Search Documents

By Akram Marseet, Andrew J. Petruska, Rik Banerjee

Underground mine roof bolting is a crucial operation for miners’ safety and mine sustainability. Since roof bolting is a manual or human-supervised operation, miners’ safety is at risk due to dust or rock falls. Traditional machine learning algorithms have shown limitations to detecting drillable areas, mainly due to harsh lighting conditions. The authors propose an adaptive deep-learning framework for autonomous roof bolting. The proposed framework is based on implementing a binary semantic segmentation algorithm on color images to classify pixels that belong to rock from those that belong to non-rock. Significantly, the proposed framework implements deep learning semantic segmentation on images from traditional and neuromorphic vision sensors in underground mines. The performance of the proposed model shows an impressive accuracy level of at least 98% at a low number of training epochs with smooth learning curves. The high accuracy enables the implementation of autonomous roof bolting, greatly improving miners’ safety and operational efficiency while reducing human exposure to safety hazards. This research will advance the use of deep learning in mining automation and has the potential to revolutionize the traditional mining industry.

Feb 1, 2024

By Jonathan K. Hrica, CASSANDRA L. HOEBBEL, Jennica L. Bellanca

Accidents involving powered haulage and mobile equipment such as haul trucks often account for the greatest number of fatalities in the mining industry each year. Despite previous analyses that have identified root causes and other contributing factors, there is still a need to better understand the events leading up to these types of accidents, what lessons may be learned, and what strategies can be employed to prevent fatal accidents from occurring. This study examines Naturalistic Decision Making (NDM) using the critical decision method (CDM). The CDM is a retrospective interview approach used to explore timelimited, high-stakes, decision-making. In this study the CDM is used to obtain more information about what happens prior to, during, and after a potentially fatal situation such as a near-miss event, loss of control, or minor accident involving equipment damage. Researchers captured firsthand accounts from 21 haul truck operators involved in near-miss events from mine sites of various sizes and commodities throughout the United States. These accounts provide rich and detailed narratives from the perspective of haul truck operators themselves and reveal insights into what decisions haul truck operators make, what sensory cues they perceive, and what strategies they employ during challenging and non-routine situations. Decision-related themes that emerged from the data are presented and discussed. These results, along with potential solutions offered by study participants, can help to inform future research, raise awareness about hidden hazards, and build more creative interventions and realistic training scenarios for use by the industry to address haul truck safety issues.

Feb 1, 2024

By Jurgen Hahn

The paper reports on both, the design and decision making for the filters in a backfill plant in Europe as well as more then 10 years of operation and maintenance experience of the vacuum disc filters installed. It begins with the first step, the filter sizing based on a sample provided by the customer, followed by the commissioning of the filter units and reports on the development of the filter performance during the following 10 years of operation. Based on the physical properties of the sample and the process targets of the client for the backfill plant, the paper shows the logic steps taken in the laboratory to ensure the target moisture, determine the required filtration area and filter size as well as the selection of the ancillary units and the formulation of the performance guarantee. The paper discusses some issues with filter operation during the commissioning phase and the actions taken. Finally, the paper concludes by highlighting the gradual change over the more than 10 years of operation and how this change affected filter performance and operating costs.

Feb 1, 2024

By Angelina Anani, Nathalie Risso, Gail Heath, Dean Riley, Nilufer Akbulut, James Werner, Víctor O. Tenorio

Implementing a Digital Twin at the San Xavier Mine Laboratory (Sahuarita, AZ), requires a network redesign with a robust architecture. The goal is to create an ecosystem in where all personnel and equipment can be monitored in real-time from the University of Arizona campus, visualizing the site in a digital terrain model. A wireless mesh will help to test robots with autonomous features. Expected outcomes include data retrieval and analytics, the evolution of communications and safety protocols, tele-operation, and an innovated approach for managing the site with new supervision challenges. A timeline with expected commissioning benchmarks is also included.

