ORIGINAL PAPER
A novel magnetite ore refined sorting method based on magnetic induction and CNN-SK-BiLSTM network
1
School of Mechanical and Electrical Engineering, Jiangxi University of Science and Technology, Ganzhou 341000, China
Submission date: 2024-08-25
Final revision date: 2024-10-31
Acceptance date: 2024-11-16
Publication date: 2025-03-18
Corresponding author
Chunrong Pan
School of Mechanical and Electrical Engineering, Jiangxi University of Science and Technology, Ganzhou 341000, China
School of Mechanical and Electrical Engineering, Jiangxi University of Science and Technology, Ganzhou 341000, China
Gospodarka Surowcami Mineralnymi – Mineral Resources Management 2025;41(1):123-139
KEYWORDS
TOPICS
ABSTRACT
Magnetite ore is a non-renewable resource that needs to be utilized effectively. Refined sorting of magnetite ore is not simply sorting it into good ore or waste ore but finely sorting it into different grades due to its magnetite content, which not only helps to improve its utilization but also reduces the energy consumption of the following process. However, traditional ore sorting methods based on optical sensors, X-ray sensors, and high-resolution cameras are challenging to achieve refined sorting for magnetite ores because of the limitations of their respective detection methods and classification algorithms. To this end, a new detection method for magnetite content is proposed in this paper; the magnetic induction signal of magnetite ore when it passes through an external magnetic field is captured by Hall sensors and made into a quantifiable time-series dataset. Meanwhile, a deep learning classification algorithm CNN-SK-BiLSTM with a multi-scale attention mechanism is proposed, which successfully sorts magnetite ore from a mine in Liaoning Province, China, into four classes finely. The experimental results show that the accuracy of the model is up to 99.44%, and the precision, recall, and F1 scores are acceptable. In addition, comparative experiments between the proposed model and other standard models were conducted. The results show that the performance of the proposed model is significantly better than the others. This paper provides ideas for the study of refined sorting of magnetite ore.
ACKNOWLEDGEMENTS
This work was supported in part by the Science and Technology Research Project GJJ2203618, 2022, Department of Education of Jiangxi Province and the Special Project for Postgraduate Innovation of Jiangxi Province (YC2021-S577).
The data that has been used is confidential.
CONFLICT OF INTEREST
The Authors have no conflict of interest to declare.
METADATA IN OTHER LANGUAGES:
Polish
Nowatorska metoda sortowania rafinowanej rudy magnetytu oparta na indukcji magnetycznej i sieci CNN-SK-BiLSTM
sortowanie rud, indukcja magnetytu, głębokie uczenie się, uwaga
Ruda magnetytu jest zasobem nieodnawialnym, który należy efektywnie wykorzystać. Rafinowane sortowanie rudy magnetytu nie polega po prostu na sortowaniu jej na dobrą rudę lub rudę odpadową, ale na dokładnym sortowaniu na różne gatunki ze względu na zawartość magnetytu, co nie tylko pomaga poprawić jej wykorzystanie, ale także zmniejsza zużycie energii w kolejnym procesie. Jednak tradycyjne metody sortowania rud oparte na czujnikach optycznych, czujnikach rentgenowskich i kamerach o wysokiej rozdzielczości stanowią wyzwanie w celu uzyskania udoskonalonego sortowania rud magnetytu ze względu na ograniczenia odpowiednich metod wykrywania i algorytmów klasyfikacji. W tym celu w artykule zaproponowano nową metodę wykrywania zawartości magnetytu. Sygnał indukcji magnetycznej rudy magnetytu przechodzącej przez zewnętrzne pole magnetyczne jest wychwytywany przez czujniki Halla i przekształcany w wymierny zbiór danych w formie szeregów czasowych. Tymczasem zaproponowano algorytm klasyfikacji głębokiego uczenia się CNN-SK-BiLSTM z wieloskalowym mechanizmem uwagi, który z powodzeniem sortuje rudę magnetytu z kopalni w prowincji Liaoning w Chinach na cztery klasy. Wyniki eksperymentów pokazują, że dokładność modelu sięga 99,44%, a precyzja, powtarzalność i wyniki F1 są akceptowalne. Dodatkowo przeprowadzono eksperymenty porównawcze zaproponowanego modelu z innymi modelami standardowymi. Wyniki pokazują, że wydajność proponowanego modelu jest znacznie lepsza od pozostałych. W artykule przedstawiono pomysły na badania rafinowanego sortowania rudy magnetytu.
