Kola Mining and Metallurgical Company JSC, Monchegorsk, Russia

M. I. Ryabushkin, First Deputy General Director, Chief Engineer
D. O. Sannikov, Director of the Department for Innovation and Digital Technologies

LLC Rocket Control, Moscow, Russia
S. A. Kovtun, Senior Data Engineer Developer of the Key Project Team, e-mail: s.kovtun@rocketcontrol.ai
F. V. Ryzhkov, Process Management Engineer, Candidate of Chemical Sciences, e-mail: f.ryzhkov@rocketcontrol.ai

The content of useful components in the flotation machine power supply has a significant impact on the efficiency of the flotation process. Depending on the geological and mineralogical properties of the incoming ore, the regime of mechanical and reagent control of flotation, stability of the useful components content and incoming raw material flows, various qualitative indicators of the flotation froth are relevant, which must be stably maintained. For precise and effective control of a process node in these conditions, solutions from the developing field of artificial intelligence are suitable, which are capable of adjusting the contour through pinpoint and frequent exposures to important control levers. Fundamentally, solutions can use predictive models that set the direction of state optimization. For a complex dynamic system, the parameters behavior prediction is important to ensure achievement of the target (optimal) state, supported by controllers, which makes it relevant to use machine learning models based on time series data. This allows to purposefully solve the task and maintain the required values of production indicators. As part of the research, a new technology of model predictive (model forecasting) control is proposed to stabilize the froth product quality in the conditions of the flotation technological  process of the JSC Kola MMC enrichment plant in Zapolyarny. The use of technology on the flotation machine of the cleaning operation of the specified site made it possible to reduce nickel content fluctuations in the froth product by 0.35% compared with the operation of the basic optimization algorithm. That made it possible to contribute to increasing the stability of the useful product content in the nutrition of the concentrate separation unit.

1. Bogdanov O. S., Maksimov I. I., Podnek A. K., Yanis N. A. Theory and technology of ore flotation. Moscow : Nedra, 1990. 363 p.
2. Vorobev N. I., Novik D. M. Mineral processing. Minsk : BGTU, 2008. 174 p.
3. Allgöwer F., Badgwell T. A., Qin J. S., Rawlings J. B. et al. Nonlinear predictive control and moving horizon estimation: an introductory overview. Advances in control: Highlights of ECC’99. Springer, 1999. pp. 391–449.
4. Piegat А. Fuzzy modeling and control. Translation from English. Iss. 2. Moscow : Binom. Laboratorija znanij, 2013. 798 p.
5. García C. E., Prett D. M., Morari M. Model predictive control: Theory and practice—A survey. Automatica. 1989. Vol. 25, Iss. 3. pp. 335–348. DOI: 10.1016/0005-1098(89)90002-2
6. Dawson P., Koorts R. Flotation control incorporating fuzzy logic and image analysis. IFAC Proceedings. 2014. Vol. 47, Iss. 3. pp. 352–357. DOI: 10.3182/20140824-6-za-1003.01864
7. Carvalho M. T., Duraö F. Control of a flotation column using fuzzy logic inference. Fuzzy sets and systems. International Journal in Information Science and Engineering. 2002. Vol. 125, Iss. 1. pp. 121–133. DOI: 10.1016/s0165-0114(01)00048-3
8. Yanhuai J., Zheng W. Application of artificial intelligence based on the fuzzy control algorithm in enterprise innovation. Helion. 2024. Vol. 10, Iss. 6. e28116.
9. Quintanilla P., Neethling S. J., Brito-Parada P. R. Modelling for froth flotation control: A review. Minerals Engineering. 2021. Vol. 162. 106718. DOI: 10.1016/j.mineng.2020.106718
10. Bergh L., Yianatos J., Pino C. Advances in developing supervisory control strategies for flotation plants. IFAC Proceedings. 2013. Vol. 46. pp. 110–115. DOI: 10.3182/20130825-4-us-2038.00003
11. Edwards R. P. Expert system control of a flotation circuit (T). University of British Columbia, 1990. 266 p.
12. Rogers B. D., Dalrymple R. A., Stansby P. K. SPH modeling of floating bodies in the surf zone. Coal. Engineering. 2009. pp. 204–215. DOI: 10.1142/9789814277426_0017
13. Foucaud Y., Badawi M., Filippov L. O., Filippova I. et al. A review of atomistic simulation methods for surface physical-chemistry phenomena applied to froth flotation. Minerals Engineering. 2019. Vol. 143. 106020. DOI: 10.1016/j.mineng.2019.106020
14. Brunton S. L., Proctor J. L., Kutz J. N. Discovering governing equations from data by sparse identification of nonlinear dynamical systems. Proceedings of the National Academy of Sciences of the United States of America. 2016. Vol. 113. pp. 3932–3937. DOI: 10.1073/pnas.1517384113
15. Fu Y., Aldrich C. Flotation froth image recognition with convolutional neural networks. Minerals Engineering. 2019. Vol. 132. pp. 183–190. DOI: 10.1016/j.mineng.2018.12.011
16. Shean B. J., Cilliers J. J. A review of froth flotation control. International Journal of Mineral Processing. 2011. Vol. 100. pp. 57–71. DOI: 10.1016/j.minpro.2011.05.002
17. Ibraeva L. K., Hisarov B. D. Modeling and identification of control objects. Almaty : AUPET, 2009. 212 p.
18. Abramov A. A. Processing, beneficiation and integrated use of solid minerals. Vol. 2. Moscow : Izdatelstvo Moskovskogo gosudarstvennogo gornogo universiteta, 2004. 510 p.
19. Cook J. Docker for data science. 2017. DOI: 10.1007/978-1-4842-3012-1