Search Documents

By E. Lentilucci, A. Dey, X. Liu, A. Datta, S. Ghosh

The segmentation of quarry blast smoke plumes is a crucial research area that has the potential to enhance safety, environmental monitoring, object detection, and regulatory compliance in the mining industry. Smoke plume segmentation can be used to identify the spatial extent and temporal evolution of smoke plumes generated by the blast. This information can be used to improve the understanding of blast dynamics and optimize blast designs. Furthermore, smoke plume segmentation can assist in identifying potential environmental and health hazards associated with the release of smoke and dust particles into the atmosphere during quarry blasts. However, many current smoke plume segmentation algorithms are designed for segmentation of smoke plumes from fires and are not suitable for segmentation of plumes from quarry blasts. The proposed method in this paper leverages the video frame captured before the blast as a stationary background image, which helps to reduce other confuser objects present in the image along with a selfadaptive clustering method. The color channels of the image are utilized based on the different contrast properties of the smoke, with a self-adaptive gradient compensation applied for edge detection of the smoke plume. This approach offers a novel and effective way to segment smoke in quarry blast analysis, with potential applications in the calculation of noxious gas volume and concentration within the plume. Compared to other algorithms, our method improves accuracy, decreases the false alarm rates, and reduces the overall miss rate. Overall, this paper contributes to the research on smoke plume image segmentation, particularly for quarry blast smoke plumes, which possess unique characteristics that require specialized algorithms for accurate segmentation.

Jan 1, 2024

By Brianna Eiter, Max Barham, Tim Bauerle

Fatigue-related risk is a persistent safety concern for the mining industry. However, fatigue and sleepiness are often treated interchangeably, which can lead to confusion and potentially less effective reduction of safety risk. To provide clarity, we present an overview of similarities and differences between work-related fatigue risk and sleepiness including definitions, theories, measurements, and mitigation strategies. As a supplement, a summative visual model which highlights these similarities and differences is presented. Expanding industry knowledge in this area will assist safety professionals in crafting more targeted risk management practices appropriate for work-related fatigue risk, sleepiness, or both.

Dec 19, 2023

By N. V. Moraka

Mining companies are required by law to improve representation of women on their boards. However, progress in this regard has been slow. Boards of directors play an important role in formulating corporate strategy, risk assessment, and effective governance for sustained financial performance. Although some women have successfully maintained board seats, others have been unable to do so. It is difficult to ascertain whether women are being appointed as tokens to satisfy social pressures, as little is known about the experiences of women board members in mining, which may inform or refute tokenism. In-depth interviews were conducted with 20 women and 16 men across six listed mining companies. Thematic data analysis revealed evidence of tokenism where recruitment to mining boards is compliance-based and is informally driven by influential directors. Adverse boardroom experiences that further confirmed tokenism were reported by women, such as disregarding their contributions, condescending behaviour, and limited influence in decision-making, while other women sought to validate their competence and oppose social exclusion. This study recommends that a strong Board Chair and nominations committee is critical to ensure sustainable recruitment of competent and suitably qualified women; and further foster a culture of inclusivity and valuing gender-diverse boards.

Dec 13, 2023

By R. M. Mashishi, T. Malehase, O. J. Okonkwo

The concentrations of selected trace elements (As, Cd, Hg, and Pb in run-of-mine (ROM) coal, discard, and coal product were investigated to assess the efficiency of beneficiation in reducing trace element concentrations and ascertain any implications regarding environmental compliance to regulatory frameworks. Samples were collected from a colliery in Mpumalanga Province twice a month for a period of 24 months. The samples were ashed to approximately 0.21 mm, then digested in a mixture of 70% HNO3 and 40% HF and analysed using inductively coupled plasma-mass spectrometry (ICP-MS). Except for Pb, the mean concentration of all elements decreased from the ROM stage to the coal product stage. The order in which trace elements occurred from the highest to lowest throughout the production chain was Pb>As>Hg>Cd. With the exception of Cd, the mean trace element concentrations in the discard were above the total concentration threshold (TCT) set for landfill disposal. The As, Hg, and Cd concentrations in the product were well below the thresholds for all land uses. However, Pb concentrations in the product coal were above legal limits, which is of concern with regard to environmental compliance and the performance and marketability of the product.

