Search Documents

By S. Frimpong, E. Gbadam, W. Nyaaba, G. Somua-Gyimah

Recent studies towards dragline excavation efficiency have focused on incrementally achieving automation of the entire excavation cycle. Initial efforts resulted in the development of an automated dragline swing system, which optimizes the swing phase time. However, the system still requires human operation for collision avoidance. For full dragline autonomy, a machine vision system is needed for collision prevention and big rock handling during the ‘swinging’ and ‘digging’ phases of the excavation operation. Previous attempts in this area focused on collision avoidance vision models which estimated the location of the bucket in space in real-time. However, these previous models use image segmentation methods that are neither scalable nor multi-purpose. In this study, a scalable and multi-purpose vision model has been developed for draglines using Convolutional Neural Networks. This vision system averages 82.6% classification accuracy and 91% detection in collision avoidance. It also achieves an 87.32% detection rate in bucket pose estimation tasks. In addition, it averages 80.9% precision and 91.3% recall performance across terrain recognition and oversized rock detection tasks. With minimal modification, the proposed vision system can be adjusted for other automated excavators. Keywords: Surface mining, dragline bucket, deep learning, machine learning, oversized rock, convolutional neural network, machine vision, object detection, earthmoving equipment. INTRODUCTION Complete excavator automation is widely regarded as the next phase in efforts toward improving earthmoving efficiency. Earthmoving often involves complex, forceful interactions between an excavator and the ground. The nature of these interactions depends largely on the type, properties and physical characteristics of the earth material. This process is further complicated by the random occurrence of tree roots, boulders and other such obstructions. Therefore, the ability of a controller to detect changes in the operating conditions, adjust the digging strategy and respond in real-time is of utmost importance. Early studies into autonomous excavation identified some key performance criteria which included the following [1]: • The autonomous excavator must be able to work in any type of earth material. • Its excavation accuracy must be within 50mm. • It should be able to handle different surface and underground obstacles autonomously. • It should be able to operate at the speed of the average operator in any condition. • Its operation should be capable of safe integration with other site systems.

Jan 1, 2019

The criteria used to design, operate and control flotation columns are significantly different from those used for conventional flotation machines. Therefore, a computer-aided design (CAD) package for column flotation has been developed. The simulator has been set up using guidelines developed at Virginia Tech and information published in literature during the past few years. The simulator is useful for determining the proper design and appropriate operating conditions for a column cell to fit a specific set of conditions. In addition, the simulator can be used for training in column operation and for trouble-shooting operational problems. Simulation results have been found to be in good agreement with experimental data obtained using laboratory columns.

Jan 1, 1990

By J. A. Irvine

In 1972, a cooperative program to map and estimate the quantity of Tertiary coal resources in southern Saskatchewan was initiated by the federal and Saskatchewan governments. More than 700 exploratory boreholes were drilled over a two-year period, and data were obtained from other organization s on an additional 1500 boreholes in the area. Stratigraphic data and data on each coal seam penetrated in the boreholes, along with technical and location data for each borehole, were stored in a computer data base for convenience of retrieval, modification and manipulation. Programs and program packages were developed or adapted to: (1) select borehole information defined by variable retrieval parameters from the data base; (2) calculate derived information , such as overburden ratios. cumulative net coal thicknesses , mean weighted center of gravity of coal seams in each borehole, etc., from the selected data ; (3) produce posting maps of either selected data or calculated information in various combinations at variable scales on a plotter or CRT display; (4) produce contoured structure, isopach or ratio maps , and perspective diagrams as appropriate , utilizing plotter or CRT display; (5) calculate and categorize quantities of coal in place within user- defined "mineability-economic-reliability" limits. The emphasis in computer utilization was placed on geological credibility of the results and ease of operation by relatively untrained geological personnel, resulting in extensive use of interactive methods. The single most important benefit derived from using computer-assisted procedures is the ability to produce new maps and quantity estimates rapidly and inexpensively as a result of availability of new data or/and changes in external criteria which impose economic or mineability constraints.

