AI赋能结构设计

卢天健; 孟晗; 姜永烽

doi:10.21656/1000-0887.470062

AI赋能结构设计

doi: 10.21656/1000-0887.470062

南京航空航天大学，南京 210016

基金项目:

国家自然科学基金（U2570248；52361165626）

详细信息

作者简介:
卢天健(1964—)，男，教授，博士，博士生导师(通信作者. E-mail: tjlu@nuaa.edu.cn)；孟晗（1989—）,女,教授,博士,博士生导师（E-mail: menghan@nuaa.edu.cn）;姜永烽（2000—）,男,博士生（E-mail: yfjiang@nuaa.edu.cn）.

通讯作者:
卢天健(1964—)，男，教授，博士，博士生导师(通信作者. E-mail: tjlu@nuaa.edu.cn)

中图分类号: O31
计量
- 文章访问数: 59
- HTML全文浏览量: 11
- PDF下载量: 29
- 被引次数: 0
出版历程
- 收稿日期: 2026-03-06
- 修回日期: 2026-03-10
- 网络出版日期: 2026-04-01
- 刊出日期: 2026-03-01

AI Enables Structural Design

Nanjing University of Aeronautics and Astronautics, Nanjing 210016, P.R.China

Funds:

The National Science Foundation of China(U2570248；52361165626)

摘要

摘要:
人工智能（artificial intelligence, AI）的快速发展，正在重塑工程结构设计的研究图景.在力学长期奠定的理论框架之上，大语言模型（large language model, LLM）等新一代AI技术，为结构设计提供了新的认知工具与方法视角.随着结构形态空间的持续拓展，以及多尺度与多物理场耦合问题的不断涌现，依赖经验积累与局部试探的传统设计路径正逐渐逼近复杂性边界.在这样的背景下，AI不仅为高维设计空间探索、知识表达与跨学科融合提供了新的可能，也在拓展人类理解复杂系统与开展决策分析的方式.面向未来，结构设计有望在AI推动下形成“设计制造测试”闭环体系、多物理场统一设计范式和智能结构系统等新的发展路径，从而为结构科学在新的时代背景下持续拓展认知边界提供重要契机.
- 力学 /
- 人工智能 /
- 大语言模型 /
- 结构设计 /
- 超结构
Abstract:
The rapid development of artificial intelligence (AI) is reshaping the research landscape of structural design. Based upon the long-established theoretical framework of mechanics, emerging AI technologies, particularly large language models (LLMs), are providing new cognitive tools and methodological perspectives for structural design. As the structural morphology space continues to expand and multi-scale, multi-physics coupling problems become increasingly prevalent, traditional design approaches relying on accumulated experience and local trial-and-error methods are approaching their limits in handling growing complexity. In this context, AI not only opens new possibilities for exploring high-dimensional design spaces, representing knowledge, and facilitating interdisciplinary integration, but also extends the way humans understand and analyze complex systems. Looking ahead, AI may enable new development pathways for structural design, including closed-loop frameworks integrating design, manufacturing, and testing, unified paradigms for multi-physics structural design, and the emergence of intelligent structural systems. These developments offer important opportunities for structural science to further expand its cognitive boundaries in an era increasingly shaped by intelligent technologies.
- mechanics /
- artificial intelligence (AI) /
- large language model (LLM) /
- structural design /
- metastructure

HTML全文

参考文献(17)

卢天健. (主编按语)桥仍在, 河向前[J]. 应用数学和力学, 2026,47(1): ⅰ-ⅳ.(LIU Tianjian.(Chief Editor’s Note) The current runs while the bridge holds[J].Applied Mathematics and Mechanics,2026,47

(1): ⅰ-ⅳ. (in Chinese))

[2]康瑞, 李雪, 孟晗, 等. 轻巧-承力-功能一体化超结构: 概念、设计及应用[J]. 应用数学和力学, 2024,45(8): 949-973.(KANG Rui, LI Xue, MENG Han, et al. Ultralight, compact, and load-bearing multifunctional metastructures: concept, design and applications[J].Applied Mathematics and Mechanics,2024,45(8): 949-973. (in Chinese))

[3]JIANG Y, LI Z, REN J, et al. Autoencoder artificial neural network for accelerated forward and inverse design of locally resonant acoustic metamaterials[J].Journal of Applied Physics,2025,137(5): 053104.

[4]SHI Z, CHEN C, ZHANG D, et al. Inverse design and spatial optimization of SFAM via deep learning[J].International Journal of Mechanical Sciences,2025,306: 110855.

[5]QIAN C, KAMINER I, CHEN H. A guidance to intelligent metamaterials and metamaterials intelligence[J].Nature Communications,〖STHZ〗16(1): 1154.

[6]WANG L, ZHANG L, HE G. Evaluations of large language models in computational fluid dynamics: leveraging, learning and creating knowledge[J].Theoretical and Applied Mechanics Letters,2025,15(3): 100597.

[7]JIANG Y, CAO S, MENG H, et al. A data-driven design for sound absorption of acoustic metamaterials based on large language models[J]. Scientific Reports,2025,16(1): 517.

[8]JI H, ZHANG Y, WANG X, et al. A few-shot deep learning framework for predicting high-velocity impact response of ultra-high molecular weight polyethylene fiber-reinforced composites[J].Aerospace Science and Technology,2025,163: 110152.

[9]LIU S, WEN T, YE B, et al. Large language models for material property predictions: elastic constant tensor prediction and materials design[J].Digital Discovery,2025,4(6): 1625-1638.

[10]WANG Z, WANG S, MA C, et al. The prediction of homogenized effective properties of continuous fiber composites based on a deep transfer learning approach[J].Composites Science and Technology,2025,262: 111050.

[11]SNAPP K L, VERDIER B, GONGORA A E, et al. Superlative mechanical energy absorbing efficiency discovered through self-driving lab-human partnership[J].Nature Communications,2024,15: 4290.

[12]LU D, MALOF J M, PADILLA W J. An agentic framework for autonomous metamaterial modeling and inverse design[J].ACS Photonics,2025,12(11): 6071-6080.

[13]孙山有铭. 混杂/层级夹层结构承载/散热特性研究及其AI算法优化[D]. 西安: 西安交通大学, 2021.(SUN S Y M. Study on load-bearing and heat dissipation characteristics of hybrid/hierarchical sandwich structures and its ai algorithm optimization[D]. Xi’an: Xi’an Jiaotong University, 2021.(in Chinese))

[14]FENG J, QIAO J, XU Q, et al. Accelerated design of acoustic-mechanical multifunctional metamaterials via neural network[J].International Journal of Mechanical Sciences,2025,287: 109920.

[15]高兆瑞, 李铮, 姜永烽, 等. 基于人工神经网络的共振吸声超材料声学性能快速预测及结构优化设计[J]. 应用数学和力学, 2024,45(8): 1058-1069.(GAO Zhaorui, LI Zheng, JIANG Yongfeng, et al. Acoustic performance rapid prediction and structural optimization for resonant sound-absorbing metamaterials based on artificial neural networks[J].Applied Mathematics and Mechanics,2024,45(8): 1058-1069. (in Chinese))

[16]WANG H, YANG Y, ZHOU X, et al. Rational design of mechanical bio-metamaterials for biomedical applications[J].Progress in Materials Science,2026,156: 101545.

[17]XIE N, TIAN J, LI Z, et al. Invited review for 20th anniversary special issue of PLRev “AI for mechanomedicine”[J].Physics of Life Reviews,2024,51: 328-342.

施引文献

资源附件(0)

访问统计