Dec 13, 2023

By Frank Pepe, Nestor Garavelli

The 2023 Rapid Excavation and Tunneling Conference (RETC) will be held in Boston, MA and will include the presentation of 133 papers divided into 20 technical sessions during the 3 day conference that were selected from over 200+ papers that were submitted to the committee. The selected papers are included within this book to provide a comprehensive recap of the conference while also providing a source of reference for contractors and engineers in the future. The selected papers will include contributions from all divisions of the tunneling industry plus risk management and best contract practices. Boston is known to the tunneling industry for its complex and difficult tunneling history, specifically of recent memory the Central Artery/Tunnel Project also know as the “Big Dig” of the 1980s and early 1990s. The selection of Boston as a host city was fitting from a historical perspective and as avenue for the city to present its upcoming tunnel projects. RETC began in 1972 with the idea of sharing ideas, design and project construction experience, and new technology, between contractors, designers, owners and suppliers. The ability to provide a single location every other year for all facets of the tunneling industry to gather and discuss the aforementioned was and still is viewed as a way to help the tunnel industry grow. The lessons learned from completed projects, the advancement of technology which could be implemented into design and the presentation of new projects by owners/designers and contractors allows for the limits and complexity of upcoming projects to be expanded beyond our current capabilities. Our industry has never been as busy as it currently is, and we are seeing more future projects coming out for solicitation across all business sectors. With the good fortune that all of us have been blessed with one of the biggest challenges that we have been faced with is the shortage of future management on all levels for our projects including the labor force. We all want our industry to continue to grow and succeed, our most important assets are the people that we employ. Investing in the education and continued training of our work forces to be up to date with the latest technology and methods is the key to our success. This also includes develop, recruiting and mentoring of young students graduating college to join our industry so that we can all continue to grow. We hope that all those attending the conference will have the ability to attend as many sessions as possible and to expand their network within our industry. As in past years, scholarship students will be in attendance and we encourage all to offer these students an opportunity to visit jobs, attend learning sessions or future internships to expand their interest in our industry that needs an injection of young engineers to contribute and lead our future Rapid Excavation and Tunneling Conferences. The program chairs wish to thank the session chairs, co-chairs, authors, and other members of the RETC Executive Committee for their dedication to this conference. Last but certainly not least, we give special thanks to the SME staff for their hard work and enthusiastic support in making this conference possible.

Dec 1, 2023

By Turker Hudaverdi, Yasar Agan

This study tries to predict blast-induced flyrock using intelligent modelling techniques. Flyrock is a site-specific phenomenon. Design variables and rock properties highly influence flyrock throw distance. Site investigations were conducted in a sandstone quarry. Fundamental operational parameters, blast dimensions, bench face condition, and rock mass structure were monitored. Stepwise regression technique was applied for variable selection. Burden–hole diameter ratio, in situ block size, and powder factor were determined as the most significant parameters. Considering output of stepwise regression, artificial neural network, ANFIS, and Gaussian process regression techniques were applied to predict flyrock throw. A comprehensive validation was performed using twelve different performance indices. Pre-determination of input parameters supported development of successful soft computing applications. Rock block size was found to be an appropriate input variable for flyrock modelling. In addition to classical performance indices, standardized and symmetric accuracy metrics were quite useful in model validation process. The intelligent models predict flyrock range with an error less than 6 m. The mean percentage errors are lower than 10%. ANFIS seems to be the best model for flyrock prediction. The calculated mean absolute error is 5.36 for ANFIS model.

Nov 12, 2023

By Arman Hazrathosseini, Ali Moradi Afrapoli

Fleet management systems (FMSs) as the pivotal part of any surface mining operation find it essential to evolve from conventional into intelligent systems because of both Mining 4.0 requirements and some shortcomings inherent in conventional methods. However, this novel transformation needs to be investigated technically and strategically. To this end, the previously published research works on intelligent FMSs are explored to track the latest status of these third generation frameworks within the surface mining context. Following that, their underlying models are compared in terms of five categories of allocation and dispatching features to pinpoint the technical gaps ignored. Having drawn the future lines of research, the present article then leverages the popular SWOT analysis method to outline the strengths, weaknesses, opportunities, and threats associated with the advent of these intelligent FMSs in mines of future. By and large, the analysis indicates that advantages outweigh disadvantages. Solutions are offered to address the existing weaknesses and threats.

