Search Documents

By JESSICA W. A. AZURE, Kansake A. Bruno, Samuel Frimpong, AZUPURI G. A. KABA, FORSYTH A. KADINGDI, PROSPER E. A. AYAWAH

Blasting generates undesirable ground vibration with adverse effects on humans, the environment, and structures. Numerous studies have used various techniques including machine learning (ML) to predict the peak particle velocity (PPV) caused by blasting in surface mines. These studies used the validation-set approach to evaluate the performance of their predictive models. In this study, Random Forest (RF), Gaussian Process (GP), and Gradient Boosting Machine (GBM) were applied for the prediction of PPV using data from a large open-pit gold mine in Ghana. An ensemble model based on a stacked generalization approach was then built on these three base models to significantly improve the PPV prediction performance. The model evaluation was done using repeated tenfold cross-validation which resolves the challenges with the validationset approach for small datasets. The dataset used in this study comprised a total of 196 measurements taken from different blasting operations. R-squared ( R2 ), mean absolute error (MAE), and root mean square error (RMSE) were used to evaluate the model performance. The results obtained from these models revealed that all the base models did not significantly outperform each other. However, the stacked generalization approach had a better performance, with RMSE = 0.12, R2 = 0.96, and MAE = 0.07. This approach presented an improvement over the existing methods and will be useful for highly accurate predictions of ground vibration during blast design. The application of this model will be useful for efficient and effective blast design.

Nov 8, 2022

If the communication/tracking system of an underground mine is severely damaged in a large mine accident or explosion, a surface based seismic system could be the only available tool for locating any trapped miners. This paper documents a field test of a seismic system at a room-and-pillar coal mine in Eastern Kentucky. Testing was performed at 6 different depths (450, 575, 750, 1175, 1525 and 1700 ft) above the mine. For each location underground, the miners pounded a ?set? of 5 ?repetitions? on the roof with each repetition being about one second apart. Over the 3 day field test, 63 sets of underground miners? pounding were detected by the seismic system at depths up to 1175 ft and at horizontal offsets up to 890 ft. Also during the testing, the observed signal strength generated by pounding with a crib blocks was seen to average 180% stronger than the observed signal strength generated by pounding with a sledgehammer. Ultimately, this field test was very successful and definitively demonstrated the ability of the seismic system to detect potential trapped miners? pounding up to a depth of 1175 ft.

Feb 23, 2014

By Daniel G. DiCanio

Investigation of the costs of industrial accidents dates back to the 1920s. The Bureau of Mines was one of the pioneers in accident cost investigation with numerous studies published between 1928 and 1933. The bulk of these early investigations dealt only with costs which appeared as payouts on company books; that is, the direct cost of insurance premiums and the costs of medical expenses which resulted from industrial accidents. These early investigations, especially the Bureau's study of accident costs in Pennsylvania and Indiana coal mines provided the first real evidence that promotion of industrial safety pays back in dollars as well as in human life and wellbeing. In the 30s and 40s, investigations began to accumulate evidence that industrial accidents impacted more than just the direct operating expenses of a company (i.e., higher insurance premiums). The work of Heinrich pioneered these investigations, and produced strong evidence that the un¬insured costs of accidents far outweighed those costs covered by insurance. Heinrich developed and published a ratio of insured to uninsured costs of 1.4. Despite his cautionary statements about blanket application of this ratio it is used even today to estimate total costs of accidents. Simonds, in the late 40s and early 50s, extended the definitions of uninsured or indirect accident costs and developed what is now regarded as the principal method for estimating the total economic impact of industrial accidents. Simonds' method served as the basis for development of the logic of the Accident Cost Indicator Model (ACIM). However, several departures from Simonds' method were made to satisfy certain requirements imposed by the Bureau of Mines, and because information gathered during the course of our investigation justified a departure. These departures are noted in the discussion which follows.

