Journals →  Gornyi Zhurnal →  2020 →  #3 →  Back

POWER SYSTEM MANAGEMENT. AUTOMATION
ArticleName Improving energy efficiency of iron ore concentrate dehydration through automation using neural technologies
DOI 10.17580/gzh.2020.03.12
ArticleAuthor Eremenko Yu. I., Khalapyan S. Yu., Anpilov A. O.
ArticleAuthorData

Ugarov Stary Oskol Technological Institute (Division), NUST MISIS, Stary Oskol, Russia:

Yu. I. Eremenko, Head of a Chair, Professor, Doctor of Engineering Sciences
S. Yu. Khalapyan, Associate Professor, Candidate of Engineering Sciences, khalapyan@ya.ru
A. O. Anpilov, Post-Graduate Student

Abstract

The study aims to enhance energy efficiency of iron ore concentrate dehydration based on the systems of automated control. Checking of disc vacuum-filters at the modern mining and metallurgical plants is executed manually by operating personnel based on the lab test data on moisture content of the concentrate. In order to ensure adherence to the technology standards, an operator has to maintain rotational speed of disc filters and/or density of initial pulp such that its moisture content is ample. As a consequence, efficiency of the filter lowers, which means increased operating expenses and product cost at the constant running cost of maintenance of the required vacuum gauge pressure in the system. The major difficulty in the automatic moisture content control is measurement of the moisture content and average thickness of filter cake. It is often technologically and economically inexpedient to install sensors and detectors. This research has revealed mutual correlation of technological parameters of vacuum-filters and feasibility of indirect humidification and weight measurement in filter cake. The hidden correlations were found by the neural network analysis of test data of a real-life object. As a
result, the neural network model was obtained, which determined the moisture content and weight of a concentrate based on the information on pressure in the vacuum system and on amplitude of vibrationat t wo points on the receiver surface with regard to the rotational speed of the disc filters and pulp density. The implemented computational experiment demonstrated the advantage of the developed system of automatic control with indirect neural network measurement of moisture content and weight of filter cake as compared with the reference system based on the laboratory research.

keywords Iron ore concentrate, dehydration, disc vacuum-filter, automatic control system, cake moisture content, indirect measurement, artificial neural network, neural network model, energy efficiency
References

1. Lozovskaya S. Yu., Chemerichko G. I., Stovpenko A. S., Narizhnykh V. Yu. Intensification of pulp separation processes on disc vacuum-filters. Vestnik Belgorodskogo gosudarstvennogo tekhnologicheskogo universiteta im. V. G. Shukhova. 2015. No. 2. pp. 95–98.
2. Patra A. S., Makhija D., Mukherjee A. K., Tiwari R., Sahoo C. R., Mohanty B. D. Improved dewatering of iron ore fines by the use of surfactants. Powder Technology. 2016. Vol. 287. pp. 43–50.
3. Volovikov A. Yu. The technique of experiments within the frame of iron ore filtration by vacuum disc filters. GIAB. 2013. No. 8. pp. 300–303.
4. Dasgupta A., Randhawa N. S., Chandrasekhar S., Goel R. P. Filtration studies for separation of solid particles from slurry generated in hydrometallurgical processing of ores and concentrates. Journal of Metallurgy and Materials Science. 2016. Vol. 58, Iss. 2. pp. 69–79.
5. Khalapyan S. Yu., Anpilov A. O. Application of neural network technologies to indirect measurements in the iron ore concentrate dewatering control. GIAB. 2017. No. 12. pp. 195–200.
6. Eremenko Yu. I., Khalapyan S. Yu., Anpilov A. O. Improvement of disc vacuum-filter efficiency. Modern problems in mining and metallurgy. Science and production : Proceedings of the 12th Russian scientific-practical conference with international participation. Staryi Oskol, 2015. Vol. 2. pp. 167–170.
7. Goldberg Yu. S., Gontarenko A. A. Dehydration of iron concentrates. Moscow : Nedra, 1986. 182 p.
8. Goldberg Yu. S., Gontarenko A. A., Barishpolets V. T., Gishchuk B. V. Processes and equipment for ore dehydration. Moscow : Nedra, 1977. 168 p.
9. Dmitrakova U. V., Kruglov A. V., Yushina T. I. Improving energy efficiency of non-ferrous metal ores dehydration process. Scientific basis and practice of processing of ores and anthropogenic raw materials : XXIV International Scientific-and-Technical Conference Proceedings. Ekaterinburg : Fort Dialog-Iset, 2019. pp. 179–184.
10. Razumov K. A. Concentration plants designing. 4th ed. Moscow : Kniga po trebovaniyu, 2019. 520 p.
11. Patskan A. Ya., Shchelinskiy A. A., Chernik Yu. P., Zolotarev A. I. Method of the automatically controlling a plate vacuum filter. Patent SSSR, No. 691156. Applied: 27.12.1976. Published: 15.10.1979. Bulletin No. 38.
12. Vishnyak B. A., Burshteyn S. M., Brodskaya S. A. Method of automatic control of process of dewatering suspension in drum or belt-type vacuum filter. Patent SSSR, No. 1725971. Applied: 23.04.1990. Published: 15.04.1992. Bulletin No. 14.
13. Goncharov Yu. G., Engel P. S., Ryasnoy A. M., Kaygorodtsev Yu. M., Miller A. A., Podoprigora V. P., Malyuta D. I., Khaletskiy B. E. Method of controlling the dehydration process in vacuum filters. Patent SSSR, No. 601029. Applied: 19.02.1973. Published: 05.04.1978. Bulletin No. 13.
14. Kononenko G. G., Bokatiy A. N., Dyadyura V. V., Sharkevich M. A., Deynega V. G. Method to automatically control operation of vacuum filter. Patent SSSR, No. 1713617. Applied: 03.01.1990. Published: 23.02.1992. Bulletin No. 7.
15. Khalapyan S. Yu., Rybak L. A., Glushchenko A. I., Mamaev Y. A. On neural network model development to solve parallel robots kinematics and control problems. International Journal of Pharmacy and Technology. 2016. Vol. 8, Iss. 4. pp. 25085–25095.
16. Koldaev A. I. The Neuro-Fuzzy Controller of Reactor Installation Management of Butanol Hydrogenation. International Journal of Engineering Research & Technology. 2014. Vol. 3, Iss. 3. pp. 1568–1571.
17. Ming-Chung Fang, Young-Zoung Zhuo, Zi-Yi Lee. The application of the self-tuning neural network PID controller on the ship roll reduction in random waves. Ocean Engineering. 2010. Vol. 37, Iss. 7. pp. 529–538.
18. Rui Cai, Xiaodong Zhang, Houfeng Wang. Bidirectional Re current Convolutional Neural Network for Relation Classification. Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics. Berlin, 2016. Vol. 1. pp. 756–765.

Language of full-text russian
Full content Buy
Back