FORECASTING THE EFFICIENCY OF PARTICLE FRACTIONATION BY A MULTI-VORTEX CLASSIFIER USING A PYTHON SOFTWARE ENVIRONMENT BASED ON CFD-MODELING
OVVFDZ
DOI:
https://doi.org/10.25712/ASTU.2072-8921.2026.01.031Keywords:
fractionation, classifier, efficiency, data approximation, data processing, neural networkAbstract
The effectiveness of dry classifiers depends on the accuracy of calculations made during equipment design, the choice of its structural implementation, and the conditions of setup and operation. Modern numerical analysis methods, as well as methods for processing got data using neural networks in Python, allow for the most accurate and effective study of the classification process. The aim of the work is to evaluate the efficiency of particle fractionation by a multi-vortex classifier through the study of a set of previously got data based on CFD modeling in Ansys Fluent and methods of their processing in Python. Approximation functions were selected that most adequately describe the dependence of the efficiency of the developed classifier on particle size and feed rate. Correction coefficients of the functions and root-mean-square deviation were determined. The coefficients of four approximating functions are determined by the least squares method. These are a changed sigmoidal function with oscillation, a combination of exponential functions, a rational function with oscillation, and a Gompertz function with oscillation. A feature of the studied functions was noted, in which the approximation coefficients are determined with an error at speeds of 12 and 16 m/s. For higher particle feed rates into the classifier, an increase in the MSD value is observed, so a code was written in Python to increase the amount of data required for analysis. Forecasting using the Gompertz function with oscillation shows the lowest MSD value between 0.0163–0.0376 for the studied conditions of air feed rate of 2–16 m/s and particles with a diameter of 5 to 100 microns. Approximation error estimates got using the sigmoidal function with oscillation solution show unsatisfactory results. The got data provide practical interest for predicting their relationship with the geometric dimensions of the multi-vortex classifier and technological parameters. This will allow selecting and designing a classifier according to various criteria under the design specification.
References
Рыбалко Р.И., Кралин А.К. Исследование рабочих процессов аэродинамической классификации сыпучих материалов в двухстадийном сепараторе // Современное промышленное и гражданское строительство. 2016. Т. 12. № 2. С. 59–69.
Терехова О.Н., Дуюнова Я.С. Пневмоцентробежная классификация дисперсных частиц в процессе переработки зерна в муку // Техника и технология пищевых производств. 2024. Т. 54. № 1. С. 124–134. doi: 10.21603/2074-9414-2024-1-2494.
Разработка классификатора с соосно расположенными трубами для разделения сыпучего материала на основе силикагеля / В.Э. Зинуров [и др.]. // Ползуновский вестник. 2021. № 2. С. 205–211. doi: 10.25712/ASTU.2072-8921.2021.02.029.
Турубаев Р.Р., Шваб А.В. Моделирование аэродинамики закрученного турбулентного течения и процесса классификации частиц в вихревой камере // Инженерно-физический журнал. 2022. Т. 95. № 5. С. 1248–1256.
Влияние поперечной неоднородности потока газа на кривую разделения гравитационного классификатора / Е.А. Шуина [и др.]. // Вестник Ивановского государственного энергетического университета. 2015. № 5. С. 60–63. doi: 10.17588/2072-2672.2015.5.060-063.
Моделирование и оптимизация регулярных многоступенчатых многопродуктовых гравитационных классификаторов / В.П. Жуков [и др.]. // Вестник Ивановского государственного энергетического университета. 2023. № 4. С. 77–84. doi: 10.17588/2072-2672.2023.4.077-084.
Separation efficiency of alumina particles in Al melt under high frequency magnetic field / Q.T. Guo [et al.]. // Transactions of Nonferrous Metals Society of China. 2010. Vol. 20. № 1. P. 153–157. doi: 10.1016/S1003-6326(09)60113-8.
Narrowing particle size distributions to enhance powder coating performance by improved classifying / X. Zhu [et al.]. // Powder Technology. 2024. Vol. 435. Article 119443. doi: 10.1016/j.powtec.2024.119443.
Fine particle sorting and classification in the cyclonic centrifugal field / P. Fu [et al.]. // Separation and Purification Technology. 2016. Vol. 158. P. 357–366. doi: 10.1016/j.seppur.2015.12.044.
Промышленные испытания фракционирования сыпучего материала в мультивихревом классификаторе-сепараторе / В.Э. Зинуров [и др.]. // Вестник технологического университета. 2022. Т. 25. № 4. С. 58–63. doi: 10.55421/1998-7072_2022_25_4_58.
