Search Documents

By Mei Long, Tawum Juvert Mbah, Jianhua Zhang, Haiwang Ye

There have been many improvements and advancements in the application of neural networks in the mining industry. In this study, two advanced deep learning neural networks called recurrent neural network (RNN) and autoregressive integrated moving average (ARIMA) were implemented in the simulation and prediction of limestone price variation. The RNN uses long shortterm memory layers (LSTM), dropout regularization, activation functions, mean square error (MSE), and the Adam optimizer to simulate the predictions. The LSTM stores previous data over time and uses it in simulating future prices based on defined parameters and algorithms. The ARIMA model is a statistical method that captures different time series based on the level, trend, and seasonality of the data. The auto ARIMA function searches for the best parameters that fit the model. Different layers and parameters are added to the model to simulate the price prediction. The performance of both network models is remarkable in terms of trend variability and factors affecting limestone price. The ARIMA model has an accuracy of 95.7% while RNN has an accuracy of 91.8%. This shows that the ARIMA model outperforms the RNN model. In addition, the time required to train the ARIMA is than that of the RNN. Predicting limestone prices may help both investors and industries in making economical and technical decisions, for example, when to invest, buy, sell, increase, and decrease production.

Jan 6, 2021

By D. R. Hanson, H. Lawson

In this study, NIOSH researchers conducted a machine learning analysis to examine whether a model could be constructed to assess the probability of dynamic failure occurrence based on geochemical and petrographic data. Random forest, cluster analysis and dimension reduction were all applied. The objective of dimensionality reduction was to explore patterns and groupings in the data and search for relations between compositional parameters. Cluster analyses were performed to determine if an algorithm could find clusters with given class memberships and to what extent misclassifications of dynamic failure status occurred. A random forest analysis performed on data from the Pennsylvania Coal Sample Databank cross-referenced with accident data from the Mine Safety and Health Administration (MSHA) determined that 7 parameters of the 18 examined exerted the most influence on results. Cluster analysis on data after dimensionality reduction resulted in a hierarchal clustering algorithm finding four clusters, with one relatively distinct dynamic failure cluster, and three clusters consisting mostly of control group members but with a small number of dynamic failure members.

Feb 1, 2023

By J Herd, J Rowe, M Cutbush, d F. Aziz

WearApp provides expertise for assessing the degradation of removable screening media panels on vibrating screens right in a user’s pocket. Through the WearApp application, users can simply take a picture of a panel using their mobile, enter some basic information and have the application assess the wear to provide a recommendation for whether the panel needs to be replaced. Where traditional mechanisms for assessing panel wear only measure a few points across the panel face and rely on having personnel with specialised knowledge available to perform the assessment, WearApp uses deep learning models to analyse the entire panel face and support the user to make an informed decision. The application works through analysing a panel image and using a model to automatically detect the corners of the panel. Given the known dimensions of the panel, this enables the application to calibrate the pixel size against a known reference. A second set of models then detect the apertures and measure their dimensions. With the aperture dimensions defined the application assesses the wear of the panel through referencing a stored 3D model of the aperture. Finally, the data from each aperture is collated and the wear of the entire panel face defined. Since WearApp performs measurement of the entire panel face, it also allows the user to perform a more detailed analysis of the panel’s wear profile. Where typically, the panel wear is only measured in a few locations, WearApp’s analysis enables users to visualise the panel wear across the entire panel face, identifying areas of high wear that could be leading to premature panel replacement. Additionally, because the entire panel face is measured the change in screening efficiency can be calculated accurately for the worn panel, even taking into account the impact of aperture pegging by small rocks.

Sep 1, 2024

By M. E. Rossi

Reconciliation of production information is critical to evaluating the effectiveness of the predictive models, and to allow for optimization of mining operations. Whether mining open pit or underground, mine-to-mill reconciliations can be one of management's better tools to perform proper accounting, and a very useful evaluation and optimization tool. This paper proposes a set of criteria and objectives for a typical reconciliation program; also presents a stepwise, logic approach to performing reconciliations, and discusses the benefits and costs associated with the up-keeping of the information. It also highlights some of the potential pitfalls involved, and methods used to avoid collecting misleading information. These issues are illustrated with an example of a model-to-mine-to-mill reconciliation program being implemented at an operating mine; the program involves multiple predictive models (Long-term and Short-term block models), different open pit and underground mines, stockpiling, and at least 23 different mill streams. This paper presents the initial results of the program implemented at Minera Michilla S.A.

