Direct reconstruction of unit-cell models from micro-CT scanning for multiscale woven prepreg forming analysis

Forming prediction of woven composites is crucial for optimizing manufacturing parameters in the fabrication of large-scale and high-quality equipment across modern automotive, aerospace and marine industries. This work presents a systematic combination of multiscale forming simulation framework and the high-fidelity finite element (FE) model to predict non-uniform yarn orientation and laminate thickness distributions. Specifically, the high-fidelity FE model based on the unit-cell reconstructed directly via a deep-learning algorithm compatible with low-resolution CT images without manual adjustment is incorporated into the integrated mesoscopic simulation model previously developed for elucidating deformation mechanisms. The indispensable characterization experiments are simplified to only uniaxial tension and consolidation tests on orthogonal composites, minimizing experimental redundancy while preserving essential data integrity. Concurrently, the coupling deformation modes impervious to direct measurement can be reliably captured through virtual characterization based on the high-fidelity FE model, leading to exceptional predictive accuracy for stress-strain relationships with 2.1% average error compared to the idealized unit-cell model established before with 15.2% average error. Subsequently, these homogenized relationships were fed into the macroscale forming simulation, which was then validated through prepreg compression molding (PCM), resulting in strong consistency between experimental and simulation data on the deformed geometry, yarn angle and thickness distribution, exhibiting respective errors of 7.59%, 4.43° and 14.7%. Therefore, the proposed multiscale simulation framework not only mitigates experimental time costs and operational complexity but also augments forming process precision by accounting for the coupling effects between distinct deformation modes, addressing a critical gap in the efficient fabrication of high-performance woven composite components.

Authors

Institutions

Publication Details

Journal
Composites Science and Technology
Published
2026-03-12
DOI
https://doi.org/10.1016/j.compscitech.2026.111601
Primary Topic
Epoxy Resin Curing Processes
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Direct reconstruction of unit-cell models from micro-CT scanning for multiscale woven prepreg forming analysis

Tianxiao Niu, Meiyu Liu, Deyong Sun, Tianyu Gao et al.
Composites Science and Technology
Epoxy Resin Curing Processes
article

Direct reconstruction of unit-cell models from micro-CT scanning for multiscale woven prepreg forming analysis

Tianxiao Niu, Meiyu Liu, Deyong Sun, Tianyu Gao, Winston Wai Shing Ma, Guoxin Fang, Weizhao Zhang, Yubo Hu, Chongrui Tang
article en

Abstract

Forming prediction of woven composites is crucial for optimizing manufacturing parameters in the fabrication of large-scale and high-quality equipment across modern automotive, aerospace and marine industries. This work presents a systematic combination of multiscale forming simulation framework and the high-fidelity finite element (FE) model to predict non-uniform yarn orientation and laminate thickness distributions. Specifically, the high-fidelity FE model based on the unit-cell reconstructed directly via a deep-learning algorithm compatible with low-resolution CT images without manual adjustment is incorporated into the integrated mesoscopic simulation model previously developed for elucidating deformation mechanisms. The indispensable characterization experiments are simplified to only uniaxial tension and consolidation tests on orthogonal composites, minimizing experimental redundancy while preserving essential data integrity. Concurrently, the coupling deformation modes impervious to direct measurement can be reliably captured through virtual characterization based on the high-fidelity FE model, leading to exceptional predictive accuracy for stress-strain relationships with 2.1% average error compared to the idealized unit-cell model established before with 15.2% average error. Subsequently, these homogenized relationships were fed into the macroscale forming simulation, which was then validated through prepreg compression molding (PCM), resulting in strong consistency between experimental and simulation data on the deformed geometry, yarn angle and thickness distribution, exhibiting respective errors of 7.59%, 4.43° and 14.7%. Therefore, the proposed multiscale simulation framework not only mitigates experimental time costs and operational complexity but also augments forming process precision by accounting for the coupling effects between distinct deformation modes, addressing a critical gap in the efficient fabrication of high-performance woven composite components.

Composites Science and TechnologyVol. 279
Chinese University of Hong Kong (HK), Zhejiang University (CN), University of Twente (NL)
Life below water
Openalex Percentile: Top 26%
Epoxy Resin Curing Processes
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.