Journals →  Obogashchenie Rud →  2018 →  #2 →  Back

BENEFICIATION PROCESSES
ArticleName Modernization of algorithms for flare detection of froth layer parameters during flotation of potash ores
DOI 10.17580/or.2018.02.07
ArticleAuthor Zatonskiy A. V., Malysheva A. V.
ArticleAuthorData

Perm National Research Polytechnic University, Berezniki Branch (Berezniki, Russia):

Zatonskiy A. V., Professor, Doctor of Engineering Sciences, zxenon2000@yandex.ru
Malysheva A. V., Postgraduate Student, akchim@mail.ru

Abstract

This study covers the scientific problem of automatic recognition of froth layer parameters in a potassium flotation machine. Based on a survey of flotation machine operators, vague rules have been compiled for flotation machine operation depending on froth type changes. The shortcomings of previously developed bubble recognition algorithms have been demonstrated, caused by the poor quality of video images or respective computational complexity. Additional requirements for the recognition algorithms associated with the froth remover motion and elimination of unnecessary areas of the image have been established. The study also demonstrates the recognition difficulties caused by the transition from recognition of static photographs taken with a flash to recognition of actual video streams. A substantiated selection of means is provided for implementing the processing algorithms for video images and for recognizing froth parameters. The single image recognition algorithm is modified and combined with the algorithm for determining the image center. This enables recognition of images with black pixel inclusions caused by matrix noise or unnecessary video stream compression. In addition, the speed of the algorithm has been increased as compared with similar solutions. The exclusion from the analysis of areas of improper illumination and minor noise has been substantiated. The possibility of identifying a transient process in a flotation machine using poor quality video images has been investigated. Based on the processing of experimental results obtained, optimal settings have been identified for establishing the adaptive binarization threshold. A linear relationship between the binarization threshold and the average frame illumination has been demonstrated. A transient process has been successfully identified. The special software used in this process recognized a change in the number of bubbles by ±6–7 % with the average frame illumination.

keywords Potassium ore, flotation, management, froth, identification, algorithm, vague rules
References

1. Romanenko S. A., Olennikov A. S. FrothMaster video system introduction experience practice at the «Green Mountain–2» Plant. Obogashchenie Rud. 2014. No. 2. pp. 23–27.
2. Zatonskiy А. V., Varlamova S. A. Use of reflection flare spots for automatic recognition of froth parameters in potassium ores flotation. Obogashchenie Rud. 2016. No. 2 (362). pp. 49–56. DOI: 10.17580/or.2016.02.09.
3. Zatonskiy А. V., Varlamova S. A., Malysheva A. V., Myasnikov A. A. Using of video graphic information to clarify a dynamic stochastic model of the potassium ore flotation process. Online Magazine «Naukovedenie». 2017. Vol. 9, No. 2. URL: http://naukovedenie.ru/PDF/53TVN217.pdf (free access).
4. Malysheva A. V. Visual computer control of bitumen production. Proceedings of the International Scientific and Practical Conference dedicated to the 60th anniversary of higher oil and gas education in the Republic of Tatarstan «Achievements, problems and prospects of oil and gas industry development». Almetyevsk: Almetyevsk State Oil Institute, 2016. Vol. 2. pp. 171–172.
5. Jahedsaravani A., Massinaei M., Marhaban M. H. An image segmentation algorithm for measurement of flotation froth bubble size distributions. Measurement. 2017. Vol. 111. pp. 29–37.
6. Tan J., Liang L., Peng Y., Xie G. The concentrate ash content analysis of coal flotation based on froth images. Minerals Engineering. 2016. Vol. 92. pp 9–20.
7. Fu Y., Aldrich C. Froth image analysis by use of transfer learning and convolutional neural networks. Minerals Engineering. 2018. Vol. 115. pp. 68–78.
8. Outotec FrothSenseTM. URL: http://www.outotec.com/globalassets/products/analyzers-and-automation/ote_frothsense_sensor_system_eng.pdf (access date: 30.03.18).
9. Minerals processing solutions. Automation. URL: http://www.metsoautomation.com/miningandconstruction/MaTobox7.nsf/DocsByID/0AF54610F6C67DDBC2257B030037F030/$File/Automation%20ENGLISH%20Low%20Res.pdf (access date: 30.03.18).
10. OpenCV. URL: https://opencv.org.
11. Fu Yi., Aldrich C. Flotation froth image analysis by use of a dynamic feature extraction algorithm. IFACPapersOnLine. 2016. Vol. 49, Iss. 20, pp. 84–89.
12. Aldrich C., Marais C., Shean B. J., Cilliers J. J. Online monitoring and control of froth flotation systems with machine vision: A review. International Journal of Mineral Processing. 2010. Vol. 96. pp. 1–13.

Language of full-text russian
Full content Buy
Back