Jan 1, 1999

By Tom Gabardi, Peter Clarke

"MillMapperTM is a new 3D laser scanning technology that has only recently been introduced into the North American mining industry. It enables mill operators to gain a much more complete understanding of the condition of their mill internals and to more accurately predict how liners will perform in the future. This paper describes the technology and techniques used to obtain data on internal mill conditions, how this data is processed into meaningful information which enables better prediction of liner life and gives examples of the reports provided by the software. Examples are given, using client sites data, on how this information is being used to reduce the costs and improve the revenue from SAG and ball mills.INTRODUCTION Scanalyse Pty Ltd developed the MillMapperTM technology in Australia over a number of years and has recently introduced it into North America with a significant take-up in Canada. The technology was commercially released in Australia in late 2006 and there has been significant technical development in the features provided since that time. MillMapperTM has been accepted as a useful technology in the Australasian region and more recently has been introduced to sites in Chile, Peru, Brazil, USA, Canada, Laos and South Africa. Over 100 different mills have been scanned and the technology has been thoroughly tested and validated. It has become an integral tool in condition monitoring and improvement programs for SAG and Ball mills across many sites and is now providing significant and measurable benefits to the mining industry. Some of these benefits will be described in this paper using actual examples."

Jan 1, 2012

By James F. Davis, Matthew A. Conte, Robert F. Tonning

This is a paper demonstrating the steps taken to combine multiple types of piles for a single site to maximize the benefits and characteristics of each method. We intend to discuss the use of grouted displacement piles in lieu of traditional pile types that were specified on an individual project. We will show how implementing this alternate pile type can both supplement and replace the originally specified pile to increase efficiency and lower costs. We will demonstrate a reduction in spoils, vibration, crew hours, and equipment size while employing methods to overcome unforeseen ground conditions.

By Brendan Demich, Brendan Macdonald, Timothy J. Orr, Jonathan K. Hrica, CASSANDRA L. HOEBBEL, Jason Navoyski, Jennica L. Bellanca

The paper describes the process of collecting detailed accounts of near-miss incidents and turning them into impactful visual stories using graphic simulation. The aim is to create high-quality training materials to reduce haul-truck-related accidents and fatalities in mining.

Sep 1, 2023

By J. K. Hrica, T. J. Orr, J. Navoyski, B. Macdonald, C. L. Hoebbel, J. L. Bellanca, B. Demich

Haul truck fatal accidents and injuries continue to be a significant concern for the mining industry. However, the availability of high-quality training materials continues to be limited. Near-miss incident accounts, if packaged well, could help fill this gap, because for every fatality, there are hundreds of reportable accidents and thousands of undocumented near misses. Researchers from the National Institute for Occupational Safety and Health (NIOSH) collected detailed accounts of 21 near-miss incidents in virtual interviews with mineworkers at surface mining operations across the country. From these interviews, researchers created four simulation videos. The videos bring these events to life through first-person retelling and various visual perspectives of actual events. Each video exemplifies a critical safety message and a common haul truck hazard. This paper describes the process of taking narratives and turning them into impactful visual stories. NIOSH plans to co-release these simulation videos with the Mine Safety and Health Administration (MSHA) to the mining industry to raise awareness and ultimately help reduce haul truck-related accidents and fatalities in mining.

Feb 1, 2023

By Brendan Demich, Brendan Macdonald, Timothy J. Orr, Jonathan K. Hrica, CASSANDRA L. HOEBBEL, Jason Navoyski, Jennica L. Bellanca

Haul truck fatal accidents and injuries continue to be a significant concern for the mining industry. However, the availability of high-quality training materials continues to be limited. Near-miss incident accounts, if packaged well, could help fill this gap, because for every fatality, there are hundreds of reportable accidents and thousands of undocumented near misses. Researchers from the National Institute for Occupational Safety and Health (NIOSH) collected detailed accounts of 21 near-miss incidents in virtual interviews with mineworkers at surface mining operations across the country. From these interviews, researchers created four simulation videos using the Unity game engine. The simulation videos bring these events to life through first-person retelling and various visual perspectives of actual events. Each video exemplifies a critical safety message and a common haul truck hazard. This paper describes the process of taking narratives and turning them into impactful visual stories using graphic simulation. NIOSH plans to co-release these simulation videos with the Mine Safety and Health Administration (MSHA) to the mining industry to raise awareness and ultimately help reduce haul truck-related accidents and fatalities in mining.

Jun 1, 2023

By L. Hales

Neural networks are a generalized, multivariable non-linear modeling technique that can be used to model mineral processing operations. Neural networks will be described as will their use in modeling column flotation. Optimization of column flotation will be explored and discussed by using the neural network column models.

Jan 1, 1996

By Felix Degold, Mark Ganster

Achieving a customer optimized fragmentation by using robust data from new technologies e.g. GPS positions of boreholes, fragmentation analysis to calibrate the geology and fragmentation prediction models. The inclusion of the collected data into the blast’s design makes it possible to accomplish a desired particle size distribution.

