Industry 4.0
Vol. 11 (2026), Issue 5, pg(s) 215-219

DOMINANT TECHNOLOGIES IN “INDUSTRY 4.0”

Application of Machine Learning Algorithms for Predicting the Hydrodynamic Characteristics of High-Voltage Electric Discharge Processing of Al–Ti Powder Mixtures with Different Mass Compositions by Spark Discharge in Ethanol

  • 1 Institute of Pulse Processes and Technologies of NAS of Ukraine, Mykolaiv, Ukraine

Abstract

This study investigates the hydrodynamic and thermal characteristics of spark discharge during electric discharge treatment of Al–Ti powder mixtures in ethanol. Machine learning methods were applied to identify relationships between mixture composition, interelectrode gap, and discharge parameters. Among the tested algorithms, the Gradient Boosting model demonstrated the highest predictive accuracy for describing nonlinear plasma-dynamic processes.
The analysis revealed a linear increase in plasma channel pressure from approximately 270 to 450 MPa as the interelectrode gap increased from 5 to 35 mm, regardless of the Al–Ti ratio. In contrast, the pressure acting on the discharge chamber walls was strongly dependent on powder composition. Titanium-rich mixtures exhibited a decrease in wall pressure with increasing gap, whereas mixtures containing 80 wt.% Al showed the opposite trend.
The developed Gradient Boosting model demonstrated that thermal characteristics of the discharge are governed primarily by pulse and liquid parameters, while peripheral hydrodynamic effects are controlled by the composition of the powder mixture. The obtained results provide new insights into the optimization of electric discharge processing of metal powder systems.

