Towards the future of physics- and data-guided AI frameworks in computational mechanics

authored by: Jinshuai Bai, Yizheng Wang, Hyogu Jeong, Shiyuan Chu, Qingxia Wang, Laith Alzubaidi, Xiaoying Zhuang, Timon Rabczuk, Yi Min Xie, Xi Qiao Feng, Yuantong Gu
Abstract: The integration of physics-based modelling and data-driven artificial intelligence (AI) has emerged as a transformative paradigm in computational mechanics, This perspective reviews the development and current status of AI-empowered frameworks, including data-driven methods, physics-informed neural networks, and neural operators, While these approaches have demonstrated significant promise, challenges remain in terms of robustness, generalisation, and computational efficiency, We delineate four promising research directions: (1) Modular neural architectures inspired by traditional computational mechanics, (2) physics informed neural operators for resolution-invariant operator learning, (3) intelligent frameworks for multiphysics and multiscale biomechanics problems, and (4) structural optimisation strategies based on physics constraints and reinforcement learning, These directions represent a shift toward foundational frameworks that combine the strengths of physics and data, opening new avenues for the modelling, simulation, and optimisation of complex physical systems.
Organisation(s): Institute of Photonics
External Organisation(s): Queensland University of Technology
Tsinghua University
University of Southern Queensland
Bauhaus-Universität Weimar
Hohai University (HHU)
Type: Article
Journal: Acta Mechanica Sinica/Lixue Xuebao
Volume: 41
ISSN: 0567-7718
Publication date: 09.07.2025
Publication status: Published
Peer reviewed: Yes
ASJC Scopus subject areas: Computational Mechanics, Mechanical Engineering
Electronic version(s): https://doi.org/10.1007/s10409-025-25340-x (Access: Closed)
: Details in the research portal "Research@Leibniz University"

BibTeX

@article{fa32e3885d8a4caaad78e6ec4221c048,
title = "Towards the future of physics- and data-guided AI frameworks in computational mechanics",
abstract = "The integration of physics-based modelling and data-driven artificial intelligence (AI) has emerged as a transformative paradigm in computational mechanics, This perspective reviews the development and current status of AI-empowered frameworks, including data-driven methods, physics-informed neural networks, and neural operators, While these approaches have demonstrated significant promise, challenges remain in terms of robustness, generalisation, and computational efficiency, We delineate four promising research directions: (1) Modular neural architectures inspired by traditional computational mechanics, (2) physics informed neural operators for resolution-invariant operator learning, (3) intelligent frameworks for multiphysics and multiscale biomechanics problems, and (4) structural optimisation strategies based on physics constraints and reinforcement learning, These directions represent a shift toward foundational frameworks that combine the strengths of physics and data, opening new avenues for the modelling, simulation, and optimisation of complex physical systems.",
keywords = "Biomechanics, Computational mechanics, Operator learning, Physics-informed neural network, Topology optimisation",
author = "Jinshuai Bai and Yizheng Wang and Hyogu Jeong and Shiyuan Chu and Qingxia Wang and Laith Alzubaidi and Xiaoying Zhuang and Timon Rabczuk and Xie, {Yi Min} and Feng, {Xi Qiao} and Yuantong Gu",
note = "Publisher Copyright: {\textcopyright} The Chinese Society of Theoretical and Applied Mechanics and Springer-Verlag GmbH Germany, part of Springer Nature 2025.",
year = "2025",
month = jul,
day = "9",
doi = "10.1007/s10409-025-25340-x",
language = "English",
volume = "41",
journal = "Acta Mechanica Sinica/Lixue Xuebao",
issn = "0567-7718",
publisher = "Springer Verlag",
number = "7",
}

Back