Feb 1, 2020

By Ping Fang, Dezhou Wei, Zichuan Guan, Wengang Liu, Yanbai Shen

A new cationic collector, bis(2-hydroxy-3-chloropropyl) dodecylamine (N23), that can be dissolved in water easily was synthesized successfully. Its flotation performance to quartz was investigated compared with dodecyl amine in hematite reverse flotation system. It was found that N23 showed stronger collecting performance to quartz than dodecyl amine. The maximum recovery of 98% to quartz was obtained at a collector concentration of 43.3 mg/L in the pulp pH range of 6-9. This collector can be used to float quartz from hematite when starch was used as depressant. Optimal separating results with 90.06% iron recovery and 66.3% iron grade to artificially mixed sample were achieved at a collector concentration of 43.3 mg/L and a starch concentration of 23.3 mg/L at neutral pulp pH when hematite and quartz were mixed at a mass ratio of 3:2. The adsorption mechanism of the collector on quartz surface was investigated by infrared spectroscopic analysis and zetapotential measurements, which revealed that electrostatic attraction and hydrogen bonding adsorption occurred between the collector and quartz surface.

Jan 1, 2014

By Sylwia Zymankowska-Kumon, Rafal Danko, Józef Danko, Mariusz Holtzer

"Burning of ceramic materials is carried out at temperatures 900+ 1300°C, in dependence of the required functional properties. The Faculty of Foundry Engineering together with the Faculty of Material Engineering and Ceramics developed the technology of production of burned ceramic materials with using spent moulding and core sands. The mentioned above waste materials of the foundry industry fulfilled -in plastic masses intended for obtaining ceramic materials -the role of the addition weakening clay raw materials. The laboratory investigations were fully verified by the results of industrial tests. It should be pointed out that none negative effects were observed at the stage of mixing raw materials and preparation of plastic masses. The measurements of exhaust gases emissions performed during burning the products containing spent moulding sands as well as during the normal production did not exhibit any significant differences, which could have indicated the harmful impact on the natural environment.IntroductionSands are used by the foundry industry to make metal casting molds and are a mixture of silica sand, bentonite clay and carbonaceous additives or resins. During the casting process, the molten metal cause thermal decomposition of carbonaceous additives or resins, which results in the formation of potentially hazardous organics, that are emitted to the atmosphere and condense in the molding sand. Each year the U.S. foundry industry landfills about 10 million tones of sand because it is no longer suitable to make metal casting molds. A large high-silica sand content in spent foundry sands (SFSs) generates rational premises for their reuse in other than foundry practice fields. Efforts to divert the SFSs from landfills and encourage their beneficial use in manufactured soils, geotechnical applications, the construction industry (road construction, highway construction), the building materials industry (cement, bricks, limestone manufacture), in the filling of mining cavities or in landfill construction (roads on landfills, permanent covers) are make [1-7]."

Jan 1, 2013

By C. H. Buck, A. J. Taylor

"In 2015 Vale’s Clarabelle Mill began the process change to enable a higher degree of Cu separation. The flowsheet modification was part of a complete divisional flowsheet change being implemented in Vale’s Sudbury Operations. This paper will centre on some of the Operational Readiness (OR) work, culminating in the commissioning of the circuit in March 2017. The emphasis will be on the implementation of an integrated test plan to prepare for the process flow change through the mill.BackgroundClarabelle Mill (CBM) began making a saleable Cu concentrate in 2006 with the installation of the Cu separation circuit. Typical separation rate was 35-40% of Cu in feed reporting to Cu concentrate based on a target Ni concentrate Cu:Ni target of 0.7-0.8. The remainder of the recovered Cu (55-60%) reported to Ni concentrate and was smelted through the Copper Cliff Smelter.In 2013 a capital project, Surface Facility Upgrade (SFU), was implemented to focus on matching the surface plants capacities with the future mine plan to maximize revenues. This was to be achieved by redistributing the metal flow, specifically Cu, through CBM to increase metal unit content through the Smelter. The amount of redistribution of Cu was determined to be based on maintaining a stable Cu to Ni ratio (Cu:Ni) in the Ni concentrate.The change in Cu:Ni ratio in the Ni concentrate would affect the Sudbury divisional flowsheet, and required process changes across the all of the surface plants. The detail that surrounds the flowsheet changes as a whole will not be discussed, but the outcome of the decisions plays a large role in the operational readiness interactions between the mill, its suppliers and its downstream customers."

Jan 1, 2018

By T. G. Joseph, M. J. Bolster

ABSTRACT This paper explores passive seismic measurement as an alternative, nonintrusive approach to conducting annual brake testing on surface mine sites. Dictated by the test site location, this evaluation adhered to the brake testing procedure as per British Columbia occupational health and safety regulations. Results showed excellent correlation to those reported via conventional practice: Comparing stopping distance, time, and velocity for the proposed instantaneous passive seismic versus conventional approaches, less than a 0.5% variance was evident. A viable alternative, the passive seismic approach, allowed an entire fleet of trucks to be sequentially evaluated for annual brake testing without disrupting operations by installing mechanical testing equipment.

