Enhancing predictive performance of artificial neural network modelling composite stiffness degradation with virtual samples
Authors:
- 8 May 2026
- journal article
- research article
- Published by SAGE Publications in Journal of Composite Materials
Abstract
This paper proposes a methodology for generating virtual samples of stiffness degradation of composite laminates to facilitate the training of artificial neural network (ANN)-based predictive models. Considering the impact of matrix cracks, delamination, and fiber fracture on the stiffness of laminated panels, a finite element model driven by the physical mechanism of fatigue damage is established for generating virtual samples of fatigue evolution. The stiffness and evolution of virtual samples for [0/90 3 /0/90 3 ] S glass fiber reinforced polymer (GFRP) laminates are obtained. By employing the fatigue evolution prediction method based on β -variational autoencoder ( β -VAE) and neural ordinary differential equation networks (Neural ODE), the proposed approach effectively learns the underlying mechanisms in fatigue evolution data and shows good correlation to experimental observation. The results indicate that physics-driven virtual sample generation significantly enhances the predictive accuracy and robustness of neural network models, broadening their practical applicability.Keywords
Funding Information
- National Natural Science Foundation of China (52235003)
- National Natural Science Foundation of China (52575181)
This publication has 32 references indexed in Scilit:
- Investigation on the effect of ply orientations on the fatigue behavior of composite laminates: from the perspective of deep learningComposites Part B: Engineering, 2026
- Fatigue damage characterization for composite laminates using deep learning and laser ultrasonicComposites Part B: Engineering, 2021
- Data-driven algorithm for real-time fatigue life prediction of structures with stochastic parametersComputer Methods in Applied Mechanics and Engineering, 2020
- Experimental investigation on damage evolution in cross-ply laminates subjected to quasi-static and fatigue loadingComposites Part B: Engineering, 2017
- Influence of ply cluster thickness and location on matrix cracking evolution in open-hole composite laminatesComposites Part B: Engineering, 2016
- Effects of layup angle and ply thickness on matrix crack interaction in contiguous plies of composite laminatesComposites Part A: Applied Science and Manufacturing, 2005
- Prediction of fatigue damage and life for composite laminates under service loading spectraComposites Science and Technology, 1996
- The analysis of survival (mortality) data: Fitting Gompertz, Weibull, and logistic functionsMechanisms of Ageing and Development, 1994
- Life Prediction for Resin-Matrix Composite Materials1991
- A simplified analysis of transverse ply cracking in cross-ply laminatesComposites Science and Technology, 1988