Artificial Intelligence Predicts Isothermal Transformation of Titanium Alloys with High Accuracy

Strategic Insights -

Researchers at Taiyuan University of Science & Technology have investigated the isothermal transformation process of TC17 alloy under different hydrogen contents using machine learning models and microstructure analysis. The study found that the transformation microstructure changes as the hydrogen content increases. A team of experts employed artificial neural networks, Gaussian process regression, and support vector machine models to predict the characteristics of the isothermal transformation curves of titanium alloys, demonstrating high predictive capabilities.

Key Takeaways:

  • The study investigated the isothermal transformation process of TC17 alloy under different hydrogen contents, revealing changes in transformation microstructure as hydrogen content increases.
  • Artificial neural networks, Gaussian process regression, and support vector machine models were used to predict the characteristics of the isothermal transformation curves of titanium alloys, achieving high predictive capabilities.
  • The results showed that all three models demonstrated high fitting accuracy, with the coefficient of determination (R2) values exceeding 0.94.
  • Specifically, the Gaussian process regression model achieved an R2 of 0.98 and a root mean square error (RMSE) of 16.89 in predicting nose temperature, while the support vector machine model achieved an R2 of 0.98 and an RMSE of 0.12 in predicting nose time.
  • The research concluded that analyzing the prediction results of nose temperature and its corresponding aging time for TC17 alloy with different hydrogen contents revealed consistent results with model predictions.

Statistics:

  • The coefficient of determination (R2) values for the three models exceeded 0.94, indicating high predictive capabilities.
  • The Gaussian process regression model achieved an R2 of 0.98 and a root mean square error (RMSE) of 16.89 in predicting nose temperature.
  • The support vector machine model achieved an R2 of 0.98 and an RMSE of 0.12 in predicting nose time.
  • The research included data on the isothermal transformation microstructure changes under different hydrogen contents, from 0.1 wt% to 0.5 wt%.

Sources:

  • Effect of Hydrogen On Time-temperature-transformation Behavior In Tc17 Titanium Alloy. International Journal of Hydrogen Energy, 2025;145:772-785.
  • International Journal of Hydrogen Energy, http://www.journals.elsevier.com/international-journal-of-hydrogen-energy/
  • Taiyuan University of Science & Technology, http://www.tust.edu.cn/