Nov 10, 2023

By Weiran Zuo, Jinsheng Guo, Chun Yu, Jinyu Zhan, Bao Guo

High-precision prediction of the target minerals’ content in the feed and concentrate products is vitally important for the efficient beneficiation of mineral resources. Visible and near-infrared (NIR) spectroscopy provides a cost-effective way for online measurement of the mineral content in an industrial process. In this investigation, simulated specimens consisting of five different types of minerals that are present in the copper oxide ore, malachite, quartz, calcite, hematite, and chlorite were prepared with a variety of malachite contents, and the mixed specimens were analyzed by a visible-NIR spectrometer in an integral sphere mode. The reflectance spectrum is used as input and the malachite content as output to build the prediction model. The obtained data was modeled by support vector machines (SVM), and a Grey Wolf Optimization (GWO) is proposed with the goal of improving the prediction accuracy. The GWO algorithm has been applied to adaptively search for the best combination of featured values. After cyclic comparison, the optimal penalty factors c and g can be quickly and accurately selected. The experimental results show that the SVM model established by the GWO algorithm has a better fitting effect and smaller prediction error, compared with other models.

Aug 3, 2023

By A. Adhikari, P. Tukkaraja, S. Jayaraman Sridharan

Shafts are critical components of the mine ventilation systems and contribute significantly to the mine ventilation pressure. Estimation of shaft pressure losses is an important aspect of mine ventilation planning and shaft frictional pressure losses are extensively studies. Shock losses contributed by shaft buntons are estimated using the interference factor model developed half a century ago. This study aims to develop a better performing prediction model for shaft buntons using state of the art experiment and numerical data, and machine learning techniques. Results show that the present prediction model is performing poorly, and the prediction performance can be improved by incorporating additional predictor variables. The Artificial Neural Network standardized Machine Learning model developed in this study is able to predict the shaft bunton shock loss with high accuracy (R2 = 0.956).

Jun 25, 2023

By Robert Bissonette, Samir Sbai

A critical component of implementing autonomous systems is the ability to use information from a robust sensory and information system to make sound, predictable, and repeatable decisions to assure safety. This means that any such system would need to evaluate what objects (people, equipment, boulders, etc.) of interest exist in that environment (with relevant data about identity, trajectory, characteristics, etc.), project a future continuum of those objects (including considerations for variabilities), and make control intervention choices to minimize risk due to exposures in the present operational area. The components of an overall system would include: a) sensor and data fusion, b) probability projection of future states, c) generation and evaluation of alternate futures, and d) implementation of best course of action to reduce risk. The National Institute for Occupational Safety and Health (NIOSH) Spokane Mining Research Division (SMRD) is studying this framework for applications for autonomous mining equipment. This paper focuses on using information from the sensor fusion engine to generate a continuous projection of future and alternate states (b and c above) incorporating the factors of confidence, accuracy, tolerance, and other variabilities. We refer to this as Machine Situational Awareness (MSA).

Jun 25, 2023

By RETC2023

RETC2023

Jun 13, 2023

By M J. Nimmo

extremely powerful tools for building predictive and generative models. ML can be used for building highly accurate regression and classification models. But without careful data science and statistical analysis, a regression or classification model could be highly biased and completely wrong – and we may not even realise. An example rock density estimation task will be used to illustrate the potential gains and possible pitfalls in using ML and AI in Mineral Resource estimation. What could possibly go wrong? Nothing! For most simple regression or classification problems, very little. For density estimation, using the global mean may be all that is possible, but the resource classification would need to reflect the uncertainty in global estimation of tonnage. Everything! From data collection and measurement, from selecting ML and AI techniques and designing the analysis workflow to our own cognitive bias. The things that could go wrong are vast and may include but not limited to asking the wrong question, not asking questions, data bias, measurement errors, insufficient data coverage, not collecting the right variables, not collecting enough observations, collecting clustered samples, unbalanced data, filtering out data without reason (outliers), clipping data (outliers), relying on automatic feature selection, including irrelevant variables, excluding relevant variables, geological interpretation, changing context, blindly following best practice, splitting small data sets, subsetting large data sets, using the wrong tool for the task, insufficient budget to complete the analysis, not allowing enough time for testing and experimentation (the science), focusing too much on building the model (the engineering), not learning from the data, our assumptions, our preconceptions, and our skill. This paper will explore the question of how training regression models to predict rock density are affected by gaps in the data (missing observations and missing variables). What could possibly go wrong?

