Donbass State Technical University, Alchevsk, Russia

N. A. Denisova, Cand. Eng., Associate Prof., Head of the Dept. of Metallurgical Machines, e-mail: Natdeny@yandex.ru

L. E. Podlipenskaya, Cand. Eng., Associate Prof., Leading Researcher, Advanced Research Directorate, e-mail: lida.podlipensky@gmail.com

Donbass State Technical University, Alchevsk, Russia¹ ; The Southern Mining and Metallurgical Complex, Alchevsk, Russia²
T. R. Kozlov, Assistant¹, Senior Foreman of the Oxygen Converter Shop², e-mail: romovaldovich@mail.ru

Donetsk National Technical University, Donetsk, Russia

A. L. Sotnikov, Dr. Eng., Prof., Dept. of Metallurgical Machines, e-mail: 0713019870@mail.ru

The paper presents the development and verification of a two-channel integrated expert system for monitoring abnormal situations, “Breakthrough,” on a slab continuous casting machine (CCM). The system’s main channel is implemented using Mamdani fuzzy inference principles and is a hierarchical model of six interconnected fuzzy inference subsystems (FIS 1-FIS 6). A parallel channel, the statistical predictor detector (PD), identifies anomalies in the dynamics of process parameters using sliding statistical thresholds and provides early warning long before the fuzzy model registers a critical state. Aggregation of the outputs of both channels produces a final hazard assessment on a seven-level scale. Verification was performed using real data from a slab CCM: a series of four heats with a “breakthrough” incident (877 time points, ~146 min), a comparative analysis of a neighboring strand without an incident, and another breakthrough event. The system detected a critical process mode violation 85 minutes before the emergency occurred, with zero false alarms across all three samples.

1. Kuklev A. V., Sosnin V. V., Vinogradov V. V., Pozdnyakov V. A. Physical model of the formation of surface cracks in slabs. Stal. 2004. No. 11. pp. 95–98.
2. Smirnov A. N., Kubersky S. V., Maksaev E. N. Some aspects of the occurrence of sticking and breakthroughs of the shell of a continuously cast slab in a mold. Elektrometallurgiya. 2013. No. 9. pp. 14–22.
3. Parshin V. M., Bulanov L. V. Continuous casting of steel. Lipetsk : NLMK, 2011. 221 p.
4. Kubersky S. V. Improving the technology of steel pouring during continuous casting to increase its quality, yield and process stability. Metallurg. 2026. No. 1. pp. 15–22.
5. Shtovba S. D. Design of fuzzy systems using MATLAB. Moscow : Goryachaya liniya - Telekom, 2007. 288 p.
6. Piegat A. Fuzzy modeling and control. Translated from English by A. G. Podvesovsky, edited by Yu. S. Kharin. Moscow : BINOM. Laboratoriya znaniy, 2009. 798 p.
7. Vishnevsky D. A., Denisova N. A., Kozlov T. R. et al. Development of an expert system for diagnosing emergency situations in a continuous casting machine based on fuzzy logic. Vestnik Cherepovetskogo gosudarstvennogo universiteta. 2025. No. 1 (124). pp. 7–23. DOI: 10.23859/1994-0637-2025-1-124-1
8. Denisova N. A., Sotnikov A. L., Podlipenskaya L. E., Kozlov T. R. Development of an expert system for diagnosing an abnormal situation “Metal overflow through the mold” for a continuous caster based on fuzzy logic. Chernye Metally. 2026. No. 1. pp. 11–18.
9. Vishnevsky D. A., Denisova N. A., Kozlov T. R. et al. Automatic control system for non-closing of the CCM tundish stopper based on fuzzy logic. Chernye Metally. 2025. No. 3. pp. 61–69.
10. Zadeh L. A. Fuzzy sets. Information and Control. 1965. Vol. 8, No. 3. pp. 338–353.
11. Mamdani E. H., Assilian S. An experiment in linguistic synthesis with a fuzzy logic controller. International Journal of Man-Machine Studies. 1975. Vol. 7, No. 1. pp. 1–13.
12. Bhattacharya A. K., Srinivas P. S., Chithra K., Jatla S. V., Das J. Recognition of fault signature patterns using fuzzy logic for prevention of breakdowns in steel continuous casting process. Lecture Notes in Computer Science. 2005. Vol. 3776. pp. 318–324. DOI: 10.1007/11590316_472005
13. Solovieva O. I., Kozhevnikov A. V. Mathematical model for predicting the safety level of steel-pouring equipment. Vestnik ChGU. 2012. Vol. 2, No. 3 (41). pp. 25–31.
14. Zhang B., Wu H., Yu H. et al. Steel breakout prediction system based on deep learning and clustering. JOM. 2025. Vol. 77. pp. 1682–1691. DOI: 10.1007/s11837-024-07093-1
15. Jin X., Ren T., Shi X., Jin R., Liu D. Breakout prediction system based on combined neural network in continuous casting. Advances in Intelligent and Soft Computing. 2012. Vol. 168. pp. 349–355. DOI: 10.1007/978-3-642-30126-1_56
16. Xudong Wang et al. Analysis and prediction of sticker breakout based on XGBoost forward iterative model. ISIJ International. 2024. Vol. 64, No. 8. pp. 1272–1278. DOI: 10.2355/isijinternational.ISIJINT-2023-449
17. Smirnov A. N., Kubersky S. V., Smirnov E. N. et al. The influence of metal level vibrations in the mold on the formation of a hard shell during slab casting. Stal. 2017. No. 7. pp. 10–14.
18. Isermann R. Fault-diagnosis applications: Model-based condition monitoring. Berlin : Springer, 2011. 354 p.