Automatic-differentiation-based inverse analysis methods, including physics-informed neural networks (PINNs) and differentiable programming, have recently shown great promise due to their ability to compute accurate gradients and convergence efficiency. However, their applicability to falling weight deflectometer (FWD) backcalculation remains unexplored. This study critically evaluates PINN-based inverse analysis for a multilayer pavement system and investigates differentiable finite element method (DiffFEM) as an alternative based on a synthetic benchmark. The standard PINN does not recover layer moduli because of the sharp domain discontinuities inherent to layered pavement systems. Although we use an extended PINN with domain decomposition (XPINN), which shows better performance on discontinuous domains, its performance remains highly sensitive to loss weighting and network architecture, and degrades under measurement noise. By contrast, DiffFEM consistently achieves more accurate, stable, and computationally efficient inversion results. These results show and quantify the expected advantage of DiffFEM, which enforces the governing physics as a hard constraint in the solver, over PINN-based approaches, where the governing physics is imposed as a soft constraint through the loss function. More broadly, the findings suggest that the choice between PINN- and DiffFEM-based inverse analysis needs careful consideration, with DiffFEM offering practical advantages when an efficient and robust differentiable forward solver is available.
A mixed formulation for physics-informed neural networks as a potential solver for engineering problems in heterogeneous domains: comparison with finite element method
Technical University of Darmstadt(达姆施塔特工业大学);RWTH Aachen University(亚琛工业大学)
Scientific Machine Learning through Physics-Informed Neural Networks: Where we are and What's next
University of Naples Federico II(那不勒斯费德里科二世大学);SISSA – International School for Advanced Studies(国际高等研究学院(SISSA));University of Colorado Boulder(科罗拉多大学博尔德分校)
Physics-Informed Neural Networks for Radial Consolidation of Combined Electroosmotic, Vacuum and Surcharge Preloading Considering Smear Effects
Department of Civil, Environmental, and Infrastructure Engineering, George Mason University(乔治·马歇尔大学土木、环境与基础设施工程系);State Key Laboratory of Cryospheric Science and Frozen Soil Engineering, Northwest Institute of Eco-Environment and Resources, Chinese Academy of Sciences(中国科学院寒区工程与冻土科学国家重点实验室,西北生态环境资源研究院);Laboratoire Navier/CERMES, École Nationale des Ponts et Chaussées, Institut Polytechnique de Paris(巴黎理工学院劳纳实验室/塞梅斯实验室,法国国家桥梁与道路学院);College of Water Conservancy and Hydropower Engineering, Hohai University(河海大学水利水电学院);School of Geosciences and Info-physics, Central South University(中南大学地球科学与信息物理学院)
Multiscale lubrication simulation based on fourier feature networks with trainable frequency
School of Mechanical Engineering, Shanghai Jiao Tong University(上海交通大学机械工程学院);Shanghai marine diesel research institute(上海船用柴油机研究所);National key laboratory of marine engine science and technology(船用发动机科学技术国家重点实验室)
针对传统 PINN 因谱偏差难以处理粗糙表面高频信号的问题,提出采用可训练 Fourier 特征网络的多尺度润滑神经网络,在多种形貌上达到与 FEM 高度一致,并优于固定频率网络。
Engineering application of physics-informed neural networks for Saint-Venant torsion
Department of Quantum System Engineering, Jeonbuk National University(全量子系统工程系,全州国立大学);Department of Mathematics, Inha University(数学系,inha大学);Electrified Propulsion Structure CAE Team, Hyundai Motor Company(电动推进结构CAE团队,现代汽车公司);Department of Mechanical Engineering, Inha University(机械工程系,inha大学);Graduate School of Integrated Energy-AI, Jeonbuk National University(整合能源-人工智能研究生院,全州国立大学)
Lagged backward-compatible physics-informed neural networks for unsaturated soil consolidation analysis
George Mason University(乔治梅森大学);National Institute of Natural Hazards, Ministry of Emergency Management(应急管理部国家自然灾害防治研究院);Northwest Institute of Eco-Environment and Resources, Chinese Academy of Sciences(中国科学院西北生态环境资源研究院);École Nationale des Ponts et Chaussées(巴黎路桥学院);Hebei University of Engineering(河北工程大学);Zhejiang University of Technology(浙江工业大学)
Revisiting Heat Flux Analysis of Tungsten Monoblock Divertor on EAST using Physics-Informed Neural Network
School of Computer Science and Technology, Anhui University(安徽大学计算机科学与技术学院);Tongling University(铜陵大学);Institute of Plasma Physics, Chinese Academy of Sciences(中国科学院等离子体物理研究所)