REFERENCES (35)
1.
Baraboshkin et al. 2020 – Baraboshkin, E.E., Ismailova, L.S., Orlov, D.M., Zhukovskaya, E.A., Kalmykov, G.A., Khotylev, O.V., Baraboshkin, E.Y. and Koroteev, D.A. 2020. Deep convolutions for in-depth automated rock typing. Computers & Geosciences 135, DOI: 10.1016/j.cageo.2019.104330.
2.
Chen et al. 2021 – Chen, X., Zhang, B., and Gao, D. 2021. Bearing fault diagnosis base on multi-scale CNN and LSTM model. Journal of Intelligent Manufacturing 32(4), pp. 971–987, DOI: 10.1007/s10845-020-01600-2.
3.
Chen et al. 2022 – Chen, Z., Guo, H., Yang, J., Jiao, H., Feng, Z., Chen, L. and Gao, T. 2022. Fast vehicle detection algorithm in traffic scene based on improved SSD. Measurement 201, DOI: 10.1016/j.measurement.2022.111655.
4.
Chen et al. 2024 – Chen, Z., Yang, J., Feng, Z. and Zhu, H. 2024. RailFOD23: A dataset for foreign object detection on railroad transmission lines. Scientific Data 11(1), DOI: 10.1038/s41597-024-02918-9.
5.
Cho et al. 2024 – Cho, H.M., Han, S., Seong, J.K. and Youn, I. 2024. Deep learning-based dynamic ventilatory threshold estimation from electrocardiograms. Comput Methods Programs Biomed., DOI: 10.1016/j.cmpb.2023.107973.
6.
Fuentes et al. 2021 – Fuentes, R., Luarte, D., Sandoval, C., Myakalwar, A.K., Yáñez, J. and Sbarbaro, D. 2021. Data fusion of Laser Induced Breakdown Spectroscopy and Diffuse Reflectance for improved analysis of mineral species in copper concentrates. Minerals Engineering 173(12), DOI: 10.1016/j.mineng.2021.107193.
7.
Henley et al. 2022 – Henley, R.W., Mernagh, T., Leys, C., Troitzsch, U., Bevitt, J., Brink, F., Gardner, J., Knuefing, L., Wheeler, J., Limaye, A., Turner, M. and Zhang, Y. 2022. Potassium silicate alteration in porphyry copper‑gold deposits: a case study at the giant maar-diatreme hosted Grasberg deposit, Indonesia. Journal of Volcanology and Geothermal Research 432, DOI: 10.1016/j.jvolgeores.2022.107710.
8.
Hu et al. 2018 – Hu, J., Shen, L. and Sun, G. 2018. Squeeze-and-excitation networks. IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 7132–7141.
9.
Jafari et al. 2023 – Jafari, M., Shoeibi, A., Khodatars, M., Bagherzadeh, S., Shalbaf, A., García, D.L., Gorriz, J.M. and Acharya, U.R. 2023. Emotion recognition in EEG signals using deep learning methods: A review. Computers in Biology and Medicine 165, DOI: 10.1016/j.compbiomed.2023.107450.
10.
Jia et al. 2024 – Jia, C., Long, M. and Liu, Y. 2024. Enhanced face morphing attack detection using error-level analysis and efficient selective kernel network. Computers & Security 137, DOI: 10.1016/j.cose.2023.103640.
11.
Kern et al. 2022 – Kern, M., Akushika, J.N.P., Godinho, J.R.A., Schmiedel, T. and Gutzmer, J. 2022. Integration of X-ray radiography and automated mineralogy data for the optimization of ore sorting routines. Minerals Engineering 186, DOI: 10.1016/j.mineng.2022.107739.
