Finite element modelling of unidirectional carbon fiber notched composite under compression using in-situ X-Ray CT fiber orientation data
Year: 2026
Type: preprint
Abstract: Preprint of Proceedings of ECCM 2025 - 22nd European Conference on Composite Materials. Oslo, Norway, June 21 - 25, 2026. Abstract Pultruded carbon fiber composites are widely used in wind turbine blade construction due to their high mechanical performance, fiber alignment, and consistent material quality. However, variations in fiber orientation can significantly reduce compressive strength, thereby limiting the structural load-bearing capacity. Accurate numerical prediction of compressive behavior, accounting for fiber orientation variability, remains a challenge. In this work, X-ray computed tomography (CT) images of a notched pultruded composite sample were acquired at the Diamond Light Source synchrotron facility and subsequently analysed using the structure tensor method to determine the fiber orientation distribution along the full sample length, enabling a more realistic representation of the composite microstructure. The extracted fiber orientations were used as input for a two-dimensional finite element model to predict the compressive strength and kink band formation. The model implemented mesh-independent material homogenization of the fiber and matrix, accounting for the fiber diameter as the material length scale. The material properties were determined through mechanical testing of the composite and its constituents and implemented in the model. The fibers were described as hyperelastic, while the matrix as hyperelasto-plastic using a Ramberg-Osgood material model. The samples’ notch geometry was included in the model. The predicted compressive stress-strain behavior, strength and kink band formation show good agreement with experimental data acquired in situ.
Institutions: Technical University of Denmark, University of Manchester
Language: English
Subfield: Mechanics of Materials
Field: Engineering
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7214929887