Prediction of High Velocity Impact Damage Behavior of 3D Angle‐Interlock Woven Fabrics Based on Deep Learning
- 23 January 2026
- journal article
- Published by Wiley in Polymer Composites
- Vol. 47 (11) , 9776-9791
Abstract
3D angle‐interlock woven fabrics (3DAWFs) have been extensively applied in aerospace and other industries because of their outstanding mechanical characteristics; thus, the investigation of their damage mechanism under impact loading has always received much attention. However, the high costs associated with experiments and the limited computational efficiency of finite element (FE) analysis have hindered the effective characterization of damage behaviors across varying parameter configurations. To address these challenges, this study presents two deep learning models. The first is a convolutional neural network (CNN) featuring an encoder‐decoder architecture enhanced by a center bias mechanism, which consists of a spatial attention module with a Gaussian enhancement mask. This model, utilizing a dataset derived from FE simulations, accurately predicts damage images of 3DAWFs, as well as deformation images that illustrate the energy absorption mechanism through the input of projectile impact angle, initial velocity, and warp/weft direction, with its mean squared error (MSE) ranging from 6.73 × 10 −3 to 9.86 × 10 −3 . The second model is a multi‐layer perceptron (MLP) that quantitatively predicts the projectile residual velocity based on the aforementioned input features, achieving a coefficient of determination ( R 2 ) of 0.9904 and a mean absolute percentage error (MAPE) of 3.72%, and explains the influence of each feature via SHapley Additive exPlanations (SHAP) interpretability analysis. Together, the CNN and MLP models successfully predict damage morphology and mechanical properties, enabling real‐time monitoring of the impact damage behavior of 3DAWFs and offering a novel approach for a comprehensive study on the dynamic damage behavior.Keywords
This publication has 36 references indexed in Scilit:
- Ballistic performance of UHMWPE laminate/modified aramid fabric composite structure based on support vector machine (SVM) methodComposites Part B: Engineering, 2025
- Deep learning paradigm shift in composite materials performance prediction: A review of state-of-the-art applicationsComposites Communications, 2025
- Deep learning-based analysis of damage mechanisms in 3D angle-interlock woven composites under variable impact conditionsComposites Science and Technology, 2025
- A few-shot deep learning framework for predicting high-velocity impact response of ultra-high molecular weight polyethylene fiber-reinforced compositesAerospace Science and Technology, 2025
- Advanced Meta-Modeling framework combining Machine Learning and Model Order Reduction towards real-time virtual testing of woven composite laminates in nonlinear regimeComposites Science and Technology, 2025
- A review on recent applications of machine learning in mechanical properties of compositesPolymer Composites, 2024
- An explainable artificial intelligence‐based approach for reliable damage detection in polymer composite structures using deep learningPolymer Composites, 2024
- Determining damage initiation of carbon fiber reinforced polymer composites using machine learningPolymer Composites, 2022
- Descriptive modeling of textiles using FE simulations and deep learningComposites Science and Technology, 2021
- A mechanistic interpretation of the comparative in-plane mechanical properties of 3D woven, stitched and pinned compositesComposites Part A: Applied Science and Manufacturing, 2010