Sign in Subscribe
Alloys

Machine Learning Model Predicts Improved Corrosion Resistance in Alloys

Strategic Insights

— 1 min read

Researchers at Hebei University in China have made a breakthrough in using machine learning to predict the corrosion resistance of high-entropy alloys. By incorporating random forest and extreme gradient boosting models, the team was able to design alloys with improved corrosion resistance, which is critical for various industrial applications. The research highlights the potential of model-driven alloy design, which could reduce the need for trial-and-error methods.

Key Takeaways:

  • The researchers used a combined random forest and extreme gradient boosting model to predict the corrosion behavior of Co-Fe-Mn-Ni-Mo-Al alloys.
  • The model was able to accurately predict the phase evolution and corrosion behavior of the alloys, with a corrosion resistance rate of 10^-7 A cm^(-2).
  • The addition of Al and Mo to the alloy resulted in improved corrosion resistance, attributed to the cathodic protection caused by multiple phase evolution.
  • The L1(2) phase served as the cathode, while the FCC and BCC phases acted as the anode.
  • The study concludes that model-driven alloy design can deliver superior compositions using fewer resources than traditional trial-and-error methods.
  • Hebei University's Hanqing Xu, Mengdi Zhang, and Gong Li worked together to conduct the research, which was funded by the National Natural Science Foundation of China and the Hebei province department of education fund.

Statistics:

  • The corrosion resistance rate of the predicted alloys was 10^-7 A cm^(-2).
  • The phase evolution of the alloys transformed from FCC+Chi to FCC+L1(2)+BCC with the addition of Al and Mo.
  • The corrosion resistance was improved by 10% with the incorporation of L1(2) phase.
  • The study used a combined random forest and extreme gradient boosting model to predict the corrosion behavior of the alloys.

Sources:

  • VerticalNews publication, "By a News Reporter-Staff News Editor at Journal of Engineering -- Researchers detail new data in Machine Learning."
  • Hebei University publication, "By Machine Learning Tailor Al/mo Ratio To Adjust Fcc+l12+bcc Structured Cofemnni-based High-entropy Alloys for Improved Corrosion Resistance."
  • Journal of Alloys and Compounds, "Journal of Alloys and Compounds, 2025;1041."
  • Hebei University publication, "By Hanqing Xu, Mengdi Zhang, and Gong Li at Hebei University, Sch Qual & Tech Supervis, Baoding 071002, People's Republic of China."

Read more