Jan 1, 2012

By P. Eger

In 1991, LTV Steel Mining Company decided to close an open pit taconite mine in northeastern Minnesota, using a passive treatment approach consisting of limiting infiltration into the stockpiles and wetland treatment to remove metals. Over 50 million metric tons of sulfide-containing waste had been stockpiled adjacent to the mine during its 30 years of operation. Drainage from the stockpiles contained elevated levels of copper, nickel, cobalt and zinc. Nickel is the major trace metal in the drainages, and prior to closure annual median concentrations ranged from 1.5-50 mg/L. Copper, cobalt and zinc are also present but are generally less than 5% of the nickel values. Median pH ranges from 5.0 to 7.5, but most of the stockpile drainages have pH greater than 6.5. Based on the chemical composition of each stockpile, a cover material was selected. The higher the potential a stockpile had to produce acid drainage, the lower the permeability of the capping material required. Covers ranged from overburden soil removed at the mine to a flexible plastic liner. Predictions of the reduction in infiltration ranged from 40% for the native soil to over 90% for the plastic liner. Five constructed wetlands have been installed since 1992, and have removed 60-90% of the nickel in the drainages. Total capital costs for all the infiltration reduction and wetlands exceeded $6.5 million, but maintenance costs are less than 1 % of those for an active treatment plant. Since mine drainage problems can continue for over a hundred years, the lower annual operating costs should pay for the construction of the wetland treatment systems within seven years.

Jan 1, 1999

By Javankhoshdel Yacoub

Piles are often used to reinforce slopes as they increase the available restoring force when calculating the factor of safety against sliding. In this paper, a 3D model of a slope containing multiple modes of failure is reinforced using pile elements. The results of the 3D Limit Equilibrium analysis containing piles are obtained using a newly introduced coupled approach which combines the results of finite element analysis (to obtain the displacement of soil along the pile) with the traditional limit equilibrium iterative formulation. The multiple modes of failure are obtained using the novel Multi-Modal Optimization (MMO) metaheuristic method which adapts the Particle Swarm Optimization (PSO) method to search for multiple local minima in a single slope model.

Oct 1, 2022

By Jorge Amaya, Nelson Morales Varela, Francisco Saavedra, Gonzalo Nelis, René Gómez

Precedence constraints are broadly used in mathematical models and algorithms for mine planning to model, for example, slope angles in open pit operations or connectivity in underground mines. Unfortunately, these constraints do not capture relevant geometrical aspects for the design of the mine like minimum bottom space of pits or smoothness of the economic envelope in block caves. In this work, we present different approaches that utilize precedence constraints generating economic envelops that are more suitable for design purposes. The numerical experiences show that the proposed techniques produce high value envelops with better geometrical properties and that the methods proposed can be computed efficiently.

Jun 25, 2023

By E. Isokangas

The production of minerals for economic use is a two-stage process, involving mining to extract the mineral from the ground, and processing to convert the mineral into a marketable product. Generally, mining and processing have been viewed as self-contained entities. Both have separate objectives, separate cost centres and key performance indicators (KPIs) that do not reflect the customer/supplier relationships that inherently exist. However, mining and processing operations are inter-connected and therefore intimately inter-related with the performance of one operation affecting the performance of another. Optimizing each stage separately without considering the whole system often misses potential economic benefits and energy savings. During the past fifteen years, the authors have been involved in implementing a holistic methodology ?Mine to Mill Process Integration and Optimisation (Mine to Mill PIO)? to maximize the overall profitability of the operation rather than just optimizing any individual process in a mining operation. Metso Process Technology and Innovation (PTI), along with their project partners, have conducted several projects to significantly increase their production- generating typically 5% to 20% higher throughput and improve the overall mine and concentrator performance through PIO methodology. This proven methodology has applications ranging from greenfield projects to long-standing operations with AG/SAG, HPGR or conventional grinding circuits. This paper explains the Mine to Mill PIO methodology and discusses the benefits of such an approach on the energy consumption, the overall costs and benefits of mining operations. The paper also summarizes several case studies illustrating the use of PTI methodology in a variety of applications. Keywords: integration, optimization, simulation, modeling, mine-to-mill, comminution, blast, drill, crushing, grinding

Sep 1, 2012

By E. Hsieh

Proximity detection in the open pit mine can be used effectively as an operator assist strategy which improves digging production through enforcement of best practices, although it is normally thought of first as a safety tool, and second as a step toward full automation when combined with object avoidance.

Feb 23, 2014