May 24, 2023

By I Sucholutsky, D M. First, D Mogilny, F Yusufali

Machine learning is creating value in all facets of the mining industry, from exploration to production. The authors provide an accessible, high-level introduction to artificial intelligence (AI), machine learning (ML), and deep learning (DL); the latter being widely recognised as one of the most powerful forms of ML. In particular, the authors will introduce deep learning models known as convolutional neural networks (CNNs), how they are applied, and the economic considerations necessary for determining when DL may be the right solution to de-risking complex block modelling problems. The authors present preliminary results from the mineral resource modelling study of the Jundee orogenic gold deposit, Yandal belt/gold province, Western Australia. The primary goal was to identify the direction and location of the narrow gold veins more accurately and demonstrate how non-linearly correlated elements are used as direct inputs to the resource model to assist with the target element’s grade prediction. This demonstrates that: (1) existing techniques for finding correlations between assayed elements do not adequately reflect the complex geology of the asset, (2) nonlinear correlations that are difficult to model as simple mathematical functions are representative of geological patterns in a deposit, and (3) non-linearly correlated assayed data, fed as inputs, increase the performance of the resource model as reconciled through blind tests. To conclude, the authors hypothesise that the patterns represented by the DL block models may be revealing the results of overprinting geological processes that generated mineral deposits. For example, the primary hydrothermal processes that deposited metals created depositional patterns that become particularly complex as a result of being overprinted, in part or in whole, by secondary physicochemical processes. This may explain why non-linear geochemical relationships have the capacity to generate more accurate block models of mineral deposits.

May 24, 2023

By A. Jahedsaravani, M. Massinaei, M. Zarie

Deep learning is a subset of machine learning that uses artificial neural networks for extracting high-level features from image data. In the present study, a soft sensor is proposed for the prediction of the flotation performance through froth features generated by the use of pre-trained convolutional neural networks. Several state-of-the-art convolutional neural networks (AlexNet, GoogLeNet, VGGNet, ResNet, and SqueezeNet) pre-trained on the ImageNet database are used to predict the metallurgical performance of two flotation systems. The first case study is a batch copper flotation system video-captured over a wide range of process conditions. The second case study is an industrial coal flotation column equipped with a continuous video recording system. The pre-trained networks are used to extract features from the froth images, and these features are subsequently used to predict the flotation conditions and performance. The prediction results by the pre-trained algorithms were compared with the traditional image processing algorithms. This demonstrates the ability of the pre-trained structures to generalize to images outside the ImageNet database. GoogLeNet outperforms other network architectures and provides more accurate predictions of the flotation process behavior and performance.

May 5, 2023

By Wei Victor Liu, Na Zhang, Chengkai Fan, Bei Jiang

The truck haulage data from open-pit mine sites are usually massive and multidimensional with multi-peak Gaussian distributions. Artificial neural networks (ANNs) are well-known machine learning algorithms to handle massive and multidimensional data for building models. Moreover, Gaussian mixture modeling (GMM) is a suitable option for processing the data under multi-peak Gaussian distributions and improving model accuracy. For the first time, this study used a back propagation neural network (BPNN), an extreme learning machine (ELM), and a Bayesian regularized neural network (BRNN) coupled with GMM to deal with the complex truck haulage data and build three weighted ensemble models (WE-BPNN, WE-ELM, and WE-BRNN models) to predict truck productivity at mine sites. Decision tree (DT), random forest (RF), gradient boosting machine (GBM), and extreme gradient boosting (XGBoost) were used to build models to be compared with the weighted ensemble models. The results showed that the WE-BRNN had a higher accuracy than the WE-BPNN and WE-ELM models. The proposed weighted ensemble models performed better than the benchmark models in predicting truck productivity, indicating that a weighted ensemble approach based on the GMM analysis significantly improved the model accuracy. Based on the relative importance analysis, haul distance was the most crucial input variable for predicting truck productivity. This study provides a new approach to predicting truck productivity, which will help mining companies make sound budget decisions and improve mine planning.

Mar 2, 2023

By Nadya Michel, Daniel Gros-Jean

This is a true story. Names have been changed to illustrate the importance and impact of carbon monoxide risk management in explosive blast projects and the community « Honey I’m not feeling good…” “What is going on Michael?” – – Michael was not answering so Julia, who was on the third level of the magnificent house, hurried down the stairs to check up on him in the basement. Today was day 2 of the construction work of the bathroom, yesterday they had excavated all the space for the shower. She started to breathe heavily as she was approaching the bathroom and saw Michael unconscious on the floor. “Michael!” – Julia immediately dialed 911. She opened a window since the air was heavy and went by Michael to hear him breathe and take his pulse. The paramedics took Michael on a stretcher and rushed him to the hospital. “I’m feeling dizzy” she thought to herself a couple of hours after the incident, and heard Matthew say: “Mom, I’m feeling sick”. Suddenly, she felt like she would pass out and felt in danger but couldn’t move. She desperately searched for her cellphone and dialed 911 but couldn’t speak. The dispatcher sent the firefighters and paramedics to the GPS location that was appearing on the screen and the emergency team was alerted that it was a second call in hours at the same address. When the firefighters arrived, they had to force the door to enter and found Julia in her bedroom on the third floor and Anette and Matthew also unconscious in their bedroom, lying on their respective bed. There was no smoke but on their multigas meter, carbon monoxide (CO) peaked at 1500 ppm on every level of the house, which is abnormally high.