12.
Keshun et al. 2024a – Keshun, Y., Zengwei, L. and Yingkui, G. 2024a. A performance-interpretable intelligent fusion of sound and vibration signals for bearing fault diagnosis via dynamic CAME. Nonlinear Dynamics 112(23), pp. 20903–20940, DOI: 10.1007/s11071-024-10157-1.
13.
Keshun et al. 2024b –Keshun, Y., Puzhou, W. and Yingkui, G. 2024b. Toward Efficient and Interpretative Rolling Bearing Fault Diagnosis via Quadratic Neural Network With Bi-LSTM. IEEE Internet of Things Journal 11(13), pp. 23002–23019, DOI: 10.1109/JIOT.2024.3377731.
14.
Keshun et al. 2025 –Keshun, Y., Puzhou, W. and Yingkui, G. 2025. A sound-vibration physical-information fusion constraint-guided deep learning method for rolling bearing fault diagnosis. Reliability Engineering & System Safety 253, DOI: 10.1016/j.ress.2024.110556.
15.
Kim, T.Y. and Cho, S.-B. 2019. Predicting residential energy consumption using CNN-LSTM neural networks. Energy 182, pp. 72–81, 2019, DOI: 10.1016/j.energy.2019.05.230.
16.
Li et al. 2019 – Li, X., Wang, W., Hu, X. and Yang, J. 2019. Selective kernel networks. Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp. 510–519, DOI: 10.48550/arXiv.1903.06586.
17.
Liu et al. 2021 – Liu, Y., Zhang, Z., Liu, X., Wang, L. and Xia, X. 2021. Performance evaluation of a deep learning based wet coal image classification. Minerals Engineering 171, DOI: 10.1016/j.mineng.2021.107126.
18.
Luo et al. 2022 – Luo, X., He, K., Zhang, Y., He, P. and Zhang, Y. 2022.A review of intelligent ore sorting technology and equipment development. International Journal of Minerals Metallurgy and Materials 29(9), pp. 1647–1655, DOI: 10.1007/s12613-022-2477-5.
19.
van der Maaten L. and Hinton, G. 2008. Visualizing Data using t-SNE. Journal of Machine Learning Research 9, pp. 2579–2605.
20.
Nie et al. 2023 – Nie, C., Jiang, J., Deng, J., Li, K., Jia, L. and Sun, T. 2023. Predicting TFe content and sorting iron ores from hyperspectral image by variational mode decomposition-based spectral feature. Journal of Cleaner Production 429, DOI: 10.1016/j.jclepro.2023.139629.
21.
Rifaat et al. 2022 – Rifaat, N, Ghosh, U.K. and Sayeed, A. 2022. Accurate gait recognition with inertial sensors using a new FCN-BiLSTM architecture. Computers and Electrical Engineering 104, DOI: 10.1016/j.compeleceng.2022.108428.
22.
Roy, A. and Todorovic, S. 2016. A multi-scale cnn for affordance segmentation in rgb images. European Conference on Computer Vision – ECCV 2016, pp. 186–201, DOI: 10.1007/978-3-319-46493-0_12.
23.
Sahu et al. 2022 – Sahu, S.N. Meikap, B.C. and Biswal, S.K., ‘Magnetization roasting of waste iron ore beneficiation plant tailings using sawdust biomass; A novel approach to produce metallurgical grade pellets. Journal of Cleaner Production 343, DOI: 10.1016/j.jclepro.2022.130894.
24.
Shatwellet al. 2023 – Shatwell, D.G., Murray, V. and Barton, A. 2023. Real-time ore sorting using color and texture analysis. International Journal of Mining Science and Technology 33(6), pp. 659–674, DOI: 10.1016/j.ijmst.2023.03.004.
25.
Vinod et al. 2024 – Vinod, P., Behera, M., Jaya Prakash, A., Hebbar, R. and Srivastav, S. 2024. A novel multitask transformer deep learning architecture for joint classification and segmentation of horticulture plantations using very High-Resolution satellite imagery. Computers and Electronics in Agriculture 227, DOI: 10.1016/j.compag.2024.109540.