Зинуров В.Э., Дмитриев А.В., Дмитриева О.С. Влияние конструктивного оформления мультивихревого классификатора-сепаратора на эффективность разделения порошка на основе силикагеля // Химическое и нефтегазовое машиностроение. 2022. № 9. С. 23–27.
Бизли Д. Python. Подробный справочник. СПб.: Символ-Плюс, 2010. 864 с.
Torres F.J. Estimation of parameters of the shifted Gompertz distribution using least squares, maximum likelihood and moments methods // Journal of Computational and Applied Mathematics. 2014. Vol. 255. P. 867–877. doi: 10.1016/j.cam.2013.07.004.
Rybalko, R.I. & Kralin, A.K. (2016). Researches of air classification of bulk materials in dry grinding aggregates in the two-stage separator. Sovremennoe promyshlennoe i grazhdanskoe stroitel'stvo, 12(2), 59–69. (In Russ.).
Terekhova, O.N. & Duyunova, Ya.S. (2024). Pneumocentrifugal classification of dispersed particles during grain milling. Tekhnika i tekhnologiya pishchevykh proizvodstv, 54(1), 124–134. (In Russ.). doi: 10.21603/2074-9414-2024-1-2494.
Zinurov, V.E., Madyshev, I.N., Ivakhnenko, A.R. & Petrova, I.V. (2021). Development of a classifier with coaxially arranged pipes for the separation of bulk material based on silica gel. Polzunovsky vеstnik, (2), 205–211. (In Russ.). doi: 10.25712/ASTU.2072-8921.2021.02.29.
Turubaev, R.R. & Shvab, A.V. (2022). Simulation of the aerodynamics of a swirling turbulent flow and of the process of classification of particles in a vortex chamber. Journal of Engineering Physics and Thermophysics, 95(5), 1230–1238.
Shuina, E.A., Mizonov, V.E. & Misbakhov, R.Sh. (2015). Influence of gas flow nonhomogeneity on the grade efficiency curve of gravity classifier. Vestnik Ivanovskogo gosudarstvennogo energeticheskogo universiteta, (5), 60–63. (In Russ.). doi: 10.17588/2072-2672.2015.5.060-063.
Zhukov, V.P., Belyakov, A.N., Ogurtsov, A.V., Shmelev, A.S., Barakovskikh, D.S. & Shuvalov, S.I. (2023). Modeling and optimization of regular multistage multiproduct gravity classifiers. Vestnik Ivanovskogo gosudarstvennogo energeticheskogo universiteta, (4), 77–84. (In Russ.). doi: 10.17588/2072-2672.2023.4.077-084.
Guo, Q.T., Hong, C.P., Cao, Z.Q., Zhang, Z.T., Li, T.J. & Jin, J.Z. (2010). Separation efficiency of alumina particles in Al melt under high frequency magnetic field. Transactions of Nonferrous Metals Society of China, 20(1), 153–157. doi: 10.1016/S1003-6326(09)60113-8.
Zhu, X., Liu, W., Zhang, H., Zhang, H. & Zhu, J. (2024). Narrowing particle size distributions to en-hance powder coating performance by improved classifying. Powder Technology, 435, 119443. doi: 10.1016/j.powtec.2024.119443.
Fu, P., Wang, F., Ma, L., Yang, X. & Wang, H. (2016). Fine particle sorting and classification in the cyclonic centrifugal field. Separation and Purification Technology, 158, 357–366. doi: 10.1016/j.seppur.2015.12.044.
Zinurov, V.E., Dmitriev, A.V., Dmitrieva, O.S. & Moiseeva, K.S. (2022). Industrial fractionation tests of bulk material in a multi vortex classifier separator. Vestnik tekhnologicheskogo universiteta, 25(4), 58–63. (In Russ.). doi: 10.55421/1998-7072_2022_25_4_58.
Zinurov, V.E., Dmitriev, A.V. & Dmitrieva, O.S. (2023). Effect of the design of a multi-vortex classifier–separator on the separation efficiency of powder based on silica gel. Chemical and Petroleum Engineering, 58(9–10), 757–765. doi: 10.1007/s10556-023-01158-9.
Bizli, D. (2010). Python. Podrobnyy spravochnik. Simvol-Plyus, Saint-Petersburg, 864 s. (In Russ.).
Torres, F.J. (2014). Estimation of parameters of the shifted Gompertz distribution using least squares, maximum likelihood and moments methods. Journal of Computational and Applied Mathematics, 255, 867–877. doi: 10.1016/j.cam.2013.07.004.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Oksana S. Dmitrieva, Andrey V. Dmitriev, Guzel R. Badretdinova, Alan A. Valeev
This work is licensed under a Creative Commons Attribution 4.0 International License.
.
This work is licensed under a 