Jan 1, 1977

By J. Parker

Gentlemen; Rainy day in Upper Michigan, a good time to catch up on reading. Am doing that, and have a couple of questions concerning the K and P paper on roof problems and supports. I wonder if the authors can enlighten me. The compressive strength of the coal was reported as 4.4 MPa (633 psi) and compressive stresses measured in the coal seam were reported at around 27.6 and 48.2 MPa (4000 and 7000 psi ). Can such a soft coal actually contain such a high level of stress? Can it be measured as if the material were elastic, or does the coal behave plastically? Can that stress level be confirmed with hydraulic load cells? And if the coal is so soft, and if it is plastic, why is the stress field not hydrostatic? The intact roofrock appears to be unusually strong, with compressive strength reported at around 145 MPa (21,000 psi), and some failures were noted as soon as the rock was exposed. Were tests run to find out how that rock responded to the addition

Jan 1, 1991

By A. Festa

Secondary crushing often gets good press for its increases in energy efficiency and SAG milling rates but for a project considering this option the proposed benefits soon get overshadowed by implementation concerns. Reverse engineering from the published case studies is risky unless the reader understands the drivers impacting stability and operability of the circuit. This paper discusses the features of secondary crushing for Greenfield and Brownfield SAG mill grinding circuits. It is the author?s view that secondary crushing rarely adds value to a Greenfield project and restricts future expansion. Its application is best retro fitted to Brownfield operations where the benefit gained for capital expended is excellent. Two case studies are presented to reaffirm these statements. The first case study shows that the implementation of partial and total secondary crushing increases SAG mill power efficiency and it offers substantial incremental increases in milling throughput with staged capital investments for the crushing and grinding circuits respectively. The paper provides a critical analysis of process conditions for secondary-crushed grinding circuits, with particular emphasis on Brownfield circuit retrofits. Furthermore, the second case study quantifies the economic impact of secondary crushing on circuit design for plant start up and it illustrates the limit imposed on plant expansion.

Feb 23, 2014

By J. P. Rider, T. W. Beck, J. A. Organiscak

"The National Institute for Occupational Safety and Health (NIOSH), Pittsburgh Mining Research Division (PMRD) has recently developed a series of models utilizing computational fluid dynamics (CFD) to study gas distribution around a continuous mining machine (CM) with various fan-powered flooded bed scrubber discharge configurations in an exhaust curtain working face. CFD models utilizing species transport model without reactions in FLUENT were constructed to evaluate the redirection of scrubber discharge toward the mining face rather than behind the return curtain. The study illustrates the gas distribution in the slab (second) cut. The following scenarios are considered in this study: 100% of the discharge redirected back towards the face on the off-curtain side of the continuous miner; 100% of the discharge redirected back towards the face, but divided equally to both sides of the machine; and 15% of the discharge redirected toward the face on the off-curtain side of the machine, with 85% directed into the return. These models were compared to a model with a conventional scrubber discharge where air is directed away from the face into the return. The models were validated based on experimental data and accurately predicted sulfur hexafluoride (SF6) gas levels at four gas monitoring locations. One additional prediction model was simulated to consider a different scrubber discharge angle for the 100% redirected, equally divided case. This paper describes the validation of the models based on experimental data and the gas distribution results of the CFD models. INTRODUCTION To control methane gas and dust concentrations on continuous mining operations, it is important to provide sufficient quantity of fresh air from the end of tubing or curtain to the active face. In this region, operating machine-mounted water sprays and fan-powered, flooded bed scrubbers can significantly increase the amount of airflow reaching the face [1-3]. However, when an entry is large, especially in higher coal seams, the water spray configuration and the scrubber capacity may not be adequate to maintain sufficient forward airflow to confine the dust cloud to the face. Dust can roll back toward the machine operator as a result [4]. Currently, several coal mines having negligible gas emissions in higher coal seams (2.4 to 3.0 m [8 to 10ft.] thick) with exhaust tubing systems redirect the scrubber discharge back towards the face in an effort to improve dust collection. NIOSH studied the redirected scrubber exhaust techniques for dust and gas control in a full scale CM gallery to assess its potential for industry-wide application [5]. For a CM face, previous studies have concluded that the behavior of gas and respirable dust is a complex process. The distribution/dispersion and transport of gas and respirable dust after they are liberated at the face are governed mainly by their mass properties, the movement pattern of the ventilation air, and interactions with the mining machine and water droplets created by the external water spray systems used for dust control [6-10]. To understand these behaviors in a complex CM environment and to evaluate the effectiveness of various control techniques, numerical modeling has become a necessary supplement to laboratory experiments and field studies."

Jan 1, 2017

By R. Ribeiro

The designers of the new Arafertil concentrator in Brazil provided for complete automatic control of the grinding circuit to meet the operating objective of maximum throughput at a specified grind. Analog control methods were selected because of proven reliability and simple maintenance, and in order to minimize commissioning time and effort. This paper summarizes the step-by-step effort required to progress from local loop control of certain individual parameters to an integrated system capable of achieving the design goal. Emphasis is placed on the practical aspects of commissioning a control system.

