Dual energy gamma-ray transmission techniques applied to on-line analysis in the coal and mineral industries

doi:10.1016/0020-708X(85)90020-1

The International Journal of Applied Radiation and Isotopes

Volume 36, Issue 11, November 1985, Pages 867-877

https://doi.org/10.1016/0020-708X(85)90020-1 Get rights and content

Abstract

The use of dual energy γ-ray transmission techniques to determine the concentration of a higher atomic number (Z) component in a lower Z matrix is investigated. Examples discussed are the determination of lead in lead-zinc ore, iron in high-grade iron ore, uranium in solution, and ash in coal. These examples are considered in relation to the main application fields of on-stream analysis of mineral slurries and solutions, on-line analysis on conveyors, and scanning of borecores.

A method for reducing errors in analysis by compensating for variations in concentration of some of the wanted or matrix components is proposed. This method depends on choice of one γ-ray energy in the photoelectric absorption region and the other in the pair production region. Calculations show that this method reduces errors in the determination of ash in coal caused by variation in concentration of iron in the ash. However, sensitivity to ash is poor, so this application is suitable only when γ-ray path lengths in the coal are large.

Preliminary results are presented of the combination of dual energy γ-ray transmission and pair production techniques to give an estimate of iron concentration in coal. This can be used to give an approximate estimate of sulphur in pyrites in coal, assuming that iron is present only as pyrites.

References (16)

R.A. Fookes
Int. J. Appl. Radiat. Isot
(1983)
J.H. Hubbell
Int. J. Appl. Radiat. Isot.
(1982)
J.S. Watt
Int. J. Appl. Radiat. Isot.
(1983)
J.S. Watt
Int. J. Appl. Radiat. Isot.
(1983)
R.J. Holmes et al.
Anal. Chim. Acta
(1983)
A. Whittaker et al.
R.A. Fookes
M. Kato
ORNL-11C-10
(1967)

There are more references available in the full text version of this article.

Cited by (27)

