Discovery of ultra-high strength aluminum alloys with high damage tolerance via interpretable chain-based machine learning
Lei Jiang, Xinbiao Zhang, Wentao Zhoutai, Zhilin Han, Minghong Mao, Wen Xue, Jianxin Xie
Abstract
This study proposes an interpretable chain-based machine learning (ICML) strategy for designing ultra-high strength aluminum alloys with high damage tolerance. Firstly, by integrating a gradient boosting regression model linking alloy composition (AC) and solution-aging processes (SAP) to tensile mechanical properties (TMP), including ultimate tensile strength σ b , yield strength σy , and elongation A , with an explicit quantitative relationship between TMP and fatigue strength ( FS ), expressed as FS = ασ b A 1/4 , a multi-scale interpretable prediction model was constructed to AC + SAP → TMP → FS . Subsequently, combining SHAP analysis, multi-objective optimization, and thermodynamic calculations, the study achieves the integrated optimization design of AC and SAP. The newly developed ultra-high strength aluminum alloy with high damage tolerance Al-10.5Zn-2.34 Mg-1.28Cu-0.11Zr-0.1Cr demonstrates outstanding mechanical properties, with a measured σ b = 794 ± 2 MPa, σ y = 765 ± 3 MPa, A = 11.9 ± 0.4 %, FS = 376 ± 14 MPa. Compared to commercial AA7050 and AA7055 alloys, the novel alloy demonstrated over 20 % enhancements in σ b , σ y , and FS , while maintaining equivalent or superior A . Furthermore, the FS = ασ b A 1/4 model reveals the influence of strength and plasticity on fatigue strength, enabling accurate fatigue strength predictions and excellent generalization capability, which can be extended to various metallic materials. This work provides a novel approach for efficiently designing high damage tolerance alloys.