{"title":"利用生成式人工智能实现数据高效的双连续复合材料机械设计","authors":"Milad Masrouri , Zhao Qin","doi":"10.1016/j.taml.2024.100492","DOIUrl":null,"url":null,"abstract":"<div><p>The distribution of material phases is crucial to determine the composite's mechanical property. While the full structure-mechanics relationship of highly ordered material distributions can be studied with finite number of cases, this relationship is difficult to be revealed for complex irregular distributions, preventing design of such material structures to meet certain mechanical requirements. The noticeable developments of artificial intelligence (AI) algorithms in material design enables to detect the hidden structure-mechanics correlations which is essential for designing composite of complex structures. It is intriguing how these tools can assist composite design. Here, we focus on the rapid generation of bicontinuous composite structures together with the stress distribution in loading. We find that generative AI, enabled through fine-tuned Low Rank Adaptation models, can be trained with a few inputs to generate both synthetic composite structures and the corresponding von Mises stress distribution. The results show that this technique is convenient in generating massive composites designs with useful mechanical information that dictate stiffness, fracture and robustness of the material with one model, and such has to be done by several different experimental or simulation tests. This research offers valuable insights for the improvement of composite design with the goal of expanding the design space and automatic screening of composite designs for improved mechanical functions.</p></div>","PeriodicalId":46902,"journal":{"name":"Theoretical and Applied Mechanics Letters","volume":null,"pages":null},"PeriodicalIF":3.2000,"publicationDate":"2024-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S2095034924000035/pdfft?md5=33fc5a8eba7d7bf17165eca971a0d917&pid=1-s2.0-S2095034924000035-main.pdf","citationCount":"0","resultStr":"{\"title\":\"Towards data-efficient mechanical design of bicontinuous composites using generative AI\",\"authors\":\"Milad Masrouri , Zhao Qin\",\"doi\":\"10.1016/j.taml.2024.100492\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><p>The distribution of material phases is crucial to determine the composite's mechanical property. While the full structure-mechanics relationship of highly ordered material distributions can be studied with finite number of cases, this relationship is difficult to be revealed for complex irregular distributions, preventing design of such material structures to meet certain mechanical requirements. The noticeable developments of artificial intelligence (AI) algorithms in material design enables to detect the hidden structure-mechanics correlations which is essential for designing composite of complex structures. It is intriguing how these tools can assist composite design. Here, we focus on the rapid generation of bicontinuous composite structures together with the stress distribution in loading. We find that generative AI, enabled through fine-tuned Low Rank Adaptation models, can be trained with a few inputs to generate both synthetic composite structures and the corresponding von Mises stress distribution. The results show that this technique is convenient in generating massive composites designs with useful mechanical information that dictate stiffness, fracture and robustness of the material with one model, and such has to be done by several different experimental or simulation tests. This research offers valuable insights for the improvement of composite design with the goal of expanding the design space and automatic screening of composite designs for improved mechanical functions.</p></div>\",\"PeriodicalId\":46902,\"journal\":{\"name\":\"Theoretical and Applied Mechanics Letters\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":3.2000,\"publicationDate\":\"2024-01-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://www.sciencedirect.com/science/article/pii/S2095034924000035/pdfft?md5=33fc5a8eba7d7bf17165eca971a0d917&pid=1-s2.0-S2095034924000035-main.pdf\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Theoretical and Applied Mechanics Letters\",\"FirstCategoryId\":\"1087\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S2095034924000035\",\"RegionNum\":3,\"RegionCategory\":\"工程技术\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"MECHANICS\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Theoretical and Applied Mechanics Letters","FirstCategoryId":"1087","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2095034924000035","RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"MECHANICS","Score":null,"Total":0}
引用次数: 0
摘要
材料相的分布是决定复合材料机械性能的关键。虽然高度有序的材料分布的完整结构-力学关系可以通过有限的案例进行研究,但对于复杂的不规则分布,这种关系却很难被揭示,从而阻碍了此类材料结构的设计以满足特定的力学要求。人工智能(AI)算法在材料设计领域的显著发展,使我们能够发现隐藏的结构-力学相关性,这对设计复杂结构的复合材料至关重要。令人感兴趣的是,这些工具如何协助复合材料设计。在这里,我们将重点放在快速生成双连续复合材料结构以及加载时的应力分布上。我们发现,生成式人工智能通过微调低等级适应模型(Low Rank Adaptation models),只需少量输入就能训练生成合成复合材料结构和相应的 von Mises 应力分布。研究结果表明,这种技术可以方便地生成大量复合材料设计,并提供有用的机械信息,通过一个模型就能确定材料的刚度、断裂和鲁棒性,而这些必须通过多个不同的实验或模拟测试才能完成。这项研究为改进复合材料设计提供了宝贵的见解,其目标是扩大设计空间和自动筛选复合材料设计,以提高机械功能。
Towards data-efficient mechanical design of bicontinuous composites using generative AI
The distribution of material phases is crucial to determine the composite's mechanical property. While the full structure-mechanics relationship of highly ordered material distributions can be studied with finite number of cases, this relationship is difficult to be revealed for complex irregular distributions, preventing design of such material structures to meet certain mechanical requirements. The noticeable developments of artificial intelligence (AI) algorithms in material design enables to detect the hidden structure-mechanics correlations which is essential for designing composite of complex structures. It is intriguing how these tools can assist composite design. Here, we focus on the rapid generation of bicontinuous composite structures together with the stress distribution in loading. We find that generative AI, enabled through fine-tuned Low Rank Adaptation models, can be trained with a few inputs to generate both synthetic composite structures and the corresponding von Mises stress distribution. The results show that this technique is convenient in generating massive composites designs with useful mechanical information that dictate stiffness, fracture and robustness of the material with one model, and such has to be done by several different experimental or simulation tests. This research offers valuable insights for the improvement of composite design with the goal of expanding the design space and automatic screening of composite designs for improved mechanical functions.
期刊介绍:
An international journal devoted to rapid communications on novel and original research in the field of mechanics. TAML aims at publishing novel, cutting edge researches in theoretical, computational, and experimental mechanics. The journal provides fast publication of letter-sized articles and invited reviews within 3 months. We emphasize highlighting advances in science, engineering, and technology with originality and rapidity. Contributions include, but are not limited to, a variety of topics such as: • Aerospace and Aeronautical Engineering • Coastal and Ocean Engineering • Environment and Energy Engineering • Material and Structure Engineering • Biomedical Engineering • Mechanical and Transportation Engineering • Civil and Hydraulic Engineering Theoretical and Applied Mechanics Letters (TAML) was launched in 2011 and sponsored by Institute of Mechanics, Chinese Academy of Sciences (IMCAS) and The Chinese Society of Theoretical and Applied Mechanics (CSTAM). It is the official publication the Beijing International Center for Theoretical and Applied Mechanics (BICTAM).