Enhancing Corrosion Resistance of Aluminum Alloys Through AI and ML Modeling

In collaboration with the U.S. Naval Nuclear Laboratory, this project developed data-driven modeling techniques to better understand and improve the corrosion resistance of aluminum alloys used in demanding engineering environments. Corrosion remains a critical challenge for structural materials in marine and nuclear systems, where long-term exposure to harsh chemical and thermal conditions can degrade performance and reliability. Marine corrosion costs about $50–80 Billion per year, according to NACE.

The WISER research team collected 331 detailed alloy composition experimental data points to analyze complex relationships between alloy composition, microstructural features, and corrosion behavior. By training predictive models on experimental and simulated materials data, the team analysed 60+ alloys across 10 marine environments to identify patterns and material parameters that influence corrosion resistance.

With a high accuracy R2 score for corrosion forecasting, the team developed a custom alloy optimizer based on reverse-engineering that recommends the best aluminum compositions for specific corrosion resistance needs, saving time on R&D trial-and-error.

WISER Research Fellows: Farnaz Kaboudvand, Maham Khalid, Nydia Assaf Aragon