Jianming Cai, Mengxia Han, Xirui Yan, Yan Chen, Daoxiu Li, Kai Zhao, Dongqing Zhang, Kaiqi Hu, Heng Han Sua, Hieng Kiat Jun, Kewei Xie, Guiliang Liu, Xiangfa Liu, Sida Liu
{"title":"高强度铝硅合金设计的过程协同主动学习框架","authors":"Jianming Cai, Mengxia Han, Xirui Yan, Yan Chen, Daoxiu Li, Kai Zhao, Dongqing Zhang, Kaiqi Hu, Heng Han Sua, Hieng Kiat Jun, Kewei Xie, Guiliang Liu, Xiangfa Liu, Sida Liu","doi":"10.1038/s41524-025-01721-3","DOIUrl":null,"url":null,"abstract":"<p>High-strength Al-Si alloys are important lightweight materials, but their optimal design is hindered by scarce-imbalance data, and complex compositional-process-property relationships. Traditional trial-and-error experimentation fails to explore this multi-dimensional design space, where processing routes (PRs) and composition must be co-optimized to achieve superior strength. This study introduces a process-synergistic active learning (PSAL) framework leveraging a conditional Wasserstein autoencoder (c-WAE) to enable the data-efficient design. By encoding PRs as conditional variables, the PSAL framework reveals exceptional synergistic effects across diverse PRs, significantly outperforming single-process approaches. The process-aware latent representation facilitates the efficient exploration of potential compositions across multi-PRs simultaneously. Through iterative active learning cycles integrating machine learning predictions with experimental validations, ultimate tensile strength is greatly improved: 459.8 MPa for gravity casting with T6 heat treatment within three iterations and 220.5 MPa for gravity casting with hot extrusion in a single iteration. This framework handles sparse datasets effectively, capturing complex process-composition-property relationships and establishing a new paradigm for accelerated multi-objective material design.</p>","PeriodicalId":19342,"journal":{"name":"npj Computational Materials","volume":"280 1","pages":""},"PeriodicalIF":11.9000,"publicationDate":"2025-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"A process-synergistic active learning framework for high-strength Al-Si alloys design\",\"authors\":\"Jianming Cai, Mengxia Han, Xirui Yan, Yan Chen, Daoxiu Li, Kai Zhao, Dongqing Zhang, Kaiqi Hu, Heng Han Sua, Hieng Kiat Jun, Kewei Xie, Guiliang Liu, Xiangfa Liu, Sida Liu\",\"doi\":\"10.1038/s41524-025-01721-3\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p>High-strength Al-Si alloys are important lightweight materials, but their optimal design is hindered by scarce-imbalance data, and complex compositional-process-property relationships. Traditional trial-and-error experimentation fails to explore this multi-dimensional design space, where processing routes (PRs) and composition must be co-optimized to achieve superior strength. This study introduces a process-synergistic active learning (PSAL) framework leveraging a conditional Wasserstein autoencoder (c-WAE) to enable the data-efficient design. By encoding PRs as conditional variables, the PSAL framework reveals exceptional synergistic effects across diverse PRs, significantly outperforming single-process approaches. The process-aware latent representation facilitates the efficient exploration of potential compositions across multi-PRs simultaneously. Through iterative active learning cycles integrating machine learning predictions with experimental validations, ultimate tensile strength is greatly improved: 459.8 MPa for gravity casting with T6 heat treatment within three iterations and 220.5 MPa for gravity casting with hot extrusion in a single iteration. This framework handles sparse datasets effectively, capturing complex process-composition-property relationships and establishing a new paradigm for accelerated multi-objective material design.</p>\",\"PeriodicalId\":19342,\"journal\":{\"name\":\"npj Computational Materials\",\"volume\":\"280 1\",\"pages\":\"\"},\"PeriodicalIF\":11.9000,\"publicationDate\":\"2025-07-14\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"npj Computational Materials\",\"FirstCategoryId\":\"88\",\"ListUrlMain\":\"https://doi.org/10.1038/s41524-025-01721-3\",\"RegionNum\":1,\"RegionCategory\":\"材料科学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"CHEMISTRY, PHYSICAL\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"npj Computational Materials","FirstCategoryId":"88","ListUrlMain":"https://doi.org/10.1038/s41524-025-01721-3","RegionNum":1,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"CHEMISTRY, PHYSICAL","Score":null,"Total":0}
A process-synergistic active learning framework for high-strength Al-Si alloys design
High-strength Al-Si alloys are important lightweight materials, but their optimal design is hindered by scarce-imbalance data, and complex compositional-process-property relationships. Traditional trial-and-error experimentation fails to explore this multi-dimensional design space, where processing routes (PRs) and composition must be co-optimized to achieve superior strength. This study introduces a process-synergistic active learning (PSAL) framework leveraging a conditional Wasserstein autoencoder (c-WAE) to enable the data-efficient design. By encoding PRs as conditional variables, the PSAL framework reveals exceptional synergistic effects across diverse PRs, significantly outperforming single-process approaches. The process-aware latent representation facilitates the efficient exploration of potential compositions across multi-PRs simultaneously. Through iterative active learning cycles integrating machine learning predictions with experimental validations, ultimate tensile strength is greatly improved: 459.8 MPa for gravity casting with T6 heat treatment within three iterations and 220.5 MPa for gravity casting with hot extrusion in a single iteration. This framework handles sparse datasets effectively, capturing complex process-composition-property relationships and establishing a new paradigm for accelerated multi-objective material design.
期刊介绍:
npj Computational Materials is a high-quality open access journal from Nature Research that publishes research papers applying computational approaches for the design of new materials and enhancing our understanding of existing ones. The journal also welcomes papers on new computational techniques and the refinement of current approaches that support these aims, as well as experimental papers that complement computational findings.
Some key features of npj Computational Materials include a 2-year impact factor of 12.241 (2021), article downloads of 1,138,590 (2021), and a fast turnaround time of 11 days from submission to the first editorial decision. The journal is indexed in various databases and services, including Chemical Abstracts Service (ACS), Astrophysics Data System (ADS), Current Contents/Physical, Chemical and Earth Sciences, Journal Citation Reports/Science Edition, SCOPUS, EI Compendex, INSPEC, Google Scholar, SCImago, DOAJ, CNKI, and Science Citation Index Expanded (SCIE), among others.