Scholarlink

Twinning-induced plasticity steel is a high-strength, high-ductility material increasingly used in automotive applications to reduce weight and improve fuel efficiency without compromising safety. Its unique mechanical properties stem from mechanical twinning, which is related to stacking fault energy (SFE) and alloying elements like aluminium, manganese, carbon, and silicon. In this study, machine learning models were used to predict yield strength of TWIP steel based on composition and grain size with random forest yielding the highest accuracy (R2 = 0.80) and lowest mean square error (MSE) among all other models. This work highlights the potential of machine learning to optimize TWIP steel for high-performance applications.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Machine Learning-Based Regression Modelling for Yield Strength Prediction in TWIP Steel

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Samjukta Sinha,
  • Prabhat Das,
  • Amitava Basu Mallick,
  • Manojit Ghosh

摘要

Twinning-induced plasticity steel is a high-strength, high-ductility material increasingly used in automotive applications to reduce weight and improve fuel efficiency without compromising safety. Its unique mechanical properties stem from mechanical twinning, which is related to stacking fault energy (SFE) and alloying elements like aluminium, manganese, carbon, and silicon. In this study, machine learning models were used to predict yield strength of TWIP steel based on composition and grain size with random forest yielding the highest accuracy (R2 = 0.80) and lowest mean square error (MSE) among all other models. This work highlights the potential of machine learning to optimize TWIP steel for high-performance applications.