Application of artificial neural networks to investigate the influence of crosshead speed and sample specifications on tensile mechanical properties of DX56D+Z100-M-C-O steel sheets
Advances in Mechanical and Materials Engineering · 2026 · 43 · 191-199
DOI: 10.7862/rm.2026.13 ↗Kurz gefasst
Fachartikel in der Zeitschrift Advances in Mechanical and Materials Engineering, 2026. Verfasst von Vít Novák und 4 weiteren Autor:innen.
Abstract (Originalsprache)
Formability of steel sheet metals depends primarily on mechanical properties of material that influence the material's ability to undergo plastic deformation without risk of fracturing. This article presents the use of multilayer artificial neural networks (ANNs) to evaluate the effect of uniaxial tensile test parameters on selected mechanical properties of 0.7-mm-thick DX56D+Z100-M-C-O steel sheets. Samples were cut from different areas of the coil and at different angles (0°, 45° and 90°) relative to the sheet rolling direction. Samples also came from different batches of materials. Yield strength, ultimate tensile strength and uniform elongation were determined under variable crosshead speeds between 2 and 600 mm/min. The quality parameters of the ANNs were the value of the coefficient of determination (the higher the better) and the prediction error (the smaller the better) determined for the test set. The predictive quality of ANNs was characterized by the coefficient of determination R2 = 0.7849-0.9864, depending on the ANN architecture, the output variable and type of neuron activation function used. As crosshead speed increased, elongation of samples decreased. However, as crosshead speed decreased, yield strength decreased. The influence of sample orientation on the analyzed mechanical parameters of the sheet was smaller than the effect of crosshead speed. This was also confirmed by sensitivity analysis of the input parameters.
Zitieren
APA
Novák, V., Tatíček, F., Żaba, K., Stejskal, O., & Mizera, J. (2026). Application of artificial neural networks to investigate the influence of crosshead speed and sample specifications on tensile mechanical properties of DX56D+Z100-M-C-O steel sheets. Advances in Mechanical and Materials Engineering, 43, 191-199. https://doi.org/10.7862/rm.2026.13
Vancouver
Novák V, Tatíček F, Żaba K, Stejskal O, Mizera J. Application of artificial neural networks to investigate the influence of crosshead speed and sample specifications on tensile mechanical properties of DX56D+Z100-M-C-O steel sheets. Advances in Mechanical and Materials Engineering. 2026;43:191-199. doi:10.7862/rm.2026.13
BibTeX
@article{Novk2026,
author = {Novák, Vít and Tatíček, František and Żaba, Krzysztof and Stejskal, Ondřej and Mizera, Jarosław},
title = {Application of artificial neural networks to investigate the influence of crosshead speed and sample specifications on tensile mechanical properties of DX56D+Z100-M-C-O steel sheets},
journal = {Advances in Mechanical and Materials Engineering},
year = {2026},
volume = {43},
pages = {191--199},
doi = {10.7862/rm.2026.13}
}
EUVIMED