Article Dans Une Revue Procedia Structural Integrity Année : 2025

Deep learning based predictions of Wöhler curve using physics-informed neural networks: self-heating determination of fatigue limits for laminated composite materials

Résumé

This study investigates the fatigue behavior of high-strength carbon/epoxy laminates using Physics-Informed Neural Networks (PINNs). Laminates composed of TR50 unidirectional fibers and R367-2 epoxy were tested under cyclic tension and fatigue at room temperature. Self-heating tests were also conducted to efficiently estimate the fatigue limits. A strong agreement was found between classical fatigue tests and the self-heating method. A PINN-based approach was employed to reconstruct S–N curves by embedding a fatigue model into the learning process. The model, implemented in Python using Keras/TensorFlow, accurately predicts fatigue performance, and a fully commented code example is provided

Fichier principal
Vignette du fichier
Gornet2025.pdf (1.07 Mo) Télécharger le fichier

Dates et versions

hal-05411712 , version 1 (24-12-2025)

Licence

Identifiants

Citer

Laurent Gornet. Deep learning based predictions of Wöhler curve using physics-informed neural networks: self-heating determination of fatigue limits for laminated composite materials. Procedia Structural Integrity, 2025, ⟨10.1016/j.prostr.2025.11.015⟩. ⟨hal-05411712⟩
98 Consultations
85 Téléchargements

Altmetric

Partager

  • More