Keywords

References

  1. Kandrotaitė Janutienė, R., Syzonenko, O., Mažeika, D., Gegeckienė, L., Venytė, I., Torpakov, A. Microstructure and Phase Composition of Ti–Al–C Materials Obtained by High Voltage Electrical Discharge/Spark Plasma Sintering. Materials. 17(1), Art. 115 (2024). DOI: 10.3390/ma17010115. URL: https://pmc.ncbi.nlm.nih.gov/articles/PMC10779920/
  2. Kandrotaitė Janutienė, R., Mažeika, D., Dlouhý, J., Syzonenko, O., Torpakov, A., Lipian, Ye., Baltušnikas, A. Investigation of the Microstructure of Sintered Ti–Al–C Composite Powder Materials under High-Voltage Electrical Discharge. Materials. 16(17), Art. 5894 (2023). DOI: 10.3390/ma16175894. URL: https://pmc.ncbi.nlm.nih.gov/articles/PMC10488492/
  3. Lobanov, L. M., Syzonenko, O. M., Prystash, M. S., Holovko, V. V., Tashev, P., Lypian, Ye. V., Pashchin, M. O., Mikhodui, O. L., Shcheretskyi, V. O., Torpakov, A. S. Pulsed-Discharge Treatment of the Al–Ti–C Modification System. Advances in Materials Science Research. 59, Chapter 9, 304 p. (2023).
  4. Syzonenko, O., Prystash, M., Zaichenko, A., Torpakov, A., Lypian, Ye., Rud, A., Kirian, I., Lakhnik, A., Shregii, E., Prokhorenko, S., Wojnarowska-Nowak, R., Kandrotaitė, R. High-Energy Synthesis of Metal-Matrix Composites Hardened by MAX Phases of the Ti–Al–C System. Machines. Technologies. Materials. 12(10), 395–397 (2018).
  5. Syzonenko, O., Prystash, M., Zaichenko, A., Lypian, Ye., Torpakov, A., Trehub, V. The Impact of Electric Field Distribution during Ti–Al–C System Blend Preparation on the Physical and Mechanical Properties of Consolidated Materials. Machines. Technologies. Materials. 12(2), 86–89 (2018).
  6. Syzonenko, O. M., Prystash, M. S., Prystash, S. F., Torpakov, A. S., Lypian, Ye. V. The Use of Machine Learning Methods to Predict the Processes and Results of High-Voltage Electric Discharge Processing of Titanium Powder in Kerosene with the Implementation of Volume-Distributed Multi-Spark Discharge. IOP Conference Series: Materials Science and Engineering. 1277, Art. 012001 (2023). DOI: 10.1088/1757-899X/1277/1/012001.
  7. Syzonenko, O., Prystash, M., Kandrotaitė Janutienė, R. The Influence of the Composition of the Working Fluid on the Characteristics of High-Voltage Electric Discharge in the "Hydrocarbon Liquid–Al Powder" Disperse System. Machines. Technologies. Materials. 17(5), 180–182 (2024). URL: https://stumejournals.com/journals/mtm/2024/6/185
  8. He, Z., Chen, B., Zhang, Y., Liu, D., Lu, X., Ostrikov, K. Data-Driven Plasma Science: A New Perspective on Modeling, Diagnostics, and Applications through Machine Learning. Plasma Processes and Polymers. Art. 202400020 (2024). DOI: 10.1002/ppap.202400020. URL: https://onlinelibrary.wiley.com/doi/10.1002/ppap.202400020
  9. Xie, Y., Jiang, Q., Gao, Y. Temperature Prediction of High-Temperature and High-Enthalpy Plasma Generators Based on Machine Learning. Physics of Plasmas. 32(1), Art. 013504 (2025). DOI: 10.1063/5.0232484. URL: https://pubs.aip.org/aip/pop/article/32/1/013504/
  10. Syzonenko, O., Prystash, M., Prystash, S., Torpakov, A., Lypian, Ye., Kandrotaitė Janutienė, R. The Use of Machine Learning Methods to Predict the Processes and Results of High-Voltage Electric Discharge Treatment of Titanium Powder in Kerosene. Machines. Technologies. Materials. 16(8), 267–269 (2022).
  11. Sizonenko, O. N., Baglyuk, G. A., Raichenko, A. I., Taftai, E. I., Lipyan, E. V., Zaichenko, A. D., Torpakov, A. S., Guseva, E. V. Variation in the Particle Size of Fe–Ti–B₄C Powders Induced by High-Voltage Electrical Discharge. Powder Metallurgy and Metal Ceramics. 51(3), 129–136 (2012).
  12. Friedman, J. H. Greedy Function Approximation: A Gradient Boosting Machine. The Annals of Statistics. 29(5), 1189–1232 (2001). DOI: 10.1214/AOS/1013203451. URL: https://www.cse.cuhk.edu.hk/irwin.king/_media/presentations/2001_greedy_function_approximation_a_gradient_boosting_machine.pdf
  13. Breiman, L. Random Forests. Machine Learning. 45(1), 5–32 (2001). DOI: 10.1023/A:1010933404324. URL: https://oa.mg/work/10.1023/a:1010933404324
  14. Maulud, D. H., Abdulazeez, A. M. A Review on Linear Regression Comprehensive in Machine Learning. Journal of Applied Science and Technology Trends. 1(4), 140–147 (2020). DOI: 10.38094/jastt1457. URL: https://jastt.org/index.php/jasttpath/article/view/57
  15. Popescu, M.-C., Balas, V. E., Perescu-Popescu, L., Mastorakis, N. Multilayer Perceptron and Neural Networks. WSEAS Transactions on Circuits and Systems. 8(7), 579–588 (2009). URL: https://www.researchgate.net/publication/228340819_Multilayer_perceptron_and_neural_networks
  16. Smola, A., Schölkopf, B. A Tutorial on Support Vector Regression. Statistics and Computing. 14(3), 199–222 (2004). DOI: 10.1023/B:STCO.0000035301.49549.88. URL: https://openresearch-repository.anu.edu.au/items/e533f210-3ec7-4be5-8694-e71057015b7d
  17. Botchkarev, A. A New Typology Design of Performance Metrics to Measure Errors in Machine Learning Regression Algorithms. Interdisciplinary Journal of Information, Knowledge, and Management. 14, 45–76 (2019). DOI: 10.28945/4184. URL: https://www.informingscience.org/Publications/4184

Article full text

Download PDF