| ArticleName |
Simulation of mechanical wear of work rolls of a wide-strip hot rolling mill using machine learning methods |
| ArticleAuthorData |
JSC Vyksa Steel Works, Vyksa, Russia ; Bauman Moscow State Technical University, Moscow, Russia:
A. E. Sevidov, Leading Software Engineer1, Postgraduate Student2, e-mail: sevidov_ae@vsw.ru
A. V. Muntin, Cand. Eng., Deputy Director for Research Activities1, Associate Prof.2, e-mail: muntin_av@omk.ru
Bauman Moscow State Technical University, Moscow, Russia:
A. G. Kolesnikov, Dr. Eng., Prof., Head of the Scientific and Educational Complex "Machine-Building Technologies" |
| Abstract |
Modern methods of modeling a mechanical wear of hot rolling mill work rolls under industrial conditions are presented. The use of machine learning algorithms based on multidimensional linear regression, decision trees, and neural networks made it possible to describe more accurately the features of work roll wear depending on many technological factors. At the same time, the data set from the production database was subjected to additional classification for each technological parameter in a certain range. This classification approach significantly improved the quality of the output model for all major metrics. The input variables for training were the main technological factors measured in the finishing group of stands during the campaign of each work roll. The output parameters – wear curves on a roll grinder measured after roll changes. 93 input parameters were used to calculate each of the 300 output values of mechanical wear along the length of the roll barrel and to train machine learning algorithms. Roll wear forecasting models based on multivariate linear regression, random forest, gradient boosting and neural networks have been developed. A comparative analysis of the developed models with classical approaches for calculating the work roll mechanical wear showed that machine learning methods have a higher accuracy. An approach and recommendations for the introduction of new models to the automated process control system of hot strip rolling mills were proposed.
The research was carried out within the framework of the program of strategic academic leadership of the Russian Federation "Priority 2030", aimed at supporting the development programs of educational institutions of higher education, the scientific project PRIOR/SN/NU/22/SP5/26 "Creation of innovative digital tools for the use of applied artificial intelligence and advanced statistical analysis of big data in technological processes of manufacture of metallurgical products", as well as within the framework of scientific and technical cooperation between JSC Vyksa Metallurgical Plant and Bauman MSTU. |
| References |
1. Reiffersheid M. Ideas, techniques and decisions for application of digital technologies in ferrous metallurgy. Chernye Metally. 2018. No. 6. pp. 62–67.
2. Belskiy S. М. Parameters of evaluation of shape cross section of hot-rolled steel strips. Message 2. Chernye Metally. 2017. No. 11. pp. 42–46.
3. Muntin А. V., Orekhov D. М., Sevidov А. Е., Tikhonov S. М., Korovin А. V., Ionov S. М. Analysis of technological factors for ensuring flatness during the rolling of ultra-thin hot-rolled strip on the wide-strip mill 1950 of JSC Vyksa Steel Works. Proizvodstvo prokata. 2019. No. 7. pp. 4–13.
4. Chekmarev А. P., Mashkovtsev R. А. Roll wear. Kharkov: Metallurgizdat, 1955. 148 p.
5. Ginzburg V. B., Ballas R. Fundamentals of flat rolling. New York: Dekker, 2000. pp. 449–455.
6. Muntin А. V., Sevidov А. Е., Tikhonov S. М. et al. Analysis of features of wear of work rolls of the finishing group of stands in the conditions of mill 1950 at the JSC VSW`s sheet rolling complex. Metallurg. 2021. No. 3. pp. 57–62.
7. Sun J., Deng J., Peng W. et al. Strip Crown Prediction in Hot Rolling Process Using Random Forest. International Journal of Precision Engineering and Manufacturing. 2021. Vol. 22. pp. 301–311.
8. Muntin А. V., Kurenkov Yu. М., Kolesnikov А. G. Modern technological solutions and equipment for production of ultra-thin hot-rolled strip. Proizvodstvo prokata. 2016. No. 8. pp. 13–21.
9. Muntin A. V. Advanced technology of combined thin slab continuous casting and steel strip hot rolling. Metallurgist. 2019. Vol. 62. No. 9-10. pp. 900–910.
10. Saravanan R., Sujatha P. A State of Art Techniques on Machine Learning Algorithms: A Perspective of Supervised Learning Approaches in Data Classification. Second International Conference on Intelligent Computing and Control Systems (ICICCS). 2018. pp. 945–949.
11. Randles B. M., Pasquetto I. V., Golshan M. S., Borgman C. L. Using the Jupyter Notebook as a Tool for Open Science: An Empirical Study. ACM/IEEE Joint Conference on Digital Libraries (JCDL). 2017. pp. 1, 2.
12. Sarker I. H. Machine learning: algorithms, real-world applications and research directions. SN Computer Science. 2021. Vol. 2. pp. 1–21.
13. McDonald G. C. Ridge regression. WIREs Computational Statatistics. 2008. Vol. l. pp. 93–100.
14. Liu Y., Wang Y., Zhang J. New Machine Learning Algorithm: Random Forest. Information Computing and Applications. ICICA 2012. Vol. 7473. Springer, Berlin, Heidelberg. p. 246.
15. Friedman J. H. Greedy Function Approximation: A Gradient Boosting Machine. The Annals of Statistics. 2001. Vol. 29, Iss. 5. pp. 1189–232.
16. Wu Yc., Feng Jw. Development and Application of Artificial Neural Network. Wireless Personal Communications. 2018. Vol. 102. pp. 1645–1656.
17. Pan F., Converse T., Ahn D., Salvetti F., Donato G. Feature selection for ranking using boosted trees. Proceedings of the 18th ACM conference on Information and knowledge management. 2009. pp. 2025–2028. |
