通过gan增强的定向约束主动学习加速发现近零烧蚀超高温陶瓷

Wenjian Guo , Fayuan Li , Lingyu Wang , Li'an Zhu , Yicong Ye , Zhen Wang , Bin Yang , Shifeng Zhang , Shuxin Bai
{"title":"通过gan增强的定向约束主动学习加速发现近零烧蚀超高温陶瓷","authors":"Wenjian Guo ,&nbsp;Fayuan Li ,&nbsp;Lingyu Wang ,&nbsp;Li'an Zhu ,&nbsp;Yicong Ye ,&nbsp;Zhen Wang ,&nbsp;Bin Yang ,&nbsp;Shifeng Zhang ,&nbsp;Shuxin Bai","doi":"10.1016/j.apmate.2025.100287","DOIUrl":null,"url":null,"abstract":"<div><div>In materials science, a significant correlation often exists between material input parameters and their corresponding performance attributes. Nevertheless, the inherent challenges associated with small data obscure these statistical correlations, impeding machine learning models from effectively capturing the underlying patterns, thereby hampering efficient optimization of material properties. This work presents a novel active learning framework that integrates generative adversarial networks (GAN) with a directionally constrained expected absolute improvement (EAI) acquisition function to accelerate the discovery of ultra-high temperature ceramics (UHTCs) using small data. The framework employs GAN for data augmentation, symbolic regression for feature weight derivation, and a self-developed EAI function that incorporates input feature importance weighting to quantify bidirectional deviations from zero ablation rate. Through only two iterations, this framework successfully identified the optimal composition of HfB<sub>2</sub>-3.52SiC-5.23TaSi<sub>2</sub>, which exhibits robust near-zero ablation rates under plasma ablation at 2500 ​°C for 200 ​s, demonstrating superior sampling efficiency compared to conventional active learning approaches. Microstructural analysis reveals that the exceptional performance stems from the formation of a highly viscous HfO<sub>2</sub>-SiO<sub>2</sub>-Ta<sub>2</sub>O<sub>5</sub>-HfSiO<sub>4</sub>-Hf<sub>3</sub>(BO<sub>3</sub>)<sub>4</sub> oxide layer, which provides effective oxygen barrier protection. This work demonstrates an efficient and universal approach for rapid materials discovery using small data.</div></div>","PeriodicalId":7283,"journal":{"name":"Advanced Powder Materials","volume":"4 3","pages":"Article 100287"},"PeriodicalIF":0.0000,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Accelerated discovery of near-zero ablation ultra-high temperature ceramics via GAN-enhanced directionally constrained active learning\",\"authors\":\"Wenjian Guo ,&nbsp;Fayuan Li ,&nbsp;Lingyu Wang ,&nbsp;Li'an Zhu ,&nbsp;Yicong Ye ,&nbsp;Zhen Wang ,&nbsp;Bin Yang ,&nbsp;Shifeng Zhang ,&nbsp;Shuxin Bai\",\"doi\":\"10.1016/j.apmate.2025.100287\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><div>In materials science, a significant correlation often exists between material input parameters and their corresponding performance attributes. Nevertheless, the inherent challenges associated with small data obscure these statistical correlations, impeding machine learning models from effectively capturing the underlying patterns, thereby hampering efficient optimization of material properties. This work presents a novel active learning framework that integrates generative adversarial networks (GAN) with a directionally constrained expected absolute improvement (EAI) acquisition function to accelerate the discovery of ultra-high temperature ceramics (UHTCs) using small data. The framework employs GAN for data augmentation, symbolic regression for feature weight derivation, and a self-developed EAI function that incorporates input feature importance weighting to quantify bidirectional deviations from zero ablation rate. Through only two iterations, this framework successfully identified the optimal composition of HfB<sub>2</sub>-3.52SiC-5.23TaSi<sub>2</sub>, which exhibits robust near-zero ablation rates under plasma ablation at 2500 ​°C for 200 ​s, demonstrating superior sampling efficiency compared to conventional active learning approaches. Microstructural analysis reveals that the exceptional performance stems from the formation of a highly viscous HfO<sub>2</sub>-SiO<sub>2</sub>-Ta<sub>2</sub>O<sub>5</sub>-HfSiO<sub>4</sub>-Hf<sub>3</sub>(BO<sub>3</sub>)<sub>4</sub> oxide layer, which provides effective oxygen barrier protection. This work demonstrates an efficient and universal approach for rapid materials discovery using small data.</div></div>\",\"PeriodicalId\":7283,\"journal\":{\"name\":\"Advanced Powder Materials\",\"volume\":\"4 3\",\"pages\":\"Article 100287\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2025-03-26\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Advanced Powder Materials\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S2772834X25000235\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Advanced Powder Materials","FirstCategoryId":"1085","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2772834X25000235","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

摘要

在材料科学中,材料输入参数与其相应的性能属性之间往往存在显著的相关性。然而,与小数据相关的固有挑战模糊了这些统计相关性,阻碍了机器学习模型有效捕获潜在模式,从而阻碍了材料性能的有效优化。这项工作提出了一种新的主动学习框架,该框架将生成对抗网络(GAN)与方向约束的预期绝对改进(EAI)获取函数集成在一起,以加速使用小数据发现超高温陶瓷(UHTCs)。该框架采用GAN进行数据增强,符号回归进行特征权重推导,以及自行开发的EAI函数,该函数结合输入特征重要性加权来量化从零消融率的双向偏差。仅通过两次迭代,该框架就成功地确定了HfB2-3.52SiC-5.23TaSi2的最佳组成,在2500°C等离子体烧蚀200 s下表现出接近零的烧蚀率,与传统的主动学习方法相比,显示出优越的采样效率。微观结构分析表明,这种优异的性能源于形成了高粘性的HfO2-SiO2-Ta2O5-HfSiO4-Hf3(BO3)4氧化层,提供了有效的氧屏障保护。这项工作展示了一种利用小数据快速发现材料的有效和通用方法。
本文章由计算机程序翻译,如有差异,请以英文原文为准。

Accelerated discovery of near-zero ablation ultra-high temperature ceramics via GAN-enhanced directionally constrained active learning

Accelerated discovery of near-zero ablation ultra-high temperature ceramics via GAN-enhanced directionally constrained active learning
In materials science, a significant correlation often exists between material input parameters and their corresponding performance attributes. Nevertheless, the inherent challenges associated with small data obscure these statistical correlations, impeding machine learning models from effectively capturing the underlying patterns, thereby hampering efficient optimization of material properties. This work presents a novel active learning framework that integrates generative adversarial networks (GAN) with a directionally constrained expected absolute improvement (EAI) acquisition function to accelerate the discovery of ultra-high temperature ceramics (UHTCs) using small data. The framework employs GAN for data augmentation, symbolic regression for feature weight derivation, and a self-developed EAI function that incorporates input feature importance weighting to quantify bidirectional deviations from zero ablation rate. Through only two iterations, this framework successfully identified the optimal composition of HfB2-3.52SiC-5.23TaSi2, which exhibits robust near-zero ablation rates under plasma ablation at 2500 ​°C for 200 ​s, demonstrating superior sampling efficiency compared to conventional active learning approaches. Microstructural analysis reveals that the exceptional performance stems from the formation of a highly viscous HfO2-SiO2-Ta2O5-HfSiO4-Hf3(BO3)4 oxide layer, which provides effective oxygen barrier protection. This work demonstrates an efficient and universal approach for rapid materials discovery using small data.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
CiteScore
33.30
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信