Jan 1, 1978

By C. Woofter

Quarry operators must continue to change and adapt environmental compliance methods to meet the increasingly demanding regulations. To fulfill industry needs, a computer-based decision support system is being developed to facilitate decision making that mine operators are required to make regarding environmental responsibilities. This system can help to determine permitting and compliance measures needed for air, surface water, groundwater, noise, aesthetics, vegetation, wetlands, hazardous wastes, storage tanks and wildlife habitats during the development, operation, and closure phases of mining. Operators will utilize the system's user-friendly interface consisting of mouse driven responses and limited keyboard entries on a desktop computer. The system is easy to upgrade according to the user's needs and changes in current laws and regulations. Data input models, user interfaces, and support of decision making are explained. The program permits inclusion of the decision maker's own insights through "what-if' analysis, in which the user can change ad hoc entries and check the related consequences. Using this approach, the user has complete control over all steps of the decision making process. A storm water example is selected to demonstrate the potential benefits of the support system to environmental decision makers.

Jan 1, 1995

By Coxhead B. A, Brandt B. A

The geological exploration and evaluation of coal reserves involves the appraisal of very large amounts of data. Computers afford an ideal means of handling and manipulating the data, thereby avoiding the laborious and expensive method of manually compiling the logs in full descriptive language. The Minad system of computer storage and retrieval of geological and analytical data is particularly suited to the description of cores and cuttings from exploration drill-holes. The encoded data is arranged in a 120-column computer output sheet format; descriptive geological information in a 54-column geolog and analytical information in a 64-column assaylog. The acquired data is quantitative, uniform and relatively complete and can be selectively retrieved for processing by a wide range of specialized computer programs developed for the coal mining industry. Visual retrieval of data for manual processing is facilitated by the use of simple codes and logical word abbreviations.

Jan 1, 1977

By D. A. Poukhovski

Two integral components of the 30 CFR 48 training requirements mandated for new miners, newly employed experienced miners, and the annual refresher training of all miners are hazard recognition and prevention of accidents. In these categories, instruction shall include the recognition and avoidance of hazards present in the mine, and a review of accidents and their causes. Unlike instruction preparation for fact-based material, program development for hazard recognition places a particular challenge before safety-training specialists because they are often limited in their ability to adequately gauge the trainees' levels awareness prior to training. Further, they must also assure that the training is relevant, and it is presented in a thought-provoking manner. Computer-based instruction is particularly well suited for assisting trainers in developing hazard-recognition topics because it can be used in pre-training and post-training modes for assessment of training effectiveness, in addition to use as a presentation tool. Therefore, this paper will present a recent approach developed at Penn State to apply this technology for assessment and instruction of accident prevention through hazard recognition, with an application prepared for surface coal mines.

Jan 1, 1997

By C. J. Bise, D. A. Poukhovski

Two integral components of the 30 CFR 48 training requirements mandated for new miners, newly employed experienced miners, and the annual refresher training of all miners are hazard recognition and prevention of accidents. In these categories, instruction shall include the recognition and avoidance of hazards present in the mine, and a review of accidents and their causes. Unlike instruction preparation for fact-based material, program development for hazard recognition places a particular challenge before safety-training specialists because they are often limited in their ability to adequately gauge the trainees' levels awareness prior to training. Further, they must also assure that the training is relevant, and it is presented in a thought-provoking manner. Computer-based instruction is particularly well suited for assisting trainers in developing hazard- recognition topics because it can be used in pre-training and post-training modes for assessment of training effectiveness, in addition to use as a presentation tool. Therefore, this paper will present a recent approach developed at Penn State to apply this technology for assessment and instruction of accident prevention through hazard recognition, with an application prepared for surface coal mines.

Feb 24, 1997

By P. N. Thompson

The past few years have seen the introduction of mini-computers on an increasing scale at collieries in the British coal industry. These are being used for controlling underground transport systems, coal preparation plants and fixed plant (pumps etc.); and for monitoring the mine environment and face operations.. A separate mini-computer (termed a primary computer) is dedicated to each of these functions. The overall system under which these computers operate is a common one which is applicable at all collieries. The information collected by the primary computers, other than that required for immediate control, will be fed to a secondary computer for analysis and processing. This overall concept, called MINOS. (MINe Operating System) is being developed by the National Coal Board's Mining Research and Development Establishment (M.R.D.E.). Eventually most of the data required by the system will be captured automatically. For example, the fact that a delay has occurred on a face, together with the duration of that delay, will be detected automatically via transducers link linked to a primary computer. However, the reason for that delay will, in all likelihood, have to be recorded manually. Such a system is currently under development with the first trial installations now being made.

