in the electric furnace, since electrical energy is much more expensive by unit of processed mineral. ò To heat the mineral as much as possible to facilitate the subsequent smelting. ò To eliminate totally the moisture content of the ore. An expert system with fuzzy logic was commissioned in 1994 for the control of the kiln and it has operated with good results ever since.

Jan 1, 1996

Predict the spontaneous combustion potential of a coal, using an expert system computer program. Background Approximately 15% of underground coal mint fires in the United States are attributed to spontaneous combustion. These spontaneous combustion fires frequently occur in worked-out or abandoned areas of a coal mine, This makes the control and extinguishment of the fire difficult and often requires large sections of the mine or even the entire mine to be scaled for long periods of time. A prior knowledge of the self-heating potential of the coal can be useful in preventing spontaneous combustion fires through the use of aggressive monitoring and control measures at existing coal mining operations, and in the planning oh new mines. Many factors can affect the spontaneous combustion process in a mine, including the self-heating potential of the coal, ventilation, geologic factors, and mining conditions and practices. Various test methods have been used to evaluate the relative self-healing potential of coals, including crossing point temperature and adiabatic heating methods, and oxygen adsorption techniques. However, these tests arc usually time-consuming and difficult to conduct, limiting their use to laboratories.

Jan 1, 1992

By Xuling Chen, Yi Wang, Xiaohui Fan

"The current control method of grate-kiln iron ore oxide pellet production is lack of accuracy and normalization due to the limited detection conditions. Therefore, mathematical models of grate-kiln temperature field are presented. With calculation of gas-particle heat transfer as well as enthalpy changes of moisture evaporation and magnetite oxidation, the thermal history of pellets throughout grate-kiln process can be acquired and used for production status estimation. Taking the calculated results as well as detected data in production as inputs, process parameters such as running speed, fan opening and coal injection rate as adjusting objects, expert system for grate-kiln pellet production is outlined. And the preheating section dominated control strategy is explained. Results of production application show that the stability rate of gas temperature in gas hood of preheating section is increased from 91.0% to 94.2%, compressive strength and first grade rate of pellet are increased by 3.62% and 2.36%, respectively, while FeO content is 18.18% lower. Control guidance of this expert system can solve production anomalies effectively, stabilize grate-kiln iron ore oxide pellet production and improve pellet quality.IntroductionSteel industry in China has developed rapidly in recent years. China’s crude steel production in 2010 is nearly 627 million, which is 44.2% of the world total production. Burden with higher quality is demanded by enlargement and modernization of blast furnaces. And oxide pellet production has become a growing concern. In 2010, the total production of pellet in China is 110 million, and about 60% is produced by grate-kiln process."

Jan 1, 2012

By A. K. Mukhopadhyay

[The mechanical status of a mine ventilation fan can be monitored effectively through vibration analysis. An expert system can be an effective tool in fault diagnosis for a variety of equipment. Computers have made it possible to implement an expert system to analyze the data for correct and appropriate diagnosis. In addition, vibration analysis can provide important information oil the condition of the machine. This paper discusses an approach for an expert system to determine on vibration analysis, practical problems that can occur with a fan. The frequency of the vibration is the only characteristics considered in the analysis, although other variables are discussed in the paper. Finally, a simple computer program, written in Turbo Prolog, is presented. A friendly query system, menu operation and a disk-resident separate knowledge base are a few of the salient features of this system.]

Jan 1, 1997

By Zhang Dekuan, Wang Haifeng, Yi Weiping, Wan Liping, Yu Yunzhen

The fundamentals of domain knowledge, knowledge base building, system structure, principle of inference mechanism, function of explanation mechanism, application analysis on Expert System of In-situ Leaching of Uranium (ESILU) are described, especially the Dempster/Shafer theory on inexact reasoning, which was used in ESILU, is detailed discussed. The application of in-situ leaching for abstraction of uranium and the analysis of parameters required for evaluation of uranium deposit. the procedure of evaluation of uranium dewsit for the purpose of in-situ leaching as well as the concept related were introduced. For building knowledge base, discussion is emphasized on production rule, which wasused in ESILU, on representation of rules and facts and application as well. From the system menu provided in the paper, the readers will, in a certain content, under- stand the functions of system structure and the way to set up. Several paragraphs are written concentrated on the procedure of backward reasoning and the principle of inference mechanisms combined with knowledge bases. Tracing interpretation and acquisition of knowledge with interactive method, which the system flexibility is considerably strengthened, are the main specific features of ESILU.

Jan 1, 1996

Expert systems have been employed in a number of areas for a considerable number of years. However, the traditional long development times associated with these systems have restricted their application. With the advent of powerful Expert System Development Tools such as COMDALE/X and the use of efficient development techniques, the time required can be drastically reduced. This results in the availability and applicability of this technology to a wide range of applications. This paper discusses some of the 'rules' which indicate amenability (or otherwise) of a particular problem to an expert system solution. Various problem types are examined and these are illustrated with reference to examples of their use at various operating sites.