Feb 1, 2024

By Milan Yekich, Emanuele Cauda, Justin Patts, Cody Wolfe

A worker’s personal exposure to respirable dust in occupational environments has traditionally been monitored using established methodologies which entail the collection of an 8-hour representative sample that is sent away for laboratory analysis. While these methods are very accurate, they only provide information on the average exposure during a specific time period, generally a worker’s shift. The availability of relatively inexpensive aerosol sensors can allow researchers and practitioners to generate real-time data with unprecedented spatial and temporal granularity. Low-cost dust monitors (LCDM) were developed and marketed for air pollution monitoring and are mostly being used to help communities understand their local and even hyper-local air quality. Most of these integrated sensing packages cost less than $300 per unit, in contrast to wearable or area dust monitors specifically built for mining applications which have been around for decades but still average around $5,000 each. At the National Institute for Occupational Safety and Health (NIOSH), we are leveraging the power of high-volume data collection from networks of LCDM to establish baseline respirable hazard levels and to monitor for changes on a seasonal basis as well as following any application of control technologies. We have seen the effective use and advantages of monitoring live data before, during, and after events like shift changes, operational changes, ventilation upgrades, adverse weather events, and machine maintenance. However, many factors have prevented a systematic adoption of LCDMs for exposure monitoring: concern for their analytical performance, the complexity of use, and lack of understanding of their value are some factors. This contribution outlines a one-year case study at a mine in Wisconsin USA, covering the installation, maintenance, data visualizations, and collaboration between NIOSH researchers and the industrial hygiene professionals at the mine.

Feb 1, 2024

By Jupneek Singh, Zachary A. Kiehl, Leila Safazadeh, Gareth Kennedy, Eleonora Widzyk-Capehart, Jeremy W. Ward

Mine workers are continuously exposed to a host of nonfatal stressors and potentially fatal hazards: noise exposure, excessive vibration, poor air quality, toxicant exposure, ignition of combustible gases, equipment-related accidents, and thermal heat stress. Wearable or portable sensors and contextual analytics provide a platform to unobtrusively collect and fuse multimodal health and safety data in real time to compute both acute safety and longitudinal health insights. This paper presents preliminary trial results, conducted by Simtars, VigiLife, Inc., and Queensland Mines Rescue Service, of the application of wearable sensors to monitor heat stress and other vital biometrics for mines rescue personnel.

Feb 1, 2024

By L. Alagha, L. A. Sanchez-Calderon

The existing methods for recycling lithium-ion batteries present initial hurdles related to the safe handling and preparation of the batteries to avert potential explosions. Submerging batteries in sodium chloride (NaCl) solution has become an acceptable industrial practice. However, the use of NaCl to discharge lithium-ion batteries rusts the batteries’ cases and releases metals which impacts the downstream processes that involve selective recovery of Li. This study aimed to determine the optimal concentrations and pH range of NaCl for the discharge of lithium-ion batteries for the purpose of minimizing the leach of Al, Co, Cu, Fe, Li, and Ni.

Feb 1, 2024

By H Arvidson, Rondinelli Sousa, Nic Pollock, Cesar Carrasco, Boz Elloy

Global public mineral reporting codes have driven strong governance of public disclosures of minerals information. Recently, we have seen an increase in disclosure of information related to Environmental, Social and Governance (ESG) issues, and governance of ESG information in mining companies has arguably reached a similar level of governance and diligence afforded mineral reporting. We define Resource Governance as the conjunction of environmental, social and minerals governance. A modern systems approach to Resource Governance, which combines mineral and ESG information, would provide not only reporting and governance capability, but also has the potential to sense/respond, predict, and manage ESG risks.

Feb 1, 2024

By Virginia T. McLemore, Abena S. Acheampong-Mensah

Critical mineral endowment of mine wastes in two mining districts in New Mexico (Copper Flat at Hillsboro and Carlisle-Center mines in the Steeple Rock district) will be characterized and estimated. “Beta-testing” of USGS procedures was performed. Potential critical minerals at these deposits include As, Bi, Te, Zn, Co, Ni, Mg, Mn, and fluorite. pH and particle size of samples were analyzed to determine weathering and migration potential of heavy metals. Soil pH was also measured to determine the potential for Acid Rock Drainage. The pH of the waste rock piles ranged from 3.66 to 5.67 which indicates fine- grained pyrite or sulfide oxidation. The samples collected from the tailings however, showed a different range of pH, from 6.30 to 8.62, probably due to the presence of carbonates. Difference in particle size fractions and its distribution along the slope are generally influenced by natural occurrences (e.g., gravity and pre-mining hydrothermal alteration) and operational activities such as material piling or dumping. Future work includes analyses of mineralogy and particle size correlation and estimation of critical mineral endowment of these mine wastes at New Mexico.