26.
Wang et al. 2019 – Wang, H., Xu, J., Yan, R. and Gao, R.X. 2019. A new intelligent bearing fault diagnosis method using SDP representation and SE-CNN. IEEE Transactions on Instrumentation and Measurement 69(5), pp. 2377–2389, DOI: 10.1109/TIM.2019.2956332.
27.
Wang et al. 2022 – Wang, Q., Xiao, J., Li, Y., Lu, Y., Jinjia Guo, J., Tian, Y. and Ren, L. 2022. Mid-level data fusion of Raman spectroscopy and laser-induced breakdown spectroscopy: Improving ores identification accuracy. Analytica Chimica Acta 1240, DOI: 10.1016/j.aca.2022.340772.
28.
Wang et al. 2023 – Wang, C., Walsh, S.D.C., Weng, Z., Haynes, M.W., Summerfield, D. and Feitz, A. 2023. Green steel: Synergies between the Australian iron ore industry and the production of green hydrogen. International Journal of Hydrogen Energy 48(10), pp. 32277–32293, DOI: 10.1016/j.ijhydene.2023.05.041.
29.
Wang et al. 2023 – Wang, X., Liesaputra, V., Liu, Z., Wang, Y. and Huang, Z. 2023. An in-depth survey on Deep Learning-based Motor Imagery Electroencephalogram (EEG) classification. Artificial Intelligence in Medicine 147, DOI: 10.1016/j.artmed.2023.102738.
30.
Xie et al. 2023 – Xie, H., Mao, Z., Xiao, D., and Li, Z. 2023. Rapid detection of molybdenum ore grade based on visible-infrared spectroscopy and MTSVD-TGJO-ELM. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 298, DOI: 10.1016/j.saa.2023.122789.
31.
Xu et al. 2023 – Xu, C., Liang, R., Wu, X., Cao, C., Chen, J., Yang, C., Zhou, Y., Wen, T., Lv, H. and Wei, C. 2023. A Hybrid Model Integrating CNN–BiLSTM and CBAM for Anchor Damage Events Recognition of Submarine Cables. IEEE Transactions on Instrumentation and Measurement 72, pp. 1–11, DOI: 10.1109/TIM.2023.3290323.
32.
Yi Y. and Bian, Y. 2021. Named Entity Recognition with Gating Mechanism and Parallel BiLSTM. Journal of Web Engineering 20(4), pp. 1219–1238, DOI: 10.13052/jwe1540-9589.20413.
33.
Yu et al. 2019 – Yu, Y., Si, X., Hu, C. and Zhang, J. 2019. A review of recurrent neural networks: LSTM cells and network architectures. Neural Computation 31(7), pp. 1235–1270, DOI: 10.1162/neco_a_01199.
34.
Zhang et al. 2017 – Zhang, W., Peng, G., Li, C., Chen, Y. and Zhang, Z. 2017. A New Deep Learning Model for Fault Diagnosis with Good Anti-Noise and Domain Adaptation Ability on Raw Vibration Signals. Sensors 17(3), DOI: 10.3390/s17020425.
35.
Zhang et al. 2024 – Zhang, Y., Francois, N., Henley, R., Turner, M., Saadatfar, M., Brink, F.J. and Knackstedt, M. 2024. In-situ study of texture-breakage coupling in a copper ore using X-ray micro-CT. Minerals Engineering 205, DOI: 10.1016/j.mineng.2023.108464.
| eISSN: | 2299-2324 |
| ISSN: | 0860-0953 |
We process personal data collected when visiting the website. The function of obtaining information about users and their behavior is carried out by voluntarily entered information in forms and saving cookies in end devices. Data, including cookies, are used to provide services, improve the user experience and to analyze the traffic in accordance with the Privacy policy. Data are also collected and processed by Google Analytics tool (more).
You can change cookies settings in your browser. Restricted use of cookies in the browser configuration may affect some functionalities of the website.
You can change cookies settings in your browser. Restricted use of cookies in the browser configuration may affect some functionalities of the website.
