AT-PINN: Advanced time-marching physics-informed neural network for structural vibration analysis

被引:18
作者
Chen, Zhaolin [1 ,3 ]
Lai, Siu-Kai [1 ,2 ]
Yang, Zhichun [3 ]
机构
[1] Hong Kong Polytech Univ, Dept Civil & Environm Engn, Kowloon, Hong Kong, Peoples R China
[2] Hong Kong Polytech Univ, Natl Rail Transit Electrificat & Automat Engn Tech, Hong Kong Branch, Kowloon, Hong Kong, Peoples R China
[3] Northwestern Polytech Univ, Sch Aeronaut, Xian, Peoples R China
基金
中国博士后科学基金;
关键词
Long-duration simulation; Physics-informed neural network; Structural vibration; Time-marching; Normalization; DEEP LEARNING FRAMEWORK;
D O I
10.1016/j.tws.2023.111423
中图分类号
TU [建筑科学];
学科分类号
0813 ;
摘要
Solving partial differential equations through deep learning has recently received wide attention, with physicsinformed neural networks (PINNs) being successfully used and showing great potential. This study focuses on the development of an efficient PINN approach for structural vibration analysis in "long -duration" simulation that is still a technical but unresolved issue of PINN. The accuracies of the standard PINN (STD-PINN) and conventional time -marching PINN (CT-PINN) methods in solving vibration equations, especially free -vibration equations, are shown to decrease to varying degrees with the simulation time. To resolve this problem, an advanced timemarching PINN (AT-PINN) approach is proposed. This method is used to solve structural vibration problems over successive time segments by adopting four key techniques: normalization of the spatiotemporal domain in each time segment, a reactivating optimization algorithm, transfer learning and the sine activation function. To illustrate the advantages of the AT-PINN approach, numerical simulations for the forced and free vibration analysis of strings, beams and plates are performed. In addition, the vibration analysis of plates under multiphysics loads is also studied. The results show that the AT-PINN approach can provide accurate solutions with lower computational cost even in long -duration simulation. The techniques adopted are verified to effectively avoid the offset of the spatiotemporal domain, reduce the accumulative error and enhance the training efficiency. The present one overcomes the drawback of the existing PINN methods and is expected to become an effective method for solving time -dependent partial differential equations in long -duration simulation.
引用
收藏
页数:29
相关论文
共 50 条
  • [21] Time-averaged flow field reconstruction based on a multifidelity model using physics-informed neural network (PINN) and nonlinear information fusion
    Rui, En-Ze
    Zeng, Guang-Zhi
    Ni, Yi-Qing
    Chen, Zheng-Wei
    Hao, Shuo
    INTERNATIONAL JOURNAL OF NUMERICAL METHODS FOR HEAT & FLUID FLOW, 2024, 34 (01) : 131 - 149
  • [22] M-PINN: A mesh-based physics-informed neural network for linear elastic problems in solid mechanics
    Wang, Lu
    Liu, Guangyan
    Wang, Guanglun
    Zhang, Kai
    INTERNATIONAL JOURNAL FOR NUMERICAL METHODS IN ENGINEERING, 2024, 125 (09)
  • [23] VC-PINN: Variable coefficient physics-informed neural network for forward and inverse problems of PDEs with variable coefficient
    Miao, Zhengwu
    Chen, Yong
    PHYSICA D-NONLINEAR PHENOMENA, 2023, 456
  • [24] CAN-PINN: A fast physics-informed neural network based on coupled-automatic-numerical differentiation method
    Chiu, Pao-Hsiung
    Wong, Jian Cheng
    Ooi, Chinchun
    Ha Dao, My
    Ong, Yew-Soon
    COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING, 2022, 395
  • [25] Physics-informed Neural Network for system identification of rotors
    Liu, Xue
    Cheng, Wei
    Xing, Ji
    Chen, Xuefeng
    Zhao, Zhibin
    Zhang, Rongyong
    Huang, Qian
    Lu, Jinqi
    Zhou, Hongpeng
    Zheng, Wei Xing
    Pan, Wei
    IFAC PAPERSONLINE, 2024, 58 (15): : 307 - 312
  • [26] Time difference physics-informed neural network for fractional water wave models
    Liu, Wenkai
    Liu, Yang
    Li, Hong
    RESULTS IN APPLIED MATHEMATICS, 2023, 17
  • [27] Time difference physics-informed neural network for fractional water wave models
    Liu, Wenkai
    Liu, Yang
    Li, Hong
    RESULTS IN APPLIED MATHEMATICS, 2023, 17
  • [28] A physics-informed neural network for Kresling origami structures
    Liu, Chen-Xu
    Wang, Xinghao
    Liu, Weiming
    Yang, Yi-Fan
    Yu, Gui-Lan
    Liu, Zhanli
    INTERNATIONAL JOURNAL OF MECHANICAL SCIENCES, 2024, 269
  • [29] Realizing the Potential of Physics-Informed Neural Network in Modelling
    Kheirandish, Zahra
    Schulz, Wolfgang
    JOURNAL OF LASER MICRO NANOENGINEERING, 2024, 19 (03): : 209 - 213
  • [30] A transfer learning-physics informed neural network (TL-PINN) for vortex-induced vibration
    Tang, Hesheng
    Liao, Yangyang
    Yang, Hu
    Xie, Liyu
    OCEAN ENGINEERING, 2022, 266