[S3] S3 Materials DX for the research and development of structural materials for extreme environment(IV)

11:55 AM - 12:10 PM JST(2:55 AM - 3:10 AM UTC)

[S3.8] Clustering Analysis of Creep Rupture Data for Long-Term Creep Life Predictionof Ferritic Heat-Resistant Steels

○Nobuaki Sekido¹, Kyosuke Yoshimi¹ (1. Tohoku Univ.)

Keywords:

heat resistant steel,Machine Learning,creep

K-means clustering was applied to creep rupture data for T91 and T92 steels compiled in the NIMS Creep Data Sheet. The stress–time-to-rupture diagrams were classified into four regions, suggesting that the boundaries near the half-yield criterion are related to reductions in rupture ductility and the development of grain-boundary damage.