Simultaneous enhancement in mechanical and corrosion properties of Al-Mg-Si alloys using machine learning

Al-Mg-Si alloys with high strength and good corrosion resistance are regarded as desirable materials for all-aluminum vehicles. However, the traditional trial-and-error experimental methods are insufficient to address the trade-off between strength and corrosion resistance. In this work, a non-dominated sorting genetic machine-learning algorithm (NSGA-II) was employed to optimize the chemical composition, so as to simultaneously improve the strength and corrosion resistance. Three high-performance Al-Mg-Si alloys with low Mg, Si, and Cu contents were successfully developed, where the yield strength (YS), ultimate tensile strength (UTS), and the elongation (& delta;) reached 375-380 MPa, 410-416 MPa, and 13.7%-15.2%, re-spectively. Compared with higher-Cu-content 6013 alloy, the YS and UTS of the present alloys increase by about 60 MPa, and the intergranular corrosion resistance is also significantly improved. Microstruc-ture characterization demonstrated that /3" and QP phases introduced a significant synergistic precipita-tion strengthening effect; the dispersoids formed by trace Mn, Cr, Fe, Zr, and Ti contributed dispersion strengthening effect; and the good pitting corrosion resistance is attributed to lower Mg and Si contents. & COPY; 2023 Published by Elsevier Ltd on behalf of The editorial office of Journal of Materials Science & Technology.