Machine learning-enhanced constitutive modeling and hot deformation behaviour of Ti-stabilized AISI 321 austenitic stainless steel
Loading...
Date
Journal Title
Journal ISSN
Volume Title
Publisher
Elsevier
Abstract
Please read abstract in the article.
HIGHLIGHTS
• ML–Arrhenius hybrid model developed for AISI 321 hot deformation.
• Random Forest predicts full flow curves with high accuracy.
• ML smoothing improves stability of Arrhenius parameters.
• Power dissipation efficiency and flow instability map defines optimal DRX hot-working conditions.
• Framework enhances constitutive modelling of Ti-stabilized steels.
Description
DATA AVAILABILITY : The data that support the findings of this study, including experimental stress–strain datasets and machine-learning prediction files, are available from the corresponding author upon reasonable request.
Keywords
Machine learning in metallurgy, Dynamic recrystallization, AISI 321 stainless steel, Power dissipation efficiency, Flow instability maps, Random forest regression
Sustainable Development Goals
SDG-12: Responsible consumption and production
Citation
Nkhoma, R., Mwale, V., Ngonda, T. & Siyasiya, C.W. 2026, 'Machine learning-enhanced constitutive modeling and hot deformation behaviour of Ti-stabilized AISI 321 austenitic stainless steel', Next Materials, vol. 11, art. 101623, pp. 1-11, doi : 10.1016/j.nxmate.2026.101623.