Jan 1, 1979

By L. Wade

Laboratory testing of the self-heating propensity of coal, using crossing-point temperature and differential thermal analysis techniques, involves the frequent recording of several temperatures. The volume of data and the desire for high accuracy of measurements motivates an automated recording system. The data which are recorded must be adjusted for calibration factors and the results presented in both graphical and tabular form. The paper describes hardware and software features of a computerised system which is being used to establish the relative propensity to self-heating of a wide variety of coals.

Jan 1, 1987

By F. S. A. de Frey

The main users of computerised financial evaluation tools in the mining industry are the mining investment analyst, the mine manager together with his financial manager and the mining student (undergruadate and post-graduate). The mining investment analyst's needs mainly centre around the-micro and macro aspects affecting the mining industry in general. The mine manager needs a tool to assist him with the planning, reporting and controlling of production, costs, efficiency, productivity and cashflow. For the mining student studying Financial Evaluation, the main need is for a computer program which will assist him in learning the basics of the subject. This paper concentrates on the needs and objectives of the course Financial Mine Evaluation, the approach to which is presented. Emphasis is placed on the practical computing sessions and the role of the FCS-EPS package as a training tool.

Jan 1, 1987

A computerised cataloguing system for geological specimens is described. The catalogue is designed for use with a compactus-type storage system. The information stored in the catalogue for each specimen includes the type of specimen, locality information in terms of standard ~ap sheets and descriptive location, geological age, stratIgraphy, and specimen descriptions. The catalogue can be interrogated at any level of detail on the basis of any combination of these specimen parameters.Other information stored for each specimen 'is the compactus address, collection date, reference information and the type of analyses performed on that specimen:Any mterrogatlon produces a listing of those specimens which satisfy the search criteria.

Jan 1, 1974

By Kathleen Y. Knoebel

This paper describes a graphical mini computer system called MPASS 2 (Mine Performance Analysis/Simulation) which has been designed to conduct analysis of underground room-and-pillar mining operations using both conventional and continuous mining methods. Dynamic, graphic displays of mine face machine activities are presented on the CRT screen to provide extensive visibility into underground mining activities. MPASS 2 consists of: an interactive inputting system that graphically creates a room-and-pillar mine and describes the mine when production is initiated; a time-step simulator that produces statistical summaries of shift activities and uses CRT screen graphics for accelerated-time display of machine activities and delays; a statistical analysis subsystem for analyzing mine performance data; an industrial engineering system that permits users to enter actual machine activity statistics and replay mine operations on the CRT screen; and a data management system to store historical performance information on mine characteristics and operations. MPASS 2 operates on a Hewlett-Packard 9845B mini computer with 450 K bytes of extended memory, and is written in H-P BASIC. It has been installed at Freeman-United Coal Mining Company in West Frankfort, Illinois and at the U. S. Department of Energy Pittsburgh Mining Technology Center. Applications include production system simulation, industrial engineering analysis, production scheduling, new equipment evaluations, and training.

Jan 1, 1982

By E. William Rossi, Audrey F. Glowacki, Richard L. Unger

The U.S. Bureau of Mines has developed a graphics-based computer model to assess operator visibility for continuous miners, shuttle cars, and scoops. Using digitized drawings of the machine, the program can analyze visibility for both 5th percentile female and 95th percentile male operators. This model was used to predict 95th percentile male operator visibility for two shuttle car designs. These results were then compared to those obtained from a manual assessment of visibility using the actual shuttle cars. The methods showed good correlation except for several cases which were attributed to imprecise digitization of the machines.

