PREDICTION MODEL FOR FATIGUE THRESHOLDS OF COMPOSITES BASED ON STRAIN ENERGY DENSITY

Composite laminates are prone to mixed-mode delamination under complex loading conditions, and reliable characterization of static fracture envelopes and fatigue delamination-growth thresholds is essential for damage-tolerance design. However, the high cost of fracture tests, insufficient utilization of mixed-mode experimental data, and reliance on empirical fitting in fatigue-threshold prediction remain major challenges. To address these issues, this study develops a unified framework for static fracture-parameter inversion and fatigue-threshold prediction based on an existing modified strain energy density criterion with a mixed-mode-ratio-dependent critical distance. The effect of the varying critical distance is incorporated into the static fracture-envelope description, enabling the pure Mode I and Mode II fracture toughness values to be inversely estimated from a limited number of mixed-mode fracture tests. The complete static mixed-mode fracture envelope can then be reconstructed without extensive pure-mode testing. Subsequently, inspired by the traction-separation relationship and failure-envelope evolution characteristics of cohesive-zone models, the similarity between the static fracture envelope and the fatigue delamination-growth threshold envelope is introduced to establish a fatigue-threshold model. The model simultaneously accounts for the effects of mixed-mode ratio and stress ratio, while an inverse estimation procedure is formulated to determine the pure-mode fatigue thresholds from limited mixed-mode fatigue data. The proposed framework is evaluated against experimental data reported in the literature for carbon-fiber-reinforced composites, glass-fiber-reinforced composites, and wood, covering different orthotropic material systems. The results demonstrate that the framework can reasonably reproduce both static mixed-mode fracture envelopes and fatigue-threshold envelopes over a broad range of mixed-mode ratios. The inversely predicted pure-mode fracture toughness and fatigue thresholds also show good agreement with experimental results. The proposed framework provides an efficient approach for extracting static and fatigue fracture parameters from limited mixed-mode tests, thereby reducing experimental cost and testing time and providing physically based threshold inputs for subsequent mixed-mode fatigue crack-growth and delamination life prediction.