Feb 1, 2024

By NORA Mining Sector Council

The National Occupational Research Agenda (NORA) is a partnership program to stimulate innovative research and workplace interventions. In combination with other initiatives, the products of this program are expected to reduce the occurrence of injuries and illnesses at work. Unveiled in 1996, NORA has become a research framework for the National Institute for Occupational Safety and Health (NIOSH) and the United States. Diverse parties collaborate to identify the most critical issues in workplace safety and health and develop research objectives for addressing those issues. NORA entered its third decade in 2016 with an enhanced structure. The 10 sectors formed for the second decade will continue to prioritize occupational safety and health research by major areas of the U.S. economy. In addition, NORA’s seven cross-sectors will continue to focus on the major health and safety issues affecting the U.S. working population. While NIOSH serves as the steward to move this effort forward, it is truly a national effort. NORA is carried out through councils comprised of members with many diverse interests that are developing and implementing research agendas for the occupational safety and health community over the third decade (2016-2026) of NORA. Councils address objectives through information exchange, partnership building, and enhanced dissemination and implementation of evidenced-based solutions. NORA groups industries into 10 sectors using North American Industry Classification System (NAICS) codes. The Mining Sector encompasses NAICS code grouping 21xxxx where “xxxx” represents the different subcodes of the mining “21” classification. Mining encompasses the extraction of solid mineral and mineral-like substances from their place in the Earth’s crust and supplies them to processing industries that produce economically valuable materials. It does not include oil and gas extraction. The Mine Safety and Health Administration, MSHA separates mining into coal, underground and surface, and metal and nonmetal, underground and surface.

Feb 1, 2024

By O. Cermak, F. Degonld

Although 3D fragmentation analysis is a relatively new technology, it is rapidly gaining popularity among mid-sized mining operations and quarries. The results provided by this technology are adequate, but like other picture analytic methods, it mainly focuses on the surface of the muck-pile. These results are incomplete or incorrect as falling rocks and boulders from the stemming zone are constantly influencing the muck-pile surface. This obstacle can be overcome by collecting a large amount of data and analyzing it over time. However, this is not always an ideal solution, as companies prefer to receive results immediately after the blast to make resource allocation decisions, production planning, and plan for the next blast. Over the past year and a half, we have conducted a study on various muck-pile fragmentation analysis results from 3D modeling in Central Europe, comparing the results between the surface and inside of the muck-pile. By analyzing multiple shot results, we were able to determine the percentage difference between the top and inside fragment sizes and calculate the range of correction coefficients for different rock types. Applying these coefficients to a single muck-pile surface fragmentation analysis allows us to determine the actual fragment sizes inside the muck-pile. This enables us to make informed decisions regarding loading, hauling, material mixing, and proper crushing or milling. By overcoming the limitation of surface-based analysis, we can now accurately evaluate blast performance, which leads to better processes optimization and increase efficiency. As 3D fragmentation analysis becomes more widespread, we expect to see even more benefits for the mining and quarrying industry

Jan 1, 2024

By D. S. Scovira, T. Gustavsson

Exiting WW2 (mid-1940s), the potential of nitrates (ammonium nitrate, calcium nitrate, sodium nitrate) and hydrogen peroxide for application as a commercial explosives for mining, construction, and other industrial purposes both became known. The technology stream advanced the development and uptake of nitrate based explosives to the point that they are the current global standard for blasting. Today, environmental issues are shaping the strategic agenda across various industries. Most global mineral resource, construction houses, and explosive manufacturers have publicly announced ESG carbon reduction commitments in support of COP26 Net Zero 2050. The mid-2010s saw a renewed interest in hydrogen peroxide explosives to eliminate NOx fume. Carbon reduction initiatives in the early 2020s provided additional impetus to investigate hydrogen peroxide based explosives for industrial use. Hypex Bio has recently pioneered the formulation, manufacture, and end-use delivery of a Hydrogen Peroxide Emulsion (HPE) on an industrial scale. The HPE is composed primarily of hydrogen peroxide (H2O2) plus a lesser amount of fuel and emulsifier phase. The base HPE is produced in a low energy intensity modular plant using industry proven mixing techniques. The base HPE is a white, highly viscous emulsion that is similar in consistency to base nitrate emulsions. To enable end blasting use, a mobile charging unit gases the base HPE to a detonable density. The HPE is compatible with current priming and initiation systems. Hydrogen peroxide explosives offer a significant reduction in total carbon emission as compared to nitrate based explosives. The production of ammonium nitrate (AN) emulsion is energy-intensive and not carbon neutral. Based on the EU Average and for the oxidizer phase only, 1 kg [2.2 lb] of AN emulsion emits 2.3 kg [5.1 lb] of CO2 as compared to the production of hydrogen peroxide emulsion that results in 0.23 kg [0.5 lb] of CO2. This is a difference of 90%. HPE contains no nitrates and does not generate post blast nitrous oxides (NOx). HPE generates no aqueous nitrate or ammonia pollution and is a solution to meeting increasingly regulated mine site water discharge limits. Hypex Bio recently conducted a successful underground HPE evaluation with a major mining company in Sweden. HPE blast designs were the same as those done with nitrate based emulsions. Rock breakage, advance, muck displacement, and excavator performance were equivalent. Ventilation and re-entry times were reduced. The mining company has expressed further interest in assessing HPE and is demonstrating industrial leadership by exploring this transformative explosive technology

Jan 1, 2024

By E. Líder, J. Pavez, J. Rivera

The application of new technologies, methodologies, and the use of new products in mining operations is a practice that is becoming more common in the industry every day, added to the new challenges and restrictions found in the operation. In the city of Tierra Amarilla, the interaction between mining operations and the community presents a constant challenge given the proximity to the community and the limitations that this presents. Explosive charge mass, in conventional blasting using traditional explosives (ANFOs, packaged explosives for example) limits the explosive charge lengths (they are reduced), impacting the explosive charge productivity. To improve the productivity of explosive charge and fragmentation, the use of low-density explosives was proposed, allowing the increase of the explosive charge length, trying to maintain the vibration levels within the established limits (2mm/s), preserving the fragmentation of the rock in the required sizes. The present study shows the results obtained from the application of low-density bulk emulsion for production blasting charges using upward and downward fan drillings. During this process, 18 blasting events were carried out, of which 8 were Up-hole loading (average density of 0.75 gr/cc) and 10 with Down-Hole loading (average density of 0.95 gr/cc). In summary, a total of 110,760 tons of ore have been broken, using low density emulsion, meeting the proposed objectives such as: 33% reduction of vibrations, 12% overbreak (within the theoretical design), P80 of 11in and 4.6% in fines.

Jan 1, 2024

By J. Sanchidrián, A. Fernández, P. Segarra

The combination of chemical analysis techniques as XRF and measurement-while-drilling (MWD) technology in underground mining is a novel approach to gather comprehensive on-line information about physical and mineralogical data. In this work their integration is investigated aiming to delineate the ore/waste boundary. The input data analyzed come from a narrow vein underground mine, where the ore correspond to fluorite, and gathered in two production levels with complex mineralogical characteristics. Two approaches are followed to build a generalizable classification methodology and a site-specific model: a k-means clustering heuristic algorithm to define rock classes based on their physicochemical features, and a machine learning (ML) ensemble model trained with drilling parameters. For an imbalance class problem, a good recognition is obtained for the ore when is related to a high content of silica and for the waste. To predict the ore with low silica content it is necessary to integrate an analysis-while-drilling method to recognize the fluorite content when this mineral is embedded in different rocks. MWD can be employing for the lithologies recognition when representing different structural and mechanical properties between them.

Jan 1, 2024

By J. Pizarro

Any organization that subjects its activities to a continuous improvement methodology, aiming to optimize their processes and services, must be able to adequately measure the results of the implemented changes to pursue better outcomes. This is no different for mining companies. When it comes to blasting, an accurate fragmentation analysis can correlate blast design and rock mass parameters to particle size distribution, enabling us to further analyze the correspondence of specific design patterns, applied to specific rock mass properties and the result in measured fragmentation indicators. Over time, many tools were developed for us to obtain PSD data, albeit each of them had their own shortcomings. The first ones to be developed involved photographing the muck pile including a scale object, uploading those photos to a PC, and using specialized software to analyze each one. As a result, analysts would get a PSD curve that served as reference for the performance of the blast. This method, which is still used, has major shortcomings. The most obvious one is that it is only able to capture a slice of the blast, which corresponds to the exposed section of the muck pile the analyst has access to, and even there its reach is limited to the areas that are safe for getting close enough for placing the scale, which are not many. Consequently, this is a very limited way to get a fragmentation performance assess for a blast. Currently, autonomous systems are capable of taking continuous pictures of the muck pile. Installed on a shovel, conveyor belt or crusher feed, these systems can take a continuous stream of pictures that are later analyzed automatically without the need of human intervention. These systems greatly increase the volume of data and, consequently, we have a PSD curve that is more representative of the blast performance. However, this technology still had issues when trying to sample front end loaders mining operations or getting results from underground blasting due to the inherent difficulties of these operations. New algorithms used for capturing and processing data for front end loaders can overcome the harsh environment of their operations and produce PSD data in high vibration environments and difficult mining angles. For underground mining, algorithms can be used to sample each truck exiting the mine, giving a reliable source of fragmentation analysis that can be traced back to their stope of origin. The development of this technology extends the existing system’s capabilities and helps us in our efforts to improve fragmentation results with each blast.

Jan 1, 2024

By P. Couceiro, J. Navarro

A new methodology to perform selective blast designs by adapting the explosive energy, hole by hole (or in the hole), to the rock properties as a combination of their strength properties through the X-Rock model has been developed based on drill-monitoring data from both autonomous and manual rotary drilling modes. For that, a thorough normalization process has been developed to minimize external influences in the data induced by the different rig system and the operator interventions during manual drilling operation. Principal component analysis has been used to combine drill monitoring information. Data from 48,385 blastholes, comprised in 102 production blasts at 7 different levels in one expansion of the mine have been used for the analysis. The methodology, called X-Energy, combines the X-Rock model outputs with actual blast design parameters (drill pattern, stemming and borehole length, explosive mass/density and others) and a constant variable obtained from a machine learning algorithm, to estimate size distribution and percentage of fines and coarse material, hole by hole, by combining and recalibrating the Kuznetsov (Cunningham, 1987) based equation. From it, a new shovel-independent diggability model has been developed as combination of the P80 size, the rock properties (from the X-Rock) and a constant variable obtained from a machine learning algorithm. The X-Energy application presented in this study have been implemented in Minera Escondida Limitada (MEL), to design and operate production blasts with the goal of varying the explosive density to the rock properties to obtain a more uniform size distribution and digging rates, hole by hole, based on a target size and rate, and improving slope care, preventing active faults to fail. The methodology has been applied and validated in 17 production blast (8,277 blasthole), unlocking an average of 12.8% reduction in explosives costs (explosive consumption), by combining a pattern expansion strategy and a better distribution of the explosive density within the blast. The results demonstrate how a better distribution of the explosive energy by varying its density in the hole according to the rock and blast geometry variables can reduce powder factor (explosive consumption) while improving or maintaining blasting and digging results.

Jan 1, 2024

By M. Araos

Non-ideal explosives can be characterised by different methods like velocity of detonation measurement, double pipe tests, underwater tests, aquarium tests, etc., to name a few. Both aquarium and velocity of detonation tests were selected to characterise a non-ideal explosive that is based on hydrogen peroxide, fuel, and void sensitisation. The detonation process in an aquarium of this sensitised explosive was captured at 80,000 frames per second using a high-speed camera. The explosive product was detonated using 51mm (2inches) diameter acrylic pipe charges. The product was detonated at different densities in an aquarium (cube box containing water). An aquarium provides a homogenous confinement for the detonation of explosives and more importantly, allows observing the shock wave as it changes the refraction index of water. The analysis of the video footage allowed the calculation of the velocity of detonation, pressure, gas rate release, and how the shock wave's shape in water changes with densities. It is suggested that the calculation of the rate of gas release could be related to the bubble data obtained from underwater tests.

Jan 1, 2024

By A. Gontijo, A. Matos, P. Lincoln, L. Fonsecail

This paper investigates how to manage and model the vibrations caused by rock blasting in a Brazilian underground mine. Considering the frequent detonations and the previous lack of effective vibration control, understanding, and reducing the impact of these seismic waves is critical. We emphasize the use of the near field model technique, which considers the varied physical properties of different rocks, optimizes blasting plans, and minimizes damage to the rock mass. We detail a case study that demonstrates a comprehensive approach to vibration control. The discussion includes the creation of a monitoring plan, the methodology used, results and the specific challenges and solutions found during the blasting process. This study provides a practical guide for similar future projects. Finally, we present a predictive model based on the collected data, offering insights into the vibrations caused by blasting. This model helps decision-makers minimize the risk of displacing rock blocks in the blasting area, improving safety and operational efficiency

Jan 1, 2024

By K. Dufresne

Blasting in geothermally high temperature ground and/or extremely reactive ground is an increasing common challenge for surface and underground mines. Many current PETN based non-electric and electronic detonators have manufacturer published in-hole temperature limits ranging from 55°C to 65°C (131°F to 150°F). When confronting blasthole temperatures above the manufacturer published limits, operations may request site-specific guidance and qualification for use. The advised safe loading window for in-hole PETN based detonators is often too short to be practicable for production blasting. To overcome limitations of PETN based initiation systems, surface mines with high temperature ground and/or reactive ground are known to load blastholes with a qualified high temperature bulk emulsion that are top primed with a cartridge of temperature qualified detonator sensitive emulsion plus an RDX detonating cord downline. As there is no in-hole delay detonator in this system, blasthole timing is accomplished with surface delays. Although this system is safe, top priming and surface delaying does influence floor quality and top size fragmentation, and safe loading windows times reduce blast block size and increase blasting frequency. A mining company with a site characterized by geothermally high temperature ground plus highly reactive ground employed a blasting system like the one described above. The site desires to improve blasting performance by moving to an in-hole delay initiation system that can withstand the site ground conditions. At the request of the mining house, DynaEnergetics (hereafter “the company”) developed an electronic initiation system for application in geothermally high temperature ground and/or extremely reactive ground conditions. The DynaEnergetics IGNEO high temperature electronic initiation system is rated for in-hole use in ground up to 150°C (302°F) for 48 hours. The detonator consists of a high strength detonator with an HMX base charge and thermally resistant downline. To make up a primer, the detonator is inserted into a booster charge assembly that is fitted with a small HMX shape charge. This high energy primer is designed to be compatible and reliable for firing bulk emulsions, including those with high percentages of urea inhibition to combat reactive ground. Up to 1000 detonators may be programmed in 1ms increments up to 1500ms. The system is controlled by a digital firing panel that can be used with either a wired or wireless trigger. The system enables high temperature and/or reactive ground sites to utilize in-hole delay initiation and realize the associated performance benefits

Jan 1, 2024

By J. A. Sanchidrián, P. Segarra, S. Gómez

Blasting is a common activity in mining that can cause significant damage in the remaining rock mass. The most commonly used method for predicting vibrations is the Swedish or Holmberg-Persson approach. It is a simple and fast method that uses empirical formulation to estimate the vibration levels based on the explosive charge weight, distance to the source, and other parameters. However, it has limited accuracy and does not consider the effect of free surfaces or other boundary conditions. On the other hand, recently developed methods as the full-field solution (FFS) can provide more accurate predictions of vibrations due to excitations provoked by cylindrical columns of explosive, but the inclusion of free surfaces in analytical methods is a challenging problem. This paper presents a new approach that includes free surfaces in the FFS to predict vibrations analytically in the near-field. The main obstacle to including free surfaces in the FFS was the belief that the solution was non-decoupleable in P- and SV-waves separately. The Heelan's approach was previously used to decouple them, but it only provides valid radiation patterns in the far-field and does not account for intrinsic attenuation. In this paper, we address this problem successfully, allowing for the inclusion of free surfaces in the FFS.

Jan 1, 2024