Journals →  Tsvetnye Metally →  2022 →  #2 →  Back

ArticleName Optimization of bulk flotation process at Talnakh Concentrator based on machine learning algorithms
DOI 10.17580/tsm.2022.02.11
ArticleAuthor Abrarov A. D., Datsiev M. S., Chikildin D. E., Fedotov D. N.

MMC Norilsk Nickel, Moscow, Russia:

A. D. Abrarov, Project Leader at the Innovative Development Centre, e-mail:


MMC Norilsk Nickel’s Polar Division, Norilsk, Russia:
M. S. Datsiev, Head of the Science and Technology Directorate, e-mail:
D. E. Chikildin, Deputy Chief Engineer – Head of the Engineering Department, e-mail:


Nornickel – Services Centre LLC, Moscow, Russia:
D. N. Fedotov, Senior Manager, e-mail:


In recent years, Russian industry has witnessed a rise of digitalization projects. This is due to a higher general quality of data, a large number of field instruments installed, availability of sufficient volumes of production history data and emergence of specialized expertise at the intersection of production technology and modern data analysis tools — for example, machine learning. Since late 2018, Talnakh Concentrator Plant at MMC Norilsk Nickel’s Polar Division has been implementing a number of artificial intelligence projects. The programme is titled Digital Plant. The projects to be implemented under the programme are divided in several categories — digital twins of the operator, computer vision sensors based on video analytics software and digital twins of the production line. Digital twins of the operator are designed to help control the production process; video analytics software based computer vision sensors deliver real-time data about the production process (e.g. the size of ore lumps transported on the conveyor); digital twins of the production line help analyze the behavior of the material in spots that cannot be observed by human eye (e.g. inside a grinding mill). The paper describes how one of the Digital Plant projects was implemented at Talnakh Concentrator Plant. Thus, a digital twin of the bulk flotation operator was created — i.e. a system that simulates the operator actions. It maintains the flotation process by exercising real-time control while the human operator can use his/her attention somewhere else. Increased stability of the flotation process contributes to the plant performance. Thus, the recovery of nickel and copper into bulk concentrate increased by 0.1 and 0.04%, correspondingly, after the algorithm had been implemented in one section (out of ten).

keywords Сoncentrator plant, flotation units, process optimization, recovery increase, data analysis, machine learning, digital twin of the operator

1. Maldonado M., Araya R., Finch J. An overview of optimizing strategies for flotation banks. Minerals. 2012. Vol. 2, Iss. 4. pp. 258–271.
2. Maldonado M., Araya R., Finch J. Optimizing flotation bank performance by recovery profiling. Minerals Engineering. 2011. Vol. 24, Iss. 8. pp. 939–943.
3. Harris A., Venkatesan L., Greyling M. A practical approach to plant-scale flotation optimization. The Journal of the Southern African Institute of Mining and Metallurgy. 2013. Vol. 113. pp. 263–272.
4. Smith C., Hadler K., Cilliers J. Flotation control using froth stability: past, present and future. 14th AusIMM Mill Operators’ Conference. Brisbane. 2018.

5. JinZhang, Zhaohui Tanga, Yongfang Xie, Mingxi Ai, Guoyong Zhang, Weihua Gui. Data-driven adaptive modeling method for industrial processes and its application in flotation reagent control. ISA Transactions. 2021. Vol. 108. pp. 305–316.
6. Oosthuizen D. J., Craig I. K. Predicting optimal operating points by modelling different flotation mechanisms. IFAC-PapersOnLine. 2019. Vol. 52, Iss. 14. pp. 60–65.
7. Kejonen I., Haavisto O., Martikainen J., Suontaka V., Musuku B. Improving grade control efficiency with rapid on-line elemental analysis. Minerals Engineering. 2018. Vol. 124. pp. 68–73.
8. Bazoev Kh. A. History of Talnakh Concentrator. Tsvetnye Metally. 2006. No. 8. pp. 29–33.
9. Golovina Yu. S., Smirnov G. N., Smirnov A. N. The Talnakh Concentrator upgrade project realization stages. Obogashchenie Rud. 2012. No. 1. pp. 43–47.
10. Tsymbal A. S., Lesnikova L. S., Volyanskiy I. V., Arabadzhi Ya. N. Phases of Talnakh concentrator development and expansion for mineral raw material processing. Tsvetnye Metally. 2015. No. 6. pp. 17–19. DOI: 10.17580/tsm.2015.06.03.
11. Volyanskiy I. V., Likhacheva T. A., Kurchukov A. M., Miller A. A. Basic aspects of Talnakh concentrator reconstruction and modernization. Tsvetnye Metally. 2020. No. 6. pp. 6–10. DOI: 10.17580/tsm.2020.06.01.
12. Couture G., Dub J.-F. Implementation of a flotation expert control system at the raglan mine. 47th Annual Canadian Mineral Processors Operators Conference. Ontario, 2015. pp. 165–172.
13. Bergh L., Yianatos J., Carrasco C. Operation of rougher flotation circuits aided by industrial simulator. 18th IFAC World Congress Milano. Italy, 2011. pp. 9917–9922.
14. Aleksandrova T. N., Arustamyan K. М., Romanenko S. A. The mathematical analysis methods application in estimation of the international practice of copperzinc and pyritic-polymetallic ores selective flotation. Obogashchenie Rud. 2017. No. 5. pp. 21–27. DOI: 10.17580/or.2017.05.04.
15. Beriashvili A. T., Pikulina V. M. A new approach to solving the problem of variable copper recovery on the example of the Zhezkazgan ore field. Obogashchenie Rud. 2018. No. 5. pp. 40–44. DOI: 10.17580/or.2018.05.07.
16. Arustamyan A. M., Sanakulov K. S. Harmonic analysis of random functions in flotation studies. Gornyi Zhurnal. 2017. No. 10. pp. 33–40. DOI: 10.17580/gzh.2017.10.07.
17. Arustamyan A. M., Romanenko S. A. Developing a mathematical model of the flotation process for pyritic copper ores. New Research in Minerals Processing and Industrial Safety. Proceedings of the conference of young researchers and specialists involving international participants. 2018. pp. 11–17.
18. Pua Y., Szmigiel А., Chen J., Apel D. B. FlotationNet: a hierarchical deep learning network for froth flotation recovery prediction. Powder Technology. 2020. Vol. 375. pp. 317–326.
19. Kurchukov A. M. A control system to control flotation of copper-nickel ores through ion content and froth optimization: Candidate of technical sciences dissertation. Saint Petersburg, 2011. 10 p.
20. Melkumova L. E., Shatskikh S. Ya. Comparing the ridge regression and LASSO techniques used for solving data processing problems. Proceedings of the 3rd International Conference and Youth School Information Technology and Nanotechnology. Samara : Novaya tekhnika, 2017. pp. 1748–1755.

Language of full-text russian
Full content Buy