Jan 1, 1991

By G. A. Kosick, C. A. Harris

"Metallurgy at the Polaris Mine is approaching an impasse, with lead and zinc concentrate grades and recoveries approaching their theoretical maximums. In an effort to increase productivity, plant throughput and operating material costs became a major focus of milling personnel.one method of increasing plant throughput was to minimize flotation circuit upsets by; standardizing operating strategy, decreasing control action response time of operators, increasing training of operators, and providing a more available decision support system for operators.To achieve these goals, Polaris personnel decided to utilize expert system technology to develop a real-time process controller for the flotation circuit which would be capable of justifying its actions to operating personnel.This paper describes the procedures followed in constructing the first process controller in the mineral industry which is based on the principles of Artificial Intelligence.IntroductionThe Polaris Concentrator was constructed in 1980 on a barge in Trois Rivieres, Quebec and towed to its present location on Little Cornwallis Island, North West Territories, about 1450 kilometers from the North Pole. The plant presently treats about 2,800 MTPD of new feed with average feed grades of 3% Pb, 13% Zn and 5% Fe. Current production averages 36,000 tonnes of lead concentrate and 220,000 tonnes of zinc concentrate annually.In an effort to increase the efficiency of the concentrator, the decision was made to implement an innovative control system on the flotation circuit. This system would embody and utilize heuristics ""rules-of-thumb"" gained from operators to make on-line process decisions. It would also justify these decisions on a demand basis."

Jan 1, 1988

To assist coal mine operators and regulatory agencies in selecting the most appropriate method to treat coal mine drainage utilizing expert system computer technology. The software package is called the Coal Mine Drainage Treatment PRO. Approach Experts in the field of coal mine drainage treatment were interviewed to determine their methods of assessing mine drainage treatment problems and their rationale for selecting a treatment protocol. These diagnostic, selection, and design criteria were codified in an interactive set of computer programs. Expert systems permit the user to input specific site data and receive treatment advice customized to the user's site and treatment objectives. How It Works The Coal Mine Drainage Treatment PRO package developed by the U.S. Bureau of Mines (USBM) consists of four programs or modules: (1) a control module, (2) a mine drainage treatment tutorial module, (3) a treatment advice module, and (4) a reference module (figure 1). A context-sensitive HELP Facility is also provided.

Jan 1, 1996

By J. L. Kohler

The knowledge-based expert systems (KBES) is a potentially powerful technique to the mining engineer in traditional areas of mine design and operations, as well as in the emerging areas of mining automation and robotics. Since KBES is still experimental in many respects, caution must be exercised as this "high tech" tool is developed and applied to mining. The objective of this paper is to demystify expert systems by introducing the elements of KBES, discussing the strengths and weaknesses, and identifying some areas of application in mining, based on the authors' experiences.

Jan 1, 1986

By Baafi E. Y

In this paper, impact of expert systems development is examined from the coal mining point of view. Some introductory concepts about expert systems are explained with a view to identifying possible beneficial applications of expert systems in coal Mining.

Jan 1, 1988

Integrated chemical thermodynamic software and computerised databank systems have evolved over the last 15 years. The extent to which these systems have automated thermodynamic calculations relevant to extractive metallurgy is examined. The scope for further automation and the role of the expert consultant is discussed.

Jan 1, 1991

By Edward F. Fries

Computerized, real -time monitoring had its beginning in the U.S. coal industry in the early 1970's. Over the last 10 years, developments in technology have made systems available that are capable of monitoring and controlling many mine parameters that can increase safety and production. However, situations currently exist where computerized monitoring systems are not being utilized to their fullest capability. They produce reams of printed output that are not sorted in a useful format, and human operators are not able to react fast enough in emergency situations tn tile large amount of information that the systems generate. A solution is the application of an expert system as the interface with real-time monitoring. An expert system would use an extensive knowledge base and human-type reasoning ability to make decisions on optimum strategies. Expert systems have been applied successfully to problems in other industries, and could provide significant benefits to mine information management systems. The use of an expert system with real-time mine monitoring is discussed in this paper.

Jan 1, 1986

By R. L. King

An expert system is a computer system that performs tasks that are considered by most people to require human expertise. Since their functioning relies heavily on large bodies of knowledge, they are often referred to as knowledge-based systems. These systems can contain the skill and knowledge of an individual human expert or the combined knowledge of a group of experts. The knowledge embodied in an expert system can consist of specific facts, HSHA regulations, heuristics or rules-of-thumb used by the expert, and databases. The purpose of this paper will be to introduce the participants in the ?Increase Competitiveness Thru Innovation? session to expert systems. This introduction will enable the participants to better appreciate the balance of the papers to be presented in this session. These papers will deal with specific mining problems and a proposed innovative solution. This paper will discuss such topics as problem-solving paradigms, rules of inference, knowledge representation, and the structure of expert systems.

Jan 1, 1989

By P. Cleyle, K. McDermott, M. Hall, C. A. Harris

"This paper describes the implementation of an expert system for monitoring and control of the NO. 4 autogeneous mill and spiral circuits at Wabush Mines.This process control system incorporated the mill operating expertize of Wabush Mines operating personnel and was implemented to improve spiral iron recovery and mill productivity by:1. Continuously monitoring the mill power draw and ensuring that the mill feedrate is regulated to prevent damage to the mill motor, while simultaneously maintaining the feedrate at or as close as possible to the maximum feedrate determined by the process supervisor.2. Improving control over the spiral density and feed rate.3. Improving control over both tbe rougher and cleaner wash water.4. Reducing the frequency with which pumps are plugged.5. Allowing modifications to be made easily and quickly to test various operating strategies.This system was installed using COMDALE/C, a real-time Expert System for Process Control. It was readily accepted by the process operators and produces higher production performance in comparison to the other 5 parallel mill lines at a lower power draw."

Jan 1, 1992