Study on photofluorescent uranium ore sorting based on deep learning
2024, Minerals Engineering
Uranium ore stands as a vital source of energy, and the process of ore sorting serves to segregate valuable uranium ore from its less valuable counterparts. This not only enhances the intrinsic ore grade but also mitigates the challenges associated with subsequent refining processes, thus leading to substantial resource cost savings. Consequently, the necessity of pre-sorting uranium ore becomes evident. While prevailing X-ray-based ore sorting technology remains at the forefront, the quest for a more efficacious method to sort uranium ore remains ongoing. This study harnesses the inherent photofluorescence properties of uranium ore by subjecting it to ultraviolet light, inducing the emission of a vibrant green fluorescence. Consequently, we propose a novel sorting technique founded on photofluorescence, augmented by deep learning to enhance the precision and categorization of uranium ore images. Two distinct models, the Convolutional Neural Network (CNN) model and the Transformer model, are developed in tandem. Empirical findings underscore the superiority of the Swin Transformer model, with training accuracy, validation accuracy, and classification rate registering at 0.9780, 0.9719, and 96.4 ms/it, respectively, surpassing the performance of alternative models like Alex Net and Res Net 34. Furthermore, the study employs visual aids, including channel visualization diagrams, confusion matrices, and Grad-CAM diagrams, to elucidate the operational intricacies of the model in the sorting of photofluorescent uranium ore images. This allows for an insightful analysis of the characteristics that influence classification weights. The potential applications of this method are indeed substantial, opening new avenues for uranium ore exploration and utilization.
A review of deep leaning in image classification for mineral exploration
2023, Minerals Engineering
Efficient sorting and optimal utilization have become the common core issues for today's mining industry. Vision-based sorting technology provides a powerful answer to this problem. Deep learning, as one of the outstanding branches of machine learning, makes it possible to design, deploy and apply the high-performance vision models, and the architecture represented by convolutional neural networks again improves the accuracy and speed of mineral sorting. Based on this, we carefully review and survey the research and applications of deep learning in digital mineral image classification between 2017 and 2022, and state the evolution and innovation of it in terms of mineral types, data qualities, and model architectures. The scope considered in this review is to predict one class label for the entire image rather than the segmentation task of pixel-by-pixel classification. This work further outlines several problems that still stand out in current industrial implementation, summarizes the relevant solutions, and discusses the application potentials of deep learning in mining domain.
Deep learning in image segmentation for mineral production: A review
2023, Computers and Geosciences
Mineral image segmentation is widely used in mining, sorting, exploration, composition analysis, and other production works. The burgeoning field of deep learning provides preferred solutions for mineral image segmentation. We present a review of recent literature in this direction, covering the module components, encoder-decoders architecture, representative networks, mineral image datasets, performance metrics, and state-of-the-art models. In the application performance survey, the review contents include mineral type, image type, image resolution, image data quantity, architecture selection, and encoder network construction, as well as summarizes the advantages of deep learning-based mineral image segmentation methods. We conducted small-scale experiments for the current mainstream architectures and visualize the segmentation results for performance comparison. We also investigated the application challenges and bottlenecks of deep learning-based methods, propose several innovative directions, and discuss promising future applications.
Research on on-site measurement factors and performance of coal calorific value based on laser ignition
2023, Fuel
The laboratory method used for coal calorific value measurement has a significant hysteresis. In this paper, an on-site measurement method of coal calorific value based on high power laser rapid ignition is proposed. The effects of oxygen transverse passage velocity, thermal insulation crucible bracket and coal samples on the measurement time, unburned rate and calorific value measurement error are gradually investigated. The experimental results shows that the measurement error of calorific value is within 2% and unburned rate is within 2% at the oxygen transverse passage velocity of 120 mm/s of Huainan bituminous coal. Then the utilization of porous thermal insulation crucible bracket has shown a good performance in heat dissipation to shorten the measurement time without increasing the unburned rate. Then, the effects of coal sample thicknesses, species and particle sizes on calorific value measurement are investigated. Results show that under the coal thickness from 4.2 to 5.4 mm, the calorific value measurement error is within 3%, and the unburned rate is within 2% of Huainan antiminous coal. Finally, the results shows that the method has a good performance on different coal species and particle sizes which the calorific value measurement error is within 5%. This research quantifies the effects of the experimental condition parameters on the measurement of coal calorific value based on laser ignition, which lays the foundation for the subsequent development of online coal calorific value meter.
Ore image classification based on small deep learning model: Evaluation and optimization of model depth, model structure and data size
2021, Minerals Engineering
Citation Excerpt :
At this stage, intelligent dry separation equipment has been applied in underground gangue discharge or ore pre-separation, significantly reducing the energy consumption in the crushing process and reagent consumption in the flotation process, further improving production efficiency and reducing pollution treatment cost. In the current market, the mainstream intelligent ore sorting equipment used into production are based on ray-based sensors, such as γ-ray (Watt and Steffner, 1985), Raman spectrum analysis (Ishikawa and Gulick, 2013), XRT (Robben et al., 2020), XRF (Li et al., 2019). Radiation-based sensors have high classification accuracy and are mostly used to identify and separate large ore blocks.
The ore image classification technology based on deep learning is an effective way to improve the image sensor-based ore sorting classification capability. However, in practice, the image sensor-based ore sorting technique often has the problem of insufficient data, and has not systematically considered the impact of model structure and dataset size on the modeling efficiency and classification performance of deep learning. Therefore, this paper attempts to explore a more suitable small deep learning model for ore image classification by considering the model depth, model structure, and dataset size. Six Convolutional Neural Networks (CNNs) models are established with different depths based on Alex Net and VGG Net and the model structure is optimized by adding BN layer. Taking the gas-coal image dataset as case study, we systematically explore the influence of model depth, model structure, dataset size on the training process efficiency and classification accuracy. Meanwhile, the operational process of coal image classifiers is analyzed visually through the ways of Channel Visualization maps, Heatmaps, Grad-CAM map, and Guided Backpropagation maps.
Performance evaluation of a deep learning based wet coal image classification
2021, Minerals Engineering
Citation Excerpt :
At present, sensor-based ore sorting technology is mainly divided into two types: ray sensor-based and machine vision-based. Specifically, the ray sensor-based ore sorting equipment mainly takes γ-ray (Watt and Steffner, 1985), Raman spectrum (Das and Hendry, 2011; Frost, 2003; Ishikawa and Gulick, 2013), X-ray diffraction (XRD) (Robben et al., 2020), and X-ray fluorescence spectrometry (XRF) as the technical foundation (Karna et al., 2016; Knapp et al., 2014). The machine vision-based ore sorting equipment mainly consists of the visible light sensor and the near-infrared sensor (Grant et al., 2018).
Moisture is one of the important influencing factors on machine vision-based mineral image classification, and it has different effects on various ore particles. At present, deep learning is an effective measure to improve classification accuracy, but the effects of moisture have not been systematically investigated. Therefore, this paper establishes deep learning-based RGB image classification models for the classification tasks of various coal particles with two density level (<1.8 g/cm³ & >1.8 g/cm³) in different water gradients, and analyzes their classification performance. Moreover, the model operational process and the change of classification weight and accuracy under different water gradients are investigated through Channel Visualization, Heatmap, Guided Backpropagation, Grad-CAM.

View all citing articles on Scopus

View full text

Dual energy gamma-ray transmission techniques applied to on-line analysis in the coal and mineral industries

Abstract

Int. J. Appl. Radiat. Isot

Int. J. Appl. Radiat. Isot.

Int. J. Appl. Radiat. Isot.

Int. J. Appl. Radiat. Isot.

Anal. Chim. Acta

ORNL-11C-10