Feb 6, 2023

By Steve Linehan, John Toedter, Kelly Williams, Mark Herman, Laura Connor, Catherine E. Johnson

Minimum safe distance (MSD) calculations are used widely in military applications to determine safe standoff distances for breaching personnel. Accurate safe distances and an understanding of blast wave propagation are critical to preventing traumatic brain injury. These calculations are often based on spherical charges in open air. However, applications including explosive breaching use planar charges rather than spherical. Previous studies analyzed the shock propagation of a square planar charge of sheet explosive and concluded the anisotropic pattern conflicted with the Kingery Bulmash (KB) and MSD spherical models. It has been determined that these predictions do not adequately predict the realistic pressures in all directions and the trend has shown to be consistent across multiple charge geometries. In this series of experiments, pressures were collected from reference spheres and rectangular geometries of detonating cord and sheet explosive commonly used in breaching and improvised demolition for this comparison. This common configuration showed pressure differences at varying angles to the charge with the highest pressures measured at the 90° position perpendicular to the “door” to be breached. Lower pressures were found at the 0° and 15° positions where the breachers would normally be standing. The reflected shock from a surface such as a door was incorporated in a semi-confined situation and added to the shock observed by the breachers. Urban environments include many surrounding structures for these reflections to occur and further testing is necessary in such an environment to determine the full-scale impact. These findings may be useful to develop improved MSD calculations for breaching applications based on charge geometry, orientation to the blast, and nearby structures.

Feb 6, 2023

By B. Eiter, M. Barham, T. Bauerle

Fatigue-related risk is a persistent safety concern for the mining industry. However, fatigue and sleepiness are often treated interchangeably, which can lead to confusion and potentially less effective reduction of safety risk. To provide clarity, we present an overview of similarities and differences between work-related fatigue risk and sleepiness including definitions, theories, measurements, and mitigation strategies. As a supplement, a summative visual model which highlights these similarities and differences is presented. Expanding industry knowledge in this area will assist safety professionals in crafting more targeted risk management practices appropriate for work-related fatigue risk, sleepiness, or both.

Feb 1, 2023

By Jian Wang, Kun Li, Hao Wu, Juan Nan

In this paper, in order to solve the problems of low accuracy and slow speed of fire flame recognition, an extreme learning machine (ELM) method based on improved Harris hawks optimization (IHHO) is proposed for fire flame recognition. A novel Harris hawks optimization (HHO) is used to solve the problem of parameter selection of ELM. In order to solve the problem that the original HHO is prone to fall into local optimum, firstly circle mapping is used to initialize the population to solve the problem of uneven distribution and small range of the initial population. Then the formula of escaping energy of HHO is modified to make the population have a large range in the middle and late iterations while gradually decreasing in the whole iteration process. Thus, the exploitation stage in HHO is improved to make the search range smaller near the optimal solution to accelerate the convergence process. Finally, at the end of each iteration, a certain number of individuals are selected to perform Gaussian and Cauchy hybrid mutation to prevent the IHHO from falling into local optimization. Through three groups of experiments, the effectiveness of the proposed IHHO and IHHO-ELM is verified. In experiment 1, the convergence performance of IHHO is significantly better than that of the other remaining algorithms in the test results of six benchmark functions. In experiments 2 and 3 of real fire flame recognition case, IHHO-ELM outperforms other remaining algorithms on the whole and has significant advantages in some indexes.

Jan 17, 2023

By Kristian E. Waters, Hao Ma, Haibei Wang

Over the past few decades, ELM based separation processes have become an attractive and efficient way to remove toxic metal ions from the various aqueous waste effluent streams. This paper provides a comprehensive review of the generation, characterization, and applications of emulsion liquid membrane (ELM) from the more than 280 papers and books published so far. First, the ELM formation and the transport mechanism are introduced. Then the ELM stability and emulsion splitting methods are discussed in detail, followed by the latest research and findings of the extraction of copper and nickel using the ELM technique during the past few decades, including a thorough discussion of ELM operating parameters that impact ELM stability and extraction. In addition, the kinetics study and industrial application of this technique for the removal of coppers/nickel solutes are also presented and discussed.

Jan 6, 2023