EUVIMEDEUVIMED entdecken
Crossref

Al-Driven Prediction of Structural Steel Mechanical Properties Using Chemical Compositions and Hardness

Haris Sufyan Bin Mohamad, Shagea Alqawzai, Zaki Bin Muhammad Sham, Thi Qui Nguyen, Mingshan Zhao, Zhang Linyun

e-Journal of Nondestructive Testing · 2026-07 · 31 (7)

DOI: 10.58286/33561 ↗

Kurz gefasst

Fachartikel in der Zeitschrift e-Journal of Nondestructive Testing, 2026-07. Verfasst von Haris Sufyan Bin Mohamad und 5 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Structural steel reuse presents a promising opportunity to reduce embodied carbon and material waste; however, its broader adoption is hindered by uncertainty in the mechanical properties of reclaimed steel members. Destructive testing provides accurate strength data but is costly, time-consuming, and impractical for large-scale assessment. Non-destructive techniques such as portable hardness testing and optical emission spectroscopy (OES) offer rapid alternatives for property evaluation. This study examines various hardness-based and chemical composition-based machine learning models for predicting mechanical properties of structural steel. Vickers hardness and OES measurements were obtained from specimens representing commonly used steel grades. For hardness-based tensile strength prediction, strong agreement was achieved with R² values of 0.92 (training) and 0.85 (testing), mean absolute percentage errors of approximately 1.6–2.1%, and 100% accuracy within ±10%. Chemical-composition-based yield strength prediction also showed good performance, with R² values of 0.92 (training) and 0.88 (testing), mean absolute percentage errors of about 3.4–4.2%, and accuracy within ±10% of approximately 92–95%. These results demonstrate that both approaches provide reliable non-destructive strength estimation, supporting informed and sustainable reuse of structural steel.

Zitieren

APA

Mohamad, H. S. B., Alqawzai, S., Sham, Z. B. M., Nguyen, T. Q., Zhao, M., & Linyun, Z. (2026). Al-Driven Prediction of Structural Steel Mechanical Properties Using Chemical Compositions and Hardness. e-Journal of Nondestructive Testing, 31(7). https://doi.org/10.58286/33561

Vancouver

Mohamad HSB, Alqawzai S, Sham ZBM, Nguyen TQ, Zhao M, Linyun Z. Al-Driven Prediction of Structural Steel Mechanical Properties Using Chemical Compositions and Hardness. e-Journal of Nondestructive Testing. 2026;31(7). doi:10.58286/33561

BibTeX
@article{Mohamad2026,
  author = {Mohamad, Haris Sufyan Bin and Alqawzai, Shagea and Sham, Zaki Bin Muhammad and Nguyen, Thi Qui and Zhao, Mingshan and Linyun, Zhang},
  title = {Al-Driven Prediction of Structural Steel Mechanical Properties Using Chemical Compositions and Hardness},
  journal = {e-Journal of Nondestructive Testing},
  year = {2026},
  volume = {31},
  number = {7},
  doi = {10.58286/33561}
}
Kostenlos starten →