Knowledge-aware design of high-strength aviation aluminum alloys via machine learning

• DOI: 10.1016/j.jmrt.2023.03.041
• Publication Date: 2023-03-10T02:25:07Z
• Authors: Juan Yong-fei; Niu Guo-shuai; Yang Yang; Dai Yong-bing; Zhang Jiao; Han Yan-feng; Sun Bao-de
• Keywords: Interpretability

The development of the aviation industry is accompanied by continuous research high-performance aluminum alloys. Stuck in vast untapped composition space and routine trial-and-error method, efficiently discovering high-strength alloys remains a significant challenge. To address this issue, we proposed knowledge-aware design system (KADS) using machine learning (ML) methods to facilitate rational An alloy database containing 5113 samples was built based on Al–Zn–Mg–Cu, Al–Cu, Al–Li series Notably, guided material knowledge, constructed feature pool (23 descriptors) improve interpretability accuracy ML models. Taking key features as input, realized transformation from "element content property" "material knowledge modeling, which first time design. According predictive results, experimentally fabricated KADS-designed (KADS-Sc) with superior mechanical strength (812 MPa for ultimate tensile 792 yield strength). Furthermore, strengthening mechanisms KADS-Sc were established quantitatively. calculations confirmed that precipitation (≈ 439 MPa) most critical final increment, agreeing microstructure analysis.