Fillet welds are extensively employed in ship hull structures. Under cyclic loading, fatigue cracks are prone to initiate and propagate at both the weld toe and root due to incomplete penetration, root gaps, and local stress concentrations. However, for weld-root fatigue assessment, the Effective Notch Stress (ENS) method requires detailed notch modeling, significantly limiting its practical applicability. To address this limitation, this study proposes a fracture mechanics-based Fatigue Crack Growth (FCG) framework incorporating a “virtual initial flaw” concept, enabling life predictions equivalent to those of the ENS method. The equivalence between the two approaches is formulated, and predictive equations for the virtual flaw size are derived as functions of plate thickness and weld leg length. The framework was validated using T-type specimens and excavator welds, demonstrating excellent agreement with ENS results, showing a deviation in fatigue life of less than 10%. Furthermore, integrating this FCG framework into a direct analysis-based hull weld design confirmed that the weld leg length could be optimized from 9.5 mm to 8.0 mm while maintaining structural integrity. Consequently, by introducing a fracture mechanics-based fatigue design framework, this study provides a comprehensive methodology applicable to both design-stage fatigue evaluation and in-service structural integrity assessment.
Kim et al. (Fri,) studied this question.