Jan 1, 1979

By S. C. Nordeng

The program described in this paper is primarily a test of the possible utility of multiple regression in ore search, It was carried out as a cooperative project between White Pine Copper Company, the Institute of Mineral Research and the Department of Geological Engineering of the Michigan College of Mining and Technology. The authors wish to thank Dr. Cleaves Byers, Associate Professor of Mathematics, MCM&T, and Mr. Robert DeBaubien, Graduate Student in Mathematics, MCM&T, for their work in programming the project for the computer,

Jan 1, 1964

By Colin Lawrence

The Port of New York and New Jersey is one of the most heavily used transportation arteries in the world, handling nearly 40 per-cent of the North Atlantic shipping trade and directly providing nearly 230,000 jobs to the local economy. In 2004, $100 billion worth of consumer goods moved through the port. To accommodate future cargo volumes in the port, which are expected to double over the next decade and possibly quadruple in 40 years, deeper shipping channels are needed to provide access for a new generation of cargo mega-ships with drafts exceeding 13.7-m (45-ft) when loaded. Current channels within the harbor range in depths up to 13.7-m (45-ft), thus preventing carriers from using these larger ships, or requiring significant reductions in cargo to achieve lesser drafts to operate safely within the harbor. As part of the Harbor Deepening Project, the Anchor-age Channel would be deepened to 15.2 m (50 ft) below mean low water (MLW), for a length of 5,790 m (19,000 ft). The New York City Department of Environmental Protection (NYCDEP) owns, operates and maintains two existing water siphons in the harbor. Due to their shallow depth, both existing siphons must be relocated before dredging of the Anchorage Channel can be completed. Tom Costanzo, manager of capital programs for the Port Authority of New York and New Jersey (PANYNJ), says that ?While the re-location of the existing siphons under the Anchorage Channel is vital to the success of the deepening program, it is also a story of cooperation among key agencies at different levels of government resulting in improvements that will benefit the people of this region greatly.?

Jun 1, 2010

By M. Albrecht

In the mining industry and particularly in the processing plants, we use some large and unusual equipment. Understanding what it is and what it does is important to good operations, but how do you teach a new employee. When I started out in this field, the information on the processing equipment was easy to come by. Equipment manufacturers readily supplied it; their salesmen, often engineers themselves, knew how important the availability of this information was. Old engineers used their notebooks to train the new engineers. Over time this has changed. Equipment manufacturers have changed hands or gone out of business, and often the older engineers are gone. As a start in reversing this, this presentation will be an introduction to sizing, selecting, and operation of some of this equipment, in particular gravity separating equipment. Major portions of this have been excerpted from A Mining Engineers Notebook: Mineral processing (2).

Feb 23, 2014

By J. F. T. Agapito

Experience at Energy West Mines in the Wasatch Plateau Coal Field, Utah, has shown that longwall face bursts occur mostly near the tailgate when extracting the third adjacent longwall in a panel section where depths exceed 610 m (2000 ft). It is believed that this is due to a combination of cover depth and side abutment stresses created by the widening of a pressure arch formed by interlocking and bridging of large blocks of" very competent strata. The 60-m-thick (200 ft) Castlegate Sandstone probably plays a key role in pressure arching. This paper presents a back-analysis of the location and magnitude of the stresses associated with face bursts that occurred when mining the 7th Right Longwall at the Deer Creek Mine. Improvement of these conditions may be achieved by widening the longwalls to attain critical subsidence width sooner while mining the second long-wall, and to reduce stresses at the face. Wider longwall faces, however, increase the chance of sandstone channel interception often associated with bursts. Satisfactory mitigation of face bursts beyond cover depths of 610 to 670 m (2000 to 2200 ft) may be difficult without using destressing techniques. The hydraulic fracturing/liquid infusion technique seems to offer the best possibilities for application in longwall mining.

Jan 1, 1997

Fundamental results obtained from research on the properties of the edge surfaces of kaolinite particles (~500 nm) are reported. Of particular significance was the development of the experimental protocol. Well-ordered kaolinite edge surfaces were prepared as an epoxy resin sandwich structure having layered kaolinite particles in the center of the epoxy resin sandwich. Images of the sectioned kaolinite edge surfaces were examined by atomic force microscopy (AFM), and the average thickness of kaolinite particles in this study was determined to be 38.3 nm ± 11.7 nm. Furthermore, the surface charge of the kaolinite edge surfaces was evaluated with a super sharp Si tip. The point of zero charge (PZC) of the kaolinite edge surface was determined to be below pH 4, in contrast to the traditional view that the edge surfaces of kaolinite particles may carry a positive charge at pH 4. This lower PZC of the kaolinite edge surface was attributed to the lack of isomorphous substitution in the silica tetrahedral layer when compared to the PZC for the muscovite edge surface. Our results are consistent with the particle aggregation and flotation behavior of kaolinite, and should provide the basis for improved flotation strategies leading to the efficient recovery and utilization of mineral and energy resources.

Feb 23, 2014

Longwall mining is a commonly used method of coal extraction that involves the complete removal of large, rectangular panels of coal. Longwall mining causes the immediate seam roof to break and fall into the void created by mining. The convergence of the surrounding strata towards the mining void causes subsidence on the surface. The subsidence magnitude is related not only to the mining depth, the topography, the geological features and the overburden materials, but also to the sequence of the strata overlying a coal seam. In additional to the vertical subsidence which is evident at the surface, there are also horizontal movements inwards toward the centre of the void, which cause the surface to tilt, bend, stretch and com-press. This paper presents a parametric investigation of the influence of the overburden stratigraphy to surface deformations due to longwall mining using 2D finite element models. Models simulate rock movement by applying equivalent edge load to the open roof.

Jan 1, 2008

By D. F. Scott

The purpose of this study was to determine if miners are at risk from exposures to chemicals used in the mining industry, and determine the nature and sources of the illnesses and injuries. The authors reviewed the Mine Safety and Health Administration?s (MSHA) Employment and Accident, Injury, and Illness database and identified 2,705 cases of chemical-related injuries and illnesses that were reported to MSHA from 1999-2006, involving 66 different chemicals. The main source (cause) of chemical-related cases was acids/alkalis (about 39%). The primary nature (effect) of chemical-related cases was chemical burns (about 57%). The job classification where workers incurred the most chemical-related injuries and illnesses was cleaning plant operator/media operator/boney preparation plant operator/crusher worker. The number of nonfatal days lost did not change significantly and overall, the incidence rate of chemical-related cases decreased by 36%. Chemical burns accounted for a large number of injuries; mining companies should carefully examine their personal protective equipment (PPE) requirements, training methods, and safety culture to insure that their workers are protected.

Jan 1, 2009

By Luttrell. G.

The profitability of coal mining operations is highly dependent on the performance of the coal blending and processing facilities. Many companies have recognized this fact and have instituted optimization programs that require large amounts of coal characterization data, extensive operational information and complex costing models. Unfortunately, this brute force approach to optimization is often overwhelming in terms of time and cost and is generally too inaccurate to be of value in day-to-day field operations. A simpler and more practical approach involves the use of ?micro-price? optimization throughout the coal supply chain. This fundamental concept assigns unit values to each type product (particle or groups of particles) passing through the production units. True optimization occurs whenever all particles with positive worth are recovered and passed to market, while all particles with negative worth are rejected. This paper describes the fundamental workings of the micro-price concept as applied to coal blending and processing operations. Several examples are also included to demonstrate the economic gains that can be realized via the application of this concept.

Feb 23, 2014

By M. A. D. Azevedo, J. D. Miller, R. Kobler, E. Mutkowska

Up to 3 million tons of automotive shredder residue (ASR) is generated annually in United States. Plastics are the most valuable component of the ASR material making up approximately 25% of the total weight. In order to reuse these plastics, a selective process is required to separate each plastic, with very high purity and favorable economics. One separation technique, which may be able to selectively recover the plastics from ASR, is flotation. The preliminary flotation results have demonstrated that acrylonitrile-butadiene-styrene (ABS) and polycarbonate (PC) can be selectively separated from the model ASR feed, the PC/ABS/PA 66 system, by appropriate control of the surface chemistry.

Jan 1, 2001

By K. Lang

A test section of the upper most lift and underlying lifts of a large scale copper heap leach operation was instrumented and monitored with advanced wireline geophysics, fiber optics and resistivity electrode arrays. Elemental Capture Spectroscopy (ESC) copper measurements, resistivity measurements from an Array Induction Tool (AIT), electrical resistivity tomography (ERT) and distributed temperature sensing (DTS) with fiber optics provided integrated near real time data about the performance of the heap. During the leaching cycle, ECS provided continuous vertical measurements of copper in the upper most and underlying lifts in the test section of the heap and was used to generate a copper recovery curve. ERT inversion modeling provided high resolution 3D resistivity measurements, which were used to evaluate moisture distribution. DTS measurements of temperature were used to track the PLS front and calculate velocity. AIT induction provided highly precise bulk resistivity measurements. In-situ samplers provided PLS chemistry at multiple points. Properly integrated monitoring data can provide the heap leach operator with information that can be used to evaluate the hydrodynamics and metallurgy of the leaching process in near real time. Problems in the heap, such as pooling, channeling, longer lag and breakthrough times, excess acid consumption, increased solution in inventory, toe saturation and slope stability can be detected in near real time with the in-situ measurements evaluated.

Feb 23, 2014

By U. Thorley

Management of mine and tailings reclamation activities by surface mining oil sands operations in Alberta, Canada is experiencing increased national and international attention. Of particular concern is the ability of existing mines to execute and fund reclamation activities. Concern stems from a lack of demonstrated tailings reclamation technologies, perceptions of a poor track-record of reclamation activity by industry, and a lack of transparency with respect to the potential cost of reclamation activities relative to the monies secured in Alberta?s Environmental Protection Security Fund (EPSF). Limited information is publicly available regarding reclamation costs and technologies. Available data typically comes from disparate sources. This paper examines the expected costs of oil sands reclamation activities through a study of corporate Asset Retirement Obligation (ARO) liabilities, the Province of Alberta?sEPSF, and published data relating to reclamation costs.Issues relating to current liability funding are introduced.

Jan 1, 2011

By Jack M. Campbell

The regulatory process by which governmental authorities acknowledge and, in turn, permit the location, construction, and operation of large scale energy-related facilities has become extremely detailed in recent years. One example of the detail that is entailed in the regulatory process is the number of permits that must be acquired by operators prior to the construction and operation of energy-related facilities. It has been estimated, for example, that 75 permits are required prior to the opening of a new coal mine in Pennsylvania while 50 are required in Kentucky. From the perspective of facility owners and operators the regulatory process often appears overly complex and may be perceived as a constraint to the corporate decision making process. From the perspective of those responsible for administering regulations, the process is perceived as a mans by which public desires - pursuant to social objectives such as environmental protection and occupational safety - can be achieved. As a result of conflicting organizational interests, the clash between energy facility owners and operators and their regulators appears to be inevitable. Furthermore, the resolution of this conflict lies in the tradeoff between objectives and the costs of regulation. The balance of objectives and costs is a subject of the political process and not a topic of consideration in this paper. This paper is concerned with the actual administration of regulations, that is, the implementation of that which arises from the interplay of organizational interests. The paper assumes that environmental control laws will not likely be repealed, that regulations to implement the intent of these laws are here to stay and thus the paper seeks to assess how regulations are administered at various levels of government pursuant to objectives. The paper hypothesizes that the administration of objectives (or policy) has an impact on decision making which affects energy facility decision; making while recognizing the fact that regulatory policy cannot be totally divorced from administration, this paper proceeds

Jan 1, 1977

By Sebastian Avalos, Julian M. Ortiz, Willy Kracht

Semi-autogenous grinding mills play a critical role in the processing stage of many mining operations. They are also one of the most intensive energy consumers of the entire process. Current forecasting techniques of energy consumption base their inferences on feeding ore mineralogical features, SAG dimensions, and operational variables. Experts recognize their capabilities to provide adequate guidelines but also their lack of accuracy when real-time forecasting is desired. As an alternative, we propose the use of real-time operational variables (feed tonnage, bearing pressure, and spindle speed) to forecast the upcoming energy consumption via machine learning and deep learning techniques. Several predictive methods were studied: polynomial regression, k-nearest neighbor, support vector machine, multilayer perceptron, long short-term memory, and gated recurrent units. A step-by-step workflow on how to deal with real datasets, and how to find optimum models and final model selection is presented. In particular, recurrent neural networks achieved the best forecasting metrics in the energy consumption prediction task. The workflow has the potential of being extended to any other temporal and multivariate mineral processing datasets.

Jun 16, 2020