A unified energy framework combining nonlocal macro-meso-scale consistent damage and cohesive zone models for failure prediction of fiber-reinforced composites
Abstract
In this study, we propose a unified numerical framework based on energy that couples the nonlocal macro-meso-scale consistent damage model (NMMD) for the bulk with the cohesive zone model (CZM) for material interfaces to simulate complex crack propagation in multi-constituent composite systems. The NMMD can accurately capture the crack initiation, growth, and coalescence within individual constituents, whereas the cohesive laws describe the interfacial separation and sliding well without overestimation of the interface thickness in nonlocal formulations. Moreover, variational coupling guarantees energy additivity and prevents double counting by mapping the cross-interface nonlocal contributions onto a zero-thickness cohesive surface; thus, the bulk and interface dissipations remain disjointed and consistent. Subsequently, numerical studies on different composite material structures solidly validate the approach, showing that the predicted crack paths and mechanical responses are sensitive to interfacial strength and correctly reflect mode transitions. The framework accurately reproduces the crack evolution in fiber-reinforced materials. Remarkably, the formulation establishes an energetically consistent connection between nonlocal bulk damage and local interface fracture, thereby providing new insight into the fracture behavior and toughening mechanisms in structured fiber-reinforced composites.
Details
- Organisationseinheit(en)
-
Institut für Photonik
- Externe Organisation(en)
-
Hohai University
- Typ
- Artikel
- Journal
- Computer Methods in Applied Mechanics and Engineering
- Band
- 460
- ISSN
- 0045-7825
- Publikationsdatum
- 27.05.2026
- Publikationsstatus
- Elektronisch veröffentlicht (E-Pub)
- Peer-reviewed
- Ja
- ASJC Scopus Sachgebiete
- Numerische Mechanik, Werkstoffmechanik, Maschinenbau, Allgemeine Physik und Astronomie, Angewandte Informatik
- Elektronische Version(en)
-
https://doi.org/10.1016/j.cma.2026.119101 (Zugang:
Geschlossen
)