Jan 1, 1992

By G. T. G. Emère

The Harmony Gold Mining Company Limited employs approximately 29 500 people, including 27000 Black workers recruited from areas both within and beyond the boundaries of the Republic of South Africa. The problems arising from the control of this diverse and complex labour force have led to the need for a fully computerized labour control system. This paper describes how, with the aid of an extensive interrelated network of satellite minicomputers, together with ancillary peripheral user terminals and printers, the management of Harmony Gold Mine intends to meet these problems.

Jan 1, 1983

By Kostas Fytas

A computer package has been developed which is capable of evaluating alternative production schedules (long and short range) of an open pit mine once the ultimate pit limits have been defined. The model user can choose the best alternative according to cash flow considerations and the particular mining conditions. The package involves a combination of simulation for long range production scheduling and linear programming for short range scheduling. The simulation model based on an iterative technique, maximizes the profits within each long scheduling period (eg. year or sixth month period), and at the same time defers stripping as long as possible. A minimum exposed ore quantity is kept at the beginning of each mining period. The final long range schedule satisfies a set of constraints for each period, mainly ore quality and quantity constraints for ore and waste. An interactive smoothing procedure at the end of each period results in a schedule that is feasible in terms of mining practice. Finally a cash flow analysis model carries out cash flow calculations so that the user can select the best alternative. The linear programming model, based on a certain number of shovel regions, defined interactively by the model user, calculates the optimum production schedule for each short period under consideration, so that a set of constraints is satisfied.

Jan 1, 1986

By Bruce C. Trent

In order to effectively model subsidence caused by underground openings, a two-dimensional computer code has been developed which couples near- and far-field response. The near-field is simulated by a discrete element block code and a finite difference code represents the continuum region in the far-field. The explicit equations allow rapid and inexpensive solutions to geometries which may lead to subsidence. Two sample problems are presented which illustrate the important features of the code.

Jan 1, 1979

By Robert A. Melton

Since many of the easily mined, continuous, thick coal seams in the Appalachian Region have been exhausted, new exploration and mining is being directed toward less continuous seams that are under more cover and are more difficult to mine. The direct results of this are that more money and time and more expertise and data are needed to plan and open new coal mines today. It is also becoming apparent that the old methods of assimilating and preparing information needed for mine planning are inadequate to meet the needs of today's modern mining engineer, especially in the areas of data gathering, storage, retrieval, preparation, and interpretation. The system described here allows assimilation of old and new core data in a relatively quick, accurate, and efficient manner. The system is a set of computer assisted procedures consisting of the following four basic steps: 1) raw data collection, computer input, storage, and graphic preparation, 2) geologic profile correlation and establishment of seam continuity, 3) computerized data posting of important seam parameters, 4) geologic and planimetric analysis for mine planning.

Jan 1, 1977

By J. E. Lawver, B. J. Clingan, R. E. Snow

Response surface methodology is a well-known and powerful technique for determining optimum conditions in flotation systems. One disadvantage is the onerous task of the numerical calculations and curve plotting. This paper describes a system in which the investigator simply selects the range of several flotation variables after which the design of the experiment, data analysis, generation of response surface equations, and response surface plots are automatically carried out using a digital computer and a computer-driven curve plotter. Several typical response surfaces of important variables in phosphate flotation are presented as examples.

Jan 8, 1979

By J. E. Lawver, B. J. Clingan, R. E. Snow

Response surface methodology is a well-known and powerful technique for determining optimum conditions in flotation systems. One disadvantage is the onerous task of the numerical calculations and curve plotting. This paper describes a system in which the investigator simply selects the range of several flotation variables after which the design of the experiment, data analysis, generation of response surface equations, and response surface plots are automatically carried out using a digital computer and a computer-driven curve plotter. Several typical response surfaces of important variables in phosphate flotation are presented as examples.

Jan 1, 1980

By James E. Lawver

Response surface methodology is a well-known and powerful technique for determining optimum conditions in flotation systems. One disadvantage is the onerous task of the numerical calculations and curve platting. This paper describes a system in which the investigator simply selects the range of several flotation variables after which the design of the experiment, data analysis, generation of response surface equations, and response surface plats are automatically carried out using a digital computer and a computer-driven curve plotter. Several typical response surfaces of important variables in phosphate flotation are presented as examples. An overview of the flow charting of the system is provided in